102 Commits
Author SHA1 Message Date
justme 699ed8c411 ci: sync main into dev [CI SKIP] 2026-10-07 16:20:25 +08:00
justme 1ff847e45a feat: declare every protocol method's direction and enforce it in the router; add the zero-trust design [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
Methods, roles and binary channels are declared in one table per plane (node and peer). RpcRouter checks direction, kind and role when handlers are registered, when messages arrive (fleet_protocol_violation, disconnect after repeated violations) and before sending. NodeConnection is split by responsibility.
2026-10-07 16:17:02 +08:00
justme 0012df7b91 ci: sync main into dev [CI SKIP] 2026-10-07 15:08:26 +08:00
justme ee22309e65 fix: keep the Backend backup out of the journal [CI SKIP] 2026-10-07 15:07:39 +08:00
justme d5967b6110 ci: sync main into dev [CI SKIP] 2026-10-07 14:57:08 +08:00
justme 9bcef56497 feat: daily node restart controlled from the admin, connect timings as a default line task [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
Agents and Probes exit in order at a configured time (default 04:30 UTC+8, switch and time zone in the admin settings) and the service manager starts them again; a local restart_at or MOSHIRO_RESTART_AT wins. The connect kind (TCP connect and TLS handshake timings) is now a default continuous line template, with a section on the line page.
2026-10-07 14:56:50 +08:00
justme 450748f3bc ci: sync main into dev [CI SKIP] 2026-10-07 14:18:03 +08:00
justme 388619b8b6 feat: add a re-runnable deploy script for container nodes and migrate listen.bind to :: [CI SKIP] 2026-10-07 14:17:50 +08:00
justme 3b73fd7474 fix: pass Backend listener port into container (#18) [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
Pass the host listener URL to the Backend container process.
2026-10-07 05:17:13 +00:00
justme e423d9b506 fix: pass Backend listener port into container [CI]
ci/woodpecker/pr/verify Pipeline was successful
ci/woodpecker/pr/web-build Pipeline was successful
2026-10-07 13:13:21 +08:00
justme 3364f7903e fix: bind Backend host service to its configured port (#17) [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
Keep the Backend host-network container reachable on its configured port.
2026-10-07 05:11:52 +00:00
justme 4e1b372e66 fix: bind Backend host service to its configured port [CI]
ci/woodpecker/pr/verify Pipeline was successful
ci/woodpecker/pr/web-build Pipeline was successful
2026-10-07 13:08:08 +08:00
justme 1b052dcaef backend: confirm signed channel updates (#16) [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
Publish signed, channel-matched Backend update offers and host IPv6 diagnostics.
2026-10-07 05:05:08 +00:00
justme 33b9782a1d ci: tolerate legacy Backend release webhook [CI]
ci/woodpecker/pr/verify Pipeline was successful
ci/woodpecker/pr/web-build Pipeline was successful
2026-10-07 13:00:35 +08:00
justme 84f9d211bc backend: require confirmation for signed channel updates [CI]
ci/woodpecker/pr/verify Pipeline failed
ci/woodpecker/pr/web-build unknown status
2026-10-07 12:57:20 +08:00
justme 7160de2a36 shared: merge IPv4 and IPv6 collection (#15) [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
Preserve independent dual-stack measurement paths; PR checks passed.
2026-10-07 03:48:39 +00:00
justme 0eda0600cc shared: add independent IPv4 and IPv6 collection [CI]
ci/woodpecker/pr/verify Pipeline was successful
ci/woodpecker/pr/web-build Pipeline was successful
2026-10-07 11:44:01 +08:00
justme 5bd53f8880 ci: sync main into dev [CI SKIP] 2026-10-07 03:36:55 +08:00
justme c974a7dddb docs: merge the Fleet docs into backend/fleet/README.md, add TODO.md, unify the Backend compose, fix stale references [CI SKIP] 2026-10-07 03:36:34 +08:00
justme 643f9e1ea3 ci: sync main into dev [CI SKIP] 2026-10-07 03:30:19 +08:00
justme ac0bcd8e1f docs: merge the Fleet docs into backend/fleet/README.md, add TODO.md, fix stale references [CI SKIP] 2026-10-07 03:30:03 +08:00
justme f4aa3f2482 docs: audit the docs, merge the Fleet docs into backend/fleet/README.md and add TODO.md [CI SKIP]
Drop the unimplemented fleet updater design, the stale project-layout sections and broken references, fix the API table, and rename the four Monitor*.cs files left over from the rename.
2026-10-07 03:29:05 +08:00
justme 180e54cb3c ci: sync main into dev [CI SKIP] 2026-10-07 03:12:48 +08:00
justme a286a453f4 fix: let the update helper write the status volume owned by the app user [CI SKIP]
The status helper runs as root with all capabilities dropped, so it could not write a volume owned by uid 1654 (mktemp: Permission denied) and the volume updater failed on rootful Podman. Add DAC_OVERRIDE only.
2026-10-07 02:58:34 +08:00
justme 7613f67fed fix: create the update directory for fresh named volumes [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
ConfigStore.Write only created the config root, so a new volume without update/ made update offers fail with IO_PathNotFound. Create the parent directory on write and ship an empty update/ in the node images.
2026-10-07 02:51:39 +08:00
justme ef75c9f510 fix: API tokens use the mos_ prefix [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
2026-10-07 02:31:32 +08:00
justme 18a6898a27 fix: build the Backend image as glibc so it can load libssl [CI]
ci/woodpecker/push/main-amd64-build Pipeline failed
The musl AotAnywhere output is a fully static executable, which cannot dlopen libssl: the Backend exited with 'No usable version of libssl was found' on start. Use linux-x64/arm64 and the Debian runtime-deps base.
2026-10-07 02:21:31 +08:00
geruipu d3fadc0bf8 fix: support rootless ICMP and verify SELinux deployment [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
2026-10-07 02:06:34 +08:00
justme 2e146d4da7 refactor!: rename the project from Monitor to MoShiro [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
Namespaces, assemblies, solution, binaries, images, systemd units, config paths, environment variables, protocol methods and the Linux user are all renamed. Breaking: existing databases, node identities and installed units must be recreated.
2026-10-07 02:03:25 +08:00
justme a19517a57d docs: rewrite the README: scope, trust model, architecture, protocol and build [SKIP TESTS]
ci/woodpecker/push/verify Pipeline was successful
ci/woodpecker/push/web-build Pipeline was successful
2026-10-07 01:51:54 +08:00
justme 83e5120377 refactor!: rename the project from Monitor to MoShiro
Namespaces, assemblies, projects and directories, the solution, binaries,
images, systemd units, config paths, environment variables, protocol method
names, the media type and the raw chunk magic all move to MoShiro/moshiro.
Existing data and deployments are not compatible and must be recreated.
Also fix the architecture tests after the solution layout change and make
the update tests independent of their order.
2026-10-07 01:49:04 +08:00
justme 17b35223aa ci: sync main into dev [CI SKIP] 2026-10-07 01:46:35 +08:00
geruipu 27ac7fb787 feat: share named volumes across Docker and rootless Podman updates [CI]
ci/woodpecker/push/main-amd64-build Pipeline was successful
2026-10-07 01:41:15 +08:00
justme 4d329f5e16 ci: sync main into dev [CI SKIP] 2026-10-07 01:30:46 +08:00
justme 90b3ca4dad docs: update the architecture, README and related docs, drop the roadmap [SKIP TESTS] 2026-10-07 01:30:41 +08:00
justme ad17170664 docs: redraw the architecture diagrams and update the README [SKIP TESTS]
ci/woodpecker/push/verify Pipeline was successful
ci/woodpecker/push/web-build Pipeline was successful
2026-10-07 01:30:28 +08:00
justme 710aa9e99f ci: sync main into dev [CI SKIP] 2026-10-07 01:28:32 +08:00
justme 71d2866b9c docs: update the architecture [SKIP TESTS] 2026-10-07 01:21:49 +08:00
geruipu 103ef27ed3 feat: use Alpine AOT images, static SQLite and Podman host metrics [CI] 2026-10-07 01:03:31 +08:00
justme 48664c0ada docs: remove the roadmap and bring the docs in line with the current code [SKIP TESTS]
Probes link to the panel directly, container nodes update through signed
releases, the frontend exists and the ports changed. The roadmap is gone; its
open items live in the README.
2026-10-06 23:27:26 +08:00
justme 198ee556d6 ci: sync main into dev [CI SKIP] 2026-10-06 23:22:52 +08:00
justme 6e945657c1 feat: choose probe/agent links from the admin, fix the overview popover [CI]
The panel decides which probes link to which agents (probes dial out): an
auto-link setting plus per-pair manual overrides with a link matrix in the
admin, and the hovered overview card is lifted above its neighbours.
2026-10-06 23:22:45 +08:00
justme da465866d5 fix(web): lift the hovered overview card above its neighbours so the popover is not hidden 2026-10-06 23:22:36 +08:00
justme 3a751a0f89 feat: choose which probes link to which agents from the admin
The panel decides the probe/agent pairs and probes always dial out. A new
fleet.auto_link setting connects every probe to every agent (default); with it
off only the pairs chosen by hand connect, and any pair can be forced on or
off. The Probes page has a link matrix with the switch and per-pair toggles.
2026-10-06 23:19:32 +08:00
justme d69581dcb5 docs: mount the host os-release and fix the agent public host after tunnel enrollment [SKIP TESTS] 2026-10-06 23:11:26 +08:00
justme 8f44e924a4 docs: describe container node deployment and the update flow [SKIP TESTS] 2026-10-06 23:02:16 +08:00
justme 0c39485a78 ci: sync main into dev [CI SKIP] 2026-10-06 22:57:11 +08:00
justme d449e50791 feat: update controls in the admin, release token scope, build info in CI images [CI]
The versions page shows each node's update state with update buttons, API
tokens get a release scope that can only register releases (used by CI), and
the CI image and release builds now pass the tag, commit and build time.
2026-10-06 22:57:04 +08:00
justme 7a71bf57f9 feat: update controls in the admin, release token scope, build info in CI images
The versions page shows each node's update state with update buttons, API
tokens get a release scope that can only register releases (used by CI), and
the CI image and release builds now pass the tag, commit and build time.
2026-10-06 22:56:56 +08:00
justme 22a98f2eb0 ci: sync main into dev [CI SKIP] 2026-10-06 22:42:20 +08:00
justme 8dd3209a65 feat!: probes link to the panel; signed release updates for container nodes [CI]
Probes connect to the panel directly and get the agent list as desired state.
CI signs the image digest of agent and probe releases, the panel forwards the
manifest, and a host updater verifies it with a local public key, pulls by
digest and rolls back when the new version is unhealthy.
2026-10-06 22:42:12 +08:00
justme 0a66f5e07a feat: signed releases and a host-side updater for container nodes
CI signs the image digest of each agent and probe release; the panel only
stores and forwards it. Nodes write the manifest to their data directory and
a host updater verifies the signature with a local public key, pulls by
digest, restarts the unit and rolls back when it is unhealthy. The panel can
list releases, ask a node to update and show the reported update state.
2026-10-06 22:37:26 +08:00
justme 48072737c9 feat!: probes connect to the backend directly
A probe now links to the backend like an agent (probe.json carries the
server address). The backend sends the agent list as desired state and the
probe reports its platform and build in its own hello. The agent no longer
relays the list or the probe platform. Existing probes must be re-created.
2026-10-06 22:25:16 +08:00
justme c645851c0e fix(ci): build native node images and add compose files [CI]
Restore with PublishAot so agent and probe images contain NativeAOT
binaries instead of a launcher that needs a .dll, fail the build when the
binary is too small, add agent and probe compose files, and document the
[CI] marker in AGENTS.md.
2026-10-06 21:43:57 +08:00
justme 604cbe3123 fix(ci): restore with PublishAot so node images are native, add compose files
Agent and probe images were built with a restore step that lacked
PublishAot, so the publish silently produced a trimmed launcher that looks
for a .dll. Restore now passes PublishAot and the build fails if the binary
is smaller than 2 MB. Also add agent and probe compose files and document
the [CI] marker in AGENTS.md.
2026-10-06 21:43:46 +08:00
justme e06462f033 ci: build main images [CI] 2026-10-06 21:25:53 +08:00
justme b242701e9e ci: sync main into dev [CI SKIP] 2026-10-06 21:24:47 +08:00
justme 14b0de3544 feat: node info, build info and acrylic web theme
Squash of recent dev work: return_session login, backend switcher, line
detail timeline, overview strips, node platform/country/ISP info with flags
and icons, build info (version, tag, time) behind site.show_build, and the
acrylic display theme.
2026-10-06 21:24:25 +08:00
justme 83f54978d2 feat(web): acrylic theme, node flags and icons, build info
Acrylic cards and topbar over a blurred background, country flags, ISP and
OS icons, per-node version tags, a build info footer and an admin versions
tab. Overview cards drop the sparkline and widen the hover area.
2026-10-06 21:22:04 +08:00
justme 1e5c660d05 feat(backend): report node platform, geo and build info
Agents and probes report OS, arch and build (version, tag, commit, time).
The backend looks up country and ISP of public addresses only, stores them
in node_meta (schema v5) and exposes display info via /api/public/nodes,
/api/public/about and /api/admin/build-info behind site.show_build.
2026-10-06 21:22:04 +08:00
justme 10634bdd80 docs: add project agent instructions [SKIP TESTS]
Merge PR #14.
2026-10-06 09:14:54 +00:00
justme f8175f5734 docs: add project agent instructions [SKIP TESTS] 2026-10-06 17:13:08 +08:00
justme 2e470607fa docs: add project agent instructions [SKIP TESTS] 2026-10-06 17:11:07 +08:00
justme a70b89b95c feat(web): show UDP and TCP strips on overview cards
Add 3/6/12 hour ranges with status strips, latency sparklines and a side popover for the hovered interval.
2026-10-06 16:59:17 +08:00
justme a9538eae4a feat(web): add backend switcher and connection status
Split the login page into backend list and credentials, show connection status in the sidebar and top bar, and render dialogs in a portal.
2026-10-06 16:59:17 +08:00
justme 91c63d0cf2 feat(api): add return_session login for cross-origin clients
Return the session token in the login response instead of a cookie, and accept it as a Bearer token without CSRF.
2026-10-06 16:59:16 +08:00
justme a2a0eee46a docs: explain development and CI workflow [SKIP TESTS]
Merge PR #13.
2026-10-06 08:46:12 +00:00
justme 929882cb04 docs: explain development and CI workflow [SKIP TESTS] 2026-10-06 16:44:27 +08:00
justme 2def8e69f9 ci: publish amd64 dev images to Zot [CI]
Merge PR #12. Publish development images separately from release images.
2026-10-06 08:28:53 +00:00
justme efa17b4485 ci: use SKIP TESTS marker for backend tests [SKIP TESTS] 2026-10-06 16:26:57 +08:00
justme 56d80e8c80 ci: build web after tests and allow skipping tests [CI SKIP] 2026-10-06 16:23:13 +08:00
justme 1da8d083c1 ci: skip backend tests only with CI SKIP suffix [CI SKIP] 2026-10-06 16:21:04 +08:00
justme e67d3294a2 ci: allow skipping verify with CI SKIP suffix [CI SKIP] 2026-10-06 16:18:59 +08:00
justme 383afa777b ci: publish main amd64 images to dev Zot [CI SKIP] 2026-10-06 16:14:50 +08:00
justme da0a0e3443 ci: parallelize opt-in main amd64 builds [CI]
Merge PR #11. Build Backend, Agent, and Probe concurrently on main.
2026-10-06 08:04:13 +00:00
justme 19c6cab104 ci: run main amd64 image builds in parallel [CI SKIP] 2026-10-06 15:59:54 +08:00
justme dd6ac4258d ci: opt in to main amd64 builds [CI]
Merge PR #10. Main Linux amd64 images are built only when explicitly marked.
2026-10-06 07:53:57 +00:00
justme 193fc3a95c feat: improve agent management and speed up tests 2026-10-06 15:48:46 +08:00
justme 46499b10b6 ci: build amd64 on opt-in main commits [CI SKIP] 2026-10-06 15:40:35 +08:00
justme 0f19d5047d ci: make main fastlane manual only [CI SKIP] 2026-10-06 15:18:12 +08:00
justme 134c06d605 merge: bring line detail timeline into dev [CI SKIP] 2026-10-06 14:52:48 +08:00
justme d1be70cda2 feat(web): 线路详情页的时间轴、EWMA 平滑和加载进度
- 先选日期,再在连续 48 小时(或 24 小时)的丢包缩略图上拖选区;选区可以跨零点,
  点轴或刻度时离得近的端点贴过去,悬停预览会动的端点;没有数据的时段画成斜纹
- 刻度下方的时长快捷档:30 分钟、1 小时、3 小时、6 小时、12 小时、整天
- 指标卡片平铺,丢包等标明 UDP/TCP;图表按 UDP、TCP、评分分组
- 图表数据刷新时原地更新(不再整张重建),毫秒到 1 秒以上显示成秒
- EWMA 平滑开关(关/轻/中/重),原始数据以淡线垫底
- 数据加载时卡片上显示进度条和占位;useQuery 区分定时刷新和用户触发的加载
- 前端可以在页面里指定后端地址和令牌(配合后端 Monitor__CorsOrigins)
2026-10-06 14:48:58 +08:00
justme 5cb64d2436 docs: clarify no-op main fastlane for docs
Describe the successful no-op when only documentation changes.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 18:58:35 +00:00
justme 9dcb95ccd0 fix(ci): trigger fastlane on merged main PRs
Handle merged Gitea PRs as well as direct pushes to main.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 18:42:39 +00:00
justme 4bde12b493 fix(ci): trigger fastlane on merged main PRs
Run the Linux amd64 fast-lane after a PR is merged to main, while leaving unmerged PRs as no-ops.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 18:35:24 +00:00
justme c4f1fd9e89 docs: clarify main fastlane merge trigger
Explain that feature PRs update main images after merging.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 18:08:49 +00:00
justme ee24593b2a fix(ci): emit native apphosts in fastlane images
Force native apphost generation and validate fastlane output files.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 17:43:22 +00:00
justme 672f6e3dcb fix(ci): emit native apphosts in fastlane images [CI SKIP] 2026-10-05 10:42:34 -07:00
justme 7cc3afbd9b fix(build): use native NativeAOT SDK in Linux fast lane
Only use the StuDev AotAnywhere SDK when explicitly enabled by Windows cross-platform builds.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 17:37:50 +00:00
justme 9a375ffd9e fix(build): select AotAnywhere explicitly [CI SKIP] 2026-10-05 10:37:10 -07:00
justme 75d6bd5540 fix(ci): use host networking for native builds
Run NativeAOT Buildah steps with host networking on the Linux agent.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 17:31:24 +00:00
justme 407c09f6a5 fix(ci): use host networking for native builds [CI SKIP] 2026-10-05 10:30:35 -07:00
justme d6b919bc42 fix(ci): handle empty changed-file lists
Use a Git diff fallback when Woodpecker provides no changed-file list.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 17:27:35 +00:00
justme fffee8937c ci: sync main into dev [CI SKIP] 2026-10-05 10:26:53 -07:00
justme 91485db500 fix(ci): fall back to git diff for empty path lists [CI SKIP] 2026-10-05 10:25:59 -07:00
justme dcf096431d ci: add dependency-aware main fastlane
Add Linux amd64 fast-lane OCI builds selected by project dependencies.

Co-authored-by: JustMe <mr.toing@outlook.com>
2026-10-05 17:22:26 +00:00
justme fa34869f18 ci: sync main into dev [CI SKIP] 2026-10-05 10:21:22 -07:00
justme deb5a97802 ci: add dependency-aware main fastlane [CI SKIP] 2026-10-05 10:18:22 -07:00
justme 479f8c25b0 feat(ci): publish agent and probe OCI images [CI SKIP] 2026-10-05 09:33:22 -07:00
justme 2b80bb66d1 ci: use local Gitea endpoint for release uploads 2026-10-05 08:44:19 -07:00
justme 078eacf1cb ci: persist Buildah cache for Linux releases 2026-10-05 08:29:08 -07:00
justme deceeb19f3 feat: add cross-platform release pipeline [CI SKIP] 2026-10-04 14:01:26 -07:00
404 changed files with 16414 additions and 3756 deletions

No files matched your search

+5
View File
@@ -0,0 +1,5 @@
ci/build-main-amd64.sh text eol=lf
ci/index-backend-image.sh text eol=lf
ci/publish-release-offer.sh text eol=lf
ci/release-linux.sh text eol=lf
packaging/update/*.sh text eol=lf
+17 -1
View File
@@ -17,11 +17,27 @@ depends_on:
- windows-release
steps:
- name: publish-agent-probe-oci
image: bash
environment:
GITEA_BASE: http://127.0.0.1:3100
GITEA_TOKEN:
from_secret: gitea_token
ZOT_REGISTRY: registry.oeo.one
ZOT_USERNAME: justme
ZOT_SECRET:
from_secret: zot_secret
commands:
- bash ci/release-node-images.sh
depends_on: []
- name: publish-gitea-release
image: bash
environment:
GITEA_BASE: http://192.168.2.162:3100
GITEA_BASE: http://127.0.0.1:3100
GITEA_TOKEN:
from_secret: gitea_token
commands:
- bash ci/publish-gitea-release.sh
depends_on:
- publish-agent-probe-oci
+5
View File
@@ -47,6 +47,11 @@ steps:
ZOT_USERNAME: justme
ZOT_SECRET:
from_secret: zot_secret
RELEASE_API_URL: https://mo.oeo.one
RELEASE_SIGNING_KEY:
from_secret: release_signing_key
RELEASE_API_TOKEN:
from_secret: release_api_token
commands:
- bash ci/index-backend-image.sh
depends_on:
+68
View File
@@ -0,0 +1,68 @@
labels:
ci: domain-linux-native
platform: linux/amd64
clone:
- name: git
image: plugin-git
settings:
lfs: false
when:
- event: push
branch: main
evaluate: 'CI_COMMIT_MESSAGE contains "[CI]"'
concurrency: 1
steps:
- name: build-linux-amd64-backend
image: bash
environment:
NUGET_PACKAGES: /home/justme/.nuget/packages
ZOT_REGISTRY: registry.oeo.one
ZOT_USERNAME: justme
ZOT_SECRET:
from_secret: zot_secret
RELEASE_API_URL: https://mo.oeo.one
RELEASE_SIGNING_KEY:
from_secret: release_signing_key
RELEASE_API_TOKEN:
from_secret: release_api_token
commands:
- bash ci/build-main-amd64.sh backend
depends_on: []
- name: build-linux-amd64-agent
image: bash
environment:
NUGET_PACKAGES: /home/justme/.nuget/packages
ZOT_REGISTRY: registry.oeo.one
ZOT_USERNAME: justme
ZOT_SECRET:
from_secret: zot_secret
RELEASE_API_URL: https://mo.oeo.one
RELEASE_SIGNING_KEY:
from_secret: release_signing_key
RELEASE_API_TOKEN:
from_secret: release_api_token
commands:
- bash ci/build-main-amd64.sh agent
depends_on: []
- name: build-linux-amd64-probe
image: bash
environment:
NUGET_PACKAGES: /home/justme/.nuget/packages
ZOT_REGISTRY: registry.oeo.one
ZOT_USERNAME: justme
ZOT_SECRET:
from_secret: zot_secret
RELEASE_API_URL: https://mo.oeo.one
RELEASE_SIGNING_KEY:
from_secret: release_signing_key
RELEASE_API_TOKEN:
from_secret: release_api_token
commands:
- bash ci/build-main-amd64.sh probe
depends_on: []
+8 -12
View File
@@ -9,22 +9,18 @@ clone:
lfs: false
when:
- event: [push, pull_request]
- event: pull_request
- event: push
branch: dev
# On Gitea PR events, CI_COMMIT_MESSAGE is the PR title.
steps:
- name: backend-tests
image: bash
# [SKIP TESTS] omits Backend tests only; web-build remains a dependent workflow.
when:
- evaluate: 'not (CI_COMMIT_MESSAGE contains "[SKIP TESTS]")'
environment:
NUGET_PACKAGES: /home/justme/.nuget/packages
commands:
- cd backend && dotnet test --solution Monitor.slnx -c Release
- name: web-build
image: bash
environment:
NPM_CONFIG_CACHE: /home/justme/.npm
commands:
- npm ci --prefix web --prefer-offline
- npm run build --prefix web
- npm run lint:css --prefix web
depends_on: []
- cd backend && dotnet test --solution MoShiro.slnx -c Release
+27
View File
@@ -0,0 +1,27 @@
labels:
ci: domain-linux-native
platform: linux/amd64
clone:
- name: git
image: plugin-git
settings:
lfs: false
when:
- event: pull_request
- event: push
branch: dev
depends_on:
- verify
steps:
- name: web-build
image: bash
environment:
NPM_CONFIG_CACHE: /home/justme/.npm
commands:
- npm ci --prefix web --prefer-offline
- npm run build --prefix web
- npm run lint:css --prefix web
+43
View File
@@ -0,0 +1,43 @@
# Agent Instructions
## Project
MoShiro is a self-hosted network path quality monitor. Backend coordinates Agents and Probes, stores measurements, and serves the REST/SSE API. `web/display` is the read-only overview; `web/admin` manages nodes, tests, and settings.
## Branch and workspace safety
- Treat `main` as the stable release branch and `dev` as the integration branch. Keep ordinary implementation work on `dev` or a feature branch; integrate into `main` through a pull request.
- A push to `main` builds images only if its commit message contains `[CI]` (for a merged PR, put it in the PR or merge commit title). Without `[CI]` nothing is built; use `[CI SKIP]` to skip the pipeline entirely. When squashing or pushing to `main`, add `[CI]` unless a build is explicitly not wanted.
- Before editing, inspect `git status`. Preserve pre-existing user changes: do not reset, clean, stash, or include unrelated files in a commit.
- Do not commit credentials, API keys, registry passwords, or generated runtime data. CI credentials belong in Woodpecker secrets (`zot_secret`, `gitea_token`, `release_signing_key` for signing release manifests, `release_api_token` with the `release` scope). The release signing private key never goes into the repository or the panel; only the public key in `packaging/update/` does.
- Keep development OCI images in the `moshiro-dev/` repository path and release images in `moshiro/`. Do not let a development build overwrite release tags such as `latest`.
## Project layout
- `backend/src/MoShiro.Backend/`: service entry point and composition.
- `backend/src/MoShiro.Backend.Api/`: REST endpoints, contracts, and OpenAPI.
- `backend/agent/`, `backend/probe/`: node applications and their READMEs.
- `backend/fleet/`, `backend/shared/`: reusable node framework, measurement engine, protocols, and data formats.
- `web/display/`, `web/admin/`: Vite applications; `web/shared/`: API client, generated types, charts, and shared UI.
- `packaging/`, `scripts/`, `.woodpecker/`, `docs/`: container/package builds, CI workflows, and project documentation.
## Local development and builds
Requirements: .NET SDK 10 and Node.js/npm. The SDK selection is defined by the root `global.json`; NuGet package versions are centralized in the root `Directory.Packages.props`.
- All .NET projects: `dotnet build MoShiro.slnx -c Release` from the repository root.
- Backend build: `cd backend && dotnet build MoShiro.slnx -c Release`
- Backend tests: `cd backend && dotnet test --solution MoShiro.slnx -c Release`
- Frontend dependencies/build: `npm ci --prefix web && npm run build --prefix web`
- Frontend style check: `npm run lint:css --prefix web`
- Local backend: from `backend/`, run `dotnet run --project src/MoShiro.Backend -- --urls http://localhost:5000`.
- Local web apps: from the repository root, run `npm ci --prefix web` then `npm run dev --prefix web`. The current dev config uses display port 5173 and admin port 5175; `/admin/` is proxied through the display server. Check the Vite configs when working on another branch because ports can differ.
- After changing API contracts, regenerate/check `backend/openapi/v1.json` and run `npm run gen:api --prefix web` so frontend API types stay aligned.
Follow the user's requested validation scope. Do not start a full tag/release build unless requested.
## CI and releases
CI triggers and image tags can change; inspect `.woodpecker/*.yml` and `docs/deployment.md` before changing or describing them. In particular, check the distinction between `[CI]` main builds, `[SKIP TESTS]`, and version-tag releases in the current branch.
A version tag `vMAJOR.MINOR.PATCH` runs the full cross-platform release pipeline. Development image builds and releases must remain isolated in their respective Zot repository paths.
@@ -1,10 +1,15 @@
<Project>
<PropertyGroup>
<ManagePackageVersionsCentrally>true</ManagePackageVersionsCentrally>
<!-- Native SQLite source is pinned separately from the JIT/test binary package. -->
<MoShiroSqliteSourceVersion>3530400</MoShiroSqliteSourceVersion>
<MoShiroSqliteSourceYear>2026</MoShiroSqliteSourceYear>
<MoShiroSqliteSourceSha256>1e71ddf93849c6a6ecf58b827c0692073d2dd7ee40196158068f7b29f422e87d</MoShiroSqliteSourceSha256>
</PropertyGroup>
<ItemGroup>
<PackageVersion Include="Microsoft.CodeAnalysis.BannedApiAnalyzers" Version="5.6.0" />
<PackageVersion Include="Microsoft.Data.Sqlite" Version="10.0.12" />
<PackageVersion Include="SQLitePCLRaw.lib.e_sqlite3" Version="3.53.3" />
<PackageVersion Include="Microsoft.AspNetCore.OpenApi" Version="10.0.12" />
<PackageVersion Include="Microsoft.Extensions.ApiDescription.Server" Version="10.0.12" />
<PackageVersion Include="Microsoft.Extensions.TimeProvider.Testing" Version="10.0.0" />
+52
View File
@@ -0,0 +1,52 @@
<Solution>
<Folder Name="/backend/fleet/src/">
<Project Path="backend/fleet/src/Fleet.Core/Fleet.Core.csproj" />
<Project Path="backend/fleet/src/Fleet.Enrollment.Client/Fleet.Enrollment.Client.csproj" />
<Project Path="backend/fleet/src/Fleet.Hosting/Fleet.Hosting.csproj" />
<Project Path="backend/fleet/src/Fleet.Identity/Fleet.Identity.csproj" />
<Project Path="backend/fleet/src/Fleet.Link/Fleet.Link.csproj" />
<Project Path="backend/fleet/src/Fleet.Rpc/Fleet.Rpc.csproj" />
<Project Path="backend/fleet/src/Fleet.Server.Abstractions/Fleet.Server.Abstractions.csproj" />
<Project Path="backend/fleet/src/Fleet.Server.Enrollment/Fleet.Server.Enrollment.csproj" />
<Project Path="backend/fleet/src/Fleet.Server.Link/Fleet.Server.Link.csproj" />
</Folder>
<Folder Name="/backend/fleet/tests/">
<Project Path="backend/fleet/tests/Fleet.Tests/Fleet.Tests.csproj" />
</Folder>
<Folder Name="/backend/fleet/tools/">
<Project Path="backend/fleet/tools/fleet-enroll/fleet-enroll.csproj" />
</Folder>
<Folder Name="/backend/shared/">
<Project Path="backend/shared/MoShiro.Measurement/MoShiro.Measurement.csproj" />
<Project Path="backend/shared/MoShiro.Protocol/MoShiro.Protocol.csproj" />
</Folder>
<Folder Name="/backend/shared/tests/">
<Project Path="backend/shared/tests/MoShiro.Measurement.Tests/MoShiro.Measurement.Tests.csproj" />
<Project Path="backend/shared/tests/MoShiro.Protocol.Tests/MoShiro.Protocol.Tests.csproj" />
</Folder>
<Folder Name="/backend/agent/src/">
<Project Path="backend/agent/src/MoShiro.Agent/MoShiro.Agent.csproj" />
</Folder>
<Folder Name="/backend/probe/src/">
<Project Path="backend/probe/src/MoShiro.Probe/MoShiro.Probe.csproj" />
</Folder>
<Folder Name="/backend/probe/tests/">
<Project Path="backend/probe/tests/MoShiro.Probe.Tests/MoShiro.Probe.Tests.csproj" />
</Folder>
<Folder Name="/backend/src/">
<Project Path="backend/src/MoShiro.Backend.Analysis/MoShiro.Backend.Analysis.csproj" />
<Project Path="backend/src/MoShiro.Backend.Api/MoShiro.Backend.Api.csproj" />
<Project Path="backend/src/MoShiro.Backend.Application/MoShiro.Backend.Application.csproj" />
<Project Path="backend/src/MoShiro.Backend.Gateway/MoShiro.Backend.Gateway.csproj" />
<Project Path="backend/src/MoShiro.Backend.Storage.Sqlite/MoShiro.Backend.Storage.Sqlite.csproj" />
<Project Path="backend/src/MoShiro.Backend/MoShiro.Backend.csproj" />
</Folder>
<Folder Name="/backend/tests/">
<Project Path="backend/tests/MoShiro.Backend.Analysis.Tests/MoShiro.Backend.Analysis.Tests.csproj" />
<Project Path="backend/tests/MoShiro.Backend.Tests/MoShiro.Backend.Tests.csproj" />
<Project Path="backend/tests/MoShiro.TestSupport/MoShiro.TestSupport.csproj" />
</Folder>
<Folder Name="/ci/">
<Project Path="ci/MoShiro.Tar/MoShiro.Tar.csproj" />
</Folder>
</Solution>
+86 -64
View File
@@ -1,94 +1,116 @@
# Monitor
# MoShiro
Monitor 是一个自托管的网络线路质量观测工具。它在 Probe 与多个 Agent 之间执行主动网络测量,记录逐包和逐事件结果;Backend 保存原始数据并生成可查询的线路指标,Web 界面用于查看、比较和管理节点。
MoShiro 是一个自托管的**网络线路质量观测工具**。它在若干 Probe(例如家宽出口)与多台 Agent(例如海外 VPS)之间做主动网络测量,只记录逐包、逐事件的原始数据;Backend 保存这些原始数据,推算延迟、丢包、抖动、吞吐等指标,并回答「如果在这条线路上部署某类服务,体验会怎样」;Web 界面用来查看、比较和管理。
它面向需要比较不同网络出口、评估线路稳定性,或观察特定网络服务体验的场景。Agent 与 Probe 专注于测量和上报;它们不提供远程终端、命令执行或文件管理能力。
## 项目边界
## 能力概览
**它做什么**
- **主动测量**:组合 TCP、UDP 等测量模块,采集延迟、丢包、抖动、吞吐和连接事件。
- **线路分析**:从原始测量数据生成时间序列指标,并比较线路在不同服务模型下的体验。
- **节点管理**:通过 Fleet 节点框架处理身份、注册、连接、任务同步和数据上报。
- **实时面板**:展示端查看线路状态和历史趋势;管理端负责节点、任务和系统配置。
- **自托管**:Backend 使用段文件和 SQLite 保存数据;提供 Linux、systemd 和 OpenWrt 相关部署文件。
- 主动测量:组合 TCP、UDP、TLS、ICMP 等采集积木,只记录时间、序号、大小这类原始事实,**不转发也不记录业务内容**。
- 线路分析:Backend 从原始数据生成时间序列指标,按服务模型(REALITY、Hysteria2、Minecraft 等)评分并对多条线路排序、对比;模型和指标可以事后重算。
- 管理:节点注册与审批、Probe 和 Agent 之间连哪些线路、任务与按需测试、节点版本与更新,都在管理端完成。
**它不做什么**
- **节点是纯采集器**:Agent 和 Probe 没有终端、命令执行、文件读写和自更新能力,也不加载外部代码。更新由节点宿主机上的更新器完成,节点进程只放面板转来的发布清单。
- **不是主机监控**:Agent 只上报 CPU、内存、磁盘、网卡流量和运行时长这类只读的汇总数字,不采集型号、主机名、进程和任何 IP;Probe 不上报主机资源。
- **不暴露地址**:公开接口里没有 IP、城市、ASN;国家和运营商只由 Backend 查询(节点自己不访问任何外部服务)。
- 暂不包含告警(见文末待办)。
## 信任关系
```text
管理员 ── 账号密码(Cookie 会话 + CSRF)或 API 令牌(read · admin · release)──▶ Backend
访客 ── 匿名只读,站点设为公开时才可用(没有 IP、城市、ASN)──────────────▶ Backend 公开接口
节点 ── 加入令牌一次性注册,之后用自己的 ECDSA P-256 密钥签名认证 ───────▶ Backend(只存公钥)
Probe ── 用同一把密钥签名建立测量会话 ───────────────────────────────────▶ Agent(只放行面板名单里的 Probe 公钥)
CI ── 发布私钥对「镜像 digest 清单」签名 ──▶ Backend 原样转发 ──▶ 节点宿主机的更新器用本机公钥验签
```
- **Backend 不信任节点上报的内容**:数据按线路隔离存放;节点的身份靠签名,数据库泄露也无法冒充节点。
- **节点不信任 Backend 的下发**:任务超出节点本机的硬上限(目标网段、码率、包速率、监听端口)就被拒绝;节点不执行任何「执行类」指令。
- **更新不依赖面板的诚实**:清单必须由发布私钥签名;节点宿主机的更新器用本机公钥验签、只认本机配置的镜像仓库、按 digest 拉取、拒绝旧的 `seq`。面板被攻破也推不出恶意镜像。
## 架构
```text
Probe <──测量会话──> Agent ──长连接与数据上报──> Backend <──REST / SSE──> Web 展示端与管理端
├── 段文件:原始数据
└── SQLite:索引、指标与配置
Web(展示端 · 管理端)
▲ REST + SSE
┌──────────────────────────────────────────────────────────┐
│ Backend(面板) │
│ Api · Registry · Planner · Links │
│ Releases · Analysis · LiveHub │
│ Store(段文件 + SQLite) │
└──────────────────────────────────────────────────────────┘
▲ ▲
│ wss 长连接(节点发起) │ wss 长连接(节点发起)
│ 任务、数据、状态、更新通知 │ 身份、Agent 列表、状态、更新通知
┌────────────────────┐ ┌──────────────────────┐
│ Agent × 10+ │ ◀── 测量会话 ── │ Probe × N │
│ 有公网 IP,监听 │ │ 家宽等出口,NAT 后 │
└────────────────────┘ └──────────────────────┘
测量会话:TLS 控制通道 + UDP/TCP 数据通道,由 Probe 主动拨出
```
## 快速开始
两种连接:**Probe 和 Agent 都用长连接连 Backend**(身份、任务、状态、更新通知),**测量只走 Probe → Agent**(Probe 主动拨出,Agent 不需要连 Probe,家宽也不需要入站端口)。哪些 Probe 连哪些 Agent 由 Backend 决定:默认自动全连,也可以在管理端手动选择。
### 环境要求
- .NET SDK 10;项目使用 `backend/global.json` 指定 SDK 版本和滚动策略。
- Node.js 与 npm,用于构建 Web 前端。
```text
采集(只记录) 传输(有界、至少一次) 分析(注册制、可重算) 展示
Probe / Agent 积木 ──▶ 原始列块(10 秒一块)──▶ Backend 段文件 ──▶ 指标与服务模型 ──▶ REST / SSE ──▶ Web
```
### 构建
Backend 用段文件保存原始数据,用 SQLite 保存索引、指标与配置。**完整的架构、数据链路、安全边界和设计决策见 [`docs/architecture.md`](docs/architecture.md)。**
从仓库根目录构建后端:
## 协议
- **控制链路(Fleet)**:节点到 Backend 的 `wss` 长连接(`/api/agent/v1/rpc`),JSON-RPC 2.0 信封加二进制通道;挑战加签名认证,`hello` 协商版本和能力,心跳;**期望状态同步**(Backend 下发完整快照,节点调和并回报已应用的版本和被拒绝的条目);**可靠发件箱**(原始块至少一次投递,按 id 去重)。Fleet 与业务无关,放在 `backend/fleet/`。
- **业务方法**:`moshiro.*`(按需测试、Probe 断开、事件、主机汇总、更新通知等),状态文档有 `task_set`(Agent 的任务集)和 `probe_agents`(Probe 的 Agent 列表)。
- **测量数据面**:Probe 到 Agent 的 TLS 控制通道、UDP 和 TCP 数据通道(默认 `7443/tcp`、`7443/udp`、`7444/tcp`);会话用 `session_id` 和 MAC 认证、不绑定五元组,NAT 映射变化后自动跟随。
- **REST + SSE**:`/api/public`(展示,可匿名)、`/api/admin`(管理);OpenAPI 在 `backend/openapi/v1.json`,前端据此生成类型。
细节见 [`docs/protocol.md`](docs/protocol.md)、[`backend/fleet/README.md`](backend/fleet/README.md)、[`docs/api.md`](docs/api.md)、[`docs/enrollment.md`](docs/enrollment.md)。
## 如何构建
环境:.NET SDK 10(`global.json` 指定版本)、Node.js 与 npm。在仓库根目录:
```sh
cd backend
dotnet build Monitor.slnx -c Release
dotnet build MoShiro.slnx -c Release # Backend、Agent、Probe、Fleet、测试和 CI 工具
npm ci --prefix web && npm run build --prefix web # 展示端和管理端,产物在 web/display/dist、web/admin/dist
```
安装前端依赖并构建展示端和管理端(在仓库根目录运行):
本地运行:先在 `backend/` 里 `dotnet run --project src/MoShiro.Backend -- --urls http://localhost:5000`,再 `npm run dev --prefix web`,展示端是 <http://localhost:5173/>,管理端是 <http://localhost:5173/admin/>。
测试和检查:
```sh
npm ci --prefix web
npm run build --prefix web
dotnet test --solution MoShiro.slnx # 后端和节点的测试
npm run lint:css --prefix web # 前端样式(颜色只能写在主题令牌里)
npm run gen:api --prefix web # 后端改了接口之后重新生成前端类型
```
前端产物位于 `web/display/dist/` 和 `web/admin/dist/`。NativeAOT 目标平台和 Windows 上使用 AotAnywhere 交叉构建的说明见 [`docs/deployment.md`](docs/deployment.md)。
Backend、Agent、Probe 以 OCI 镜像发布(Linux amd64、arm64),Agent 和 Probe 另有 Linux glibc/musl、macOS、Windows 的二进制包。NativeAOT 的构建细节、CI 触发规则(`[CI]`、`[SKIP TESTS]`、版本 tag)、镜像路径、容器节点的运行和更新见 [`docs/deployment.md`](docs/deployment.md);前端见 [`web/README.md`](web/README.md),后端见 [`backend/README.md`](backend/README.md)。
Backend 以 OCI 镜像发布,支持 Linux/amd64 与 Linux/arm64;Agent 和 Probe 发布 Linux glibc/musl x64、arm64,macOS x64/arm64,以及 Windows x64/arm64 二进制包。容器部署配置见 [`compose.yaml`](compose.yaml),发布说明见 [`docs/deployment.md`](docs/deployment.md)。
## 文档
### 本地运行
1. **启动 Backend。** 在仓库根目录运行:
```sh
cd backend
dotnet run --project src/Monitor.Backend -- --urls http://localhost:5000
```
2. **启动前端。** 保持 Backend 运行,在另一个终端的仓库根目录运行:
```sh
npm ci --prefix web
npm run dev --prefix web
```
打开 <http://localhost:5173/> 使用展示端,或访问 <http://localhost:5174/admin/> 进入管理端。依赖安装和前端配置见 [`web/README.md`](web/README.md);Backend 的首次启动与配置见 [`backend/README.md`](backend/README.md)。
## 项目布局
| 路径 | 说明 |
| 文档 | 内容 |
| --- | --- |
| [`backend/`](backend/README.md) | Backend 服务、Fleet 框架、共享测量代码与 .NET 测试 |
| [`backend/agent/`](backend/agent/README.md) | Agent 节点:注册到 Backend,接受 Probe 测量并上报数据 |
| [`backend/probe/`](backend/probe/README.md) | Probe 节点:发起线路测量并将采集数据交给 Agent |
| `backend/fleet/` | 可复用的节点身份、注册和连接框架 |
| `backend/shared/` | 测量引擎、协议和原始数据格式 |
| `web/display/`、`web/admin/` | 展示端和管理端 |
| `web/shared/` | 前端 API、图表、组件与主题 |
| `packaging/`、`scripts/` | 部署、构建与维护脚本 |
| `docs/` | [架构](docs/architecture.md)、[测量模型](docs/measurement.md)、[API](docs/api.md)、[部署](docs/deployment.md)等文档 |
| [`docs/architecture.md`](docs/architecture.md) | 架构、数据链路、安全边界、设计决策 |
| [`docs/security.md`](docs/security.md) | 零信任设计稿(尚未实现):威胁模型、名册和操作员凭证、监听隐身、extension 授权 |
| [`docs/protocol.md`](docs/protocol.md)、[`backend/fleet/README.md`](backend/fleet/README.md) | 线协议、Fleet 节点框架(身份、注册、连接) |
| [`docs/enrollment.md`](docs/enrollment.md) | 注册、身份、Probe 与 Agent 的接入 |
| [`docs/measurement.md`](docs/measurement.md)、[`docs/analysis.md`](docs/analysis.md) | 采集引擎与任务模型、分析层与指标 |
| [`docs/storage.md`](docs/storage.md)、[`docs/query.md`](docs/query.md) | 存储、展示数据查询 |
| [`docs/api.md`](docs/api.md)、[`docs/api-guidelines.md`](docs/api-guidelines.md) | API 规范与约束 |
| [`docs/deployment.md`](docs/deployment.md) | 构建、发布、CI、部署与更新 |
| [`docs/project-layout.md`](docs/project-layout.md)、[`docs/coding-style.md`](docs/coding-style.md) | 仓库结构与依赖规则、代码风格 |
## 测试
代码目录:`backend/`(Backend、`fleet/` 节点框架、`shared/` 测量引擎和协议、`agent/`、`probe/`、`tests/`)、`web/`(`display` 展示端、`admin` 管理端、`shared`)、`packaging/` 和 `scripts/`(部署与维护)、`ci/` 和 `.woodpecker/`(流水线)。
```sh
cd backend
dotnet test --solution Monitor.slnx
```
## 待办
```sh
# 前端样式检查(仓库根目录)
npm run lint:css --prefix web
```
见 [`TODO.md`](TODO.md)。
## 许可证
+48
View File
@@ -0,0 +1,48 @@
# TODO
按主题记录还没做的事,做完就删掉对应的条目。已经实现的东西以 [`docs/`](docs/) 和代码为准。
## 测量
- 路由跳数、路径漂移和 trace
- QUIC 握手耗时:TCP 建连和 TLS 握手已经有了(`connect` 任务)。.NET 的 QUIC 依赖系统里的 msquic 原生库,和节点「只依赖 OpenSSL、NativeAOT 单文件」冲突,要做只能自己写最小的 QUIC Initial 握手探测
- 网页长截图
- pcap / eBPF 采集
## 任务与面板
- 单个 Agent 的任务(目前是线路模板加按 Agent 的覆盖,没有只属于某一个 Agent 的独立任务)
- 原生 webhook 投递;告警(基于 webhook)。实现之前先定约束(只允许 HTTPS、禁止私网和回环、不跟随重定向、限制超时和响应大小、解析后再检查地址),见 [`docs/security.md`](docs/security.md) 第 9.2 节
## 部署
- 后端自动下载前端资源:用 Podman、Docker 部署时,默认拉取最新的前端资源(现在前端要自己构建,镜像里没有)。
[security.md](docs/security.md)
## 访客页面
- 人机验证:访客页面要通过验证才能拉数据,前端访问时静默触发。候选 tiago2/cap 和 Cloudflare Turnstile。
## 更新
- 更新遥测(现在面板只有每个节点的「当前更新状态」)
- **回滚路径的实机测试**:容器节点更新后观察 20 秒、失败回滚,只有单元和脚本测试,没有在真实节点上触发过。
- **更新的灰度和自动策略**:按标签分批发布;`POLICY=auto` 的长期验证。
- **纯二进制节点的更新**(`fleet-updater`、`fleet-release`、Windows 服务和 OpenWrt procd 适配器)。设计要点,实现时照这个来:
- 信任来自签名:只接受离线发布密钥签名的清单,发布源可以是面板、GitHub Releases 或任意 HTTP 服务;面板被攻破也推不出恶意二进制。
- 本地说了算:更新策略(`auto` / `notify` / `pinned:<version>`)和信任的公钥写在节点本地配置里,面板只能「建议」;支持多把公钥以便轮换。
- 清单带 `app`、`channel`、`version`、`min_from`、各 RID 制品的 `sha256` 和大小、`key_id`、签名(对规范化 JSON 签名);本地记录已装过的最高版本,拒绝降级和旧清单重放。
- 流程:下载 → 校验 → 解包到 `versions/<v>/` → 原子切换 `current`(Linux 符号链接,Windows 目录联结)→ 重启服务 → 等健康信标确认 → 失败切回旧版本并把该版本标记为坏版本。
- 权限分离:更新进程只能写安装目录和重启被管理的服务,读不到节点身份私钥。
- OpenWrt:`keep_versions = 1`,下载前检查闪存剩余空间。
- 容器节点已经实现的那一套(清单签名、`seq` 防重放、按 digest 拉取、回滚)见 [`docs/deployment.md`](docs/deployment.md)。
## 安全
- **零信任**:设计稿在 [`docs/security.md`](docs/security.md),按其中第 12 节的顺序实现(阶段零「协议方向声明表」已完成):Probe 自己生成密钥、Agent 预握手敲门、操作员凭证和名册、测量数据的签名和归属校验、入库配额和管理端 CSP、Backend 应用层身份、extension 框架(敏感能力都放进可选安装的独立 extension,核心保持 NativeAOT);可选的 UDP 单包授权、UDP 会话迁移的路径验证。
- **extension**(Web SSH、TTY、更新遥测,敏感 operator):设计时必须保住「节点是纯采集器」这条边界,见 [`README.md`](README.md) 的项目边界。
- 对 UDP 包、控制帧、RawChunk、JSON-RPC 做模糊测试;伪造签名和重放的专项测试。
## 打包
- OpenWrt:procd 脚本、ipk/apk 包、sysupgrade 保留列表,在 aarch64 路由器上长时间运行验证。
- Windows 服务适配。
+1 -1
View File
@@ -13,7 +13,7 @@
<UseArtifactsOutput>true</UseArtifactsOutput>
<ArtifactsPath>$(MSBuildThisFileDirectory)artifacts</ArtifactsPath>
<Deterministic>true</Deterministic>
<Version>0.1.0</Version>
<Version>0.1.1</Version>
<NuGetAudit>false</NuGetAudit>
</PropertyGroup>
<ItemGroup>
+16 -16
View File
@@ -17,33 +17,33 @@
<Project Path="fleet/tools/fleet-enroll/fleet-enroll.csproj" />
</Folder>
<Folder Name="/shared/">
<Project Path="shared/Monitor.Measurement/Monitor.Measurement.csproj" />
<Project Path="shared/Monitor.Protocol/Monitor.Protocol.csproj" />
<Project Path="shared/MoShiro.Measurement/MoShiro.Measurement.csproj" />
<Project Path="shared/MoShiro.Protocol/MoShiro.Protocol.csproj" />
</Folder>
<Folder Name="/shared/tests/">
<Project Path="shared/tests/Monitor.Measurement.Tests/Monitor.Measurement.Tests.csproj" />
<Project Path="shared/tests/Monitor.Protocol.Tests/Monitor.Protocol.Tests.csproj" />
<Project Path="shared/tests/MoShiro.Measurement.Tests/MoShiro.Measurement.Tests.csproj" />
<Project Path="shared/tests/MoShiro.Protocol.Tests/MoShiro.Protocol.Tests.csproj" />
</Folder>
<Folder Name="/agent/src/">
<Project Path="agent/src/Monitor.Agent/Monitor.Agent.csproj" />
<Project Path="agent/src/MoShiro.Agent/MoShiro.Agent.csproj" />
</Folder>
<Folder Name="/probe/src/">
<Project Path="probe/src/Monitor.Probe/Monitor.Probe.csproj" />
<Project Path="probe/src/MoShiro.Probe/MoShiro.Probe.csproj" />
</Folder>
<Folder Name="/probe/tests/">
<Project Path="probe/tests/Monitor.Probe.Tests/Monitor.Probe.Tests.csproj" />
<Project Path="probe/tests/MoShiro.Probe.Tests/MoShiro.Probe.Tests.csproj" />
</Folder>
<Folder Name="/backend/src/">
<Project Path="src/Monitor.Backend.Analysis/Monitor.Backend.Analysis.csproj" />
<Project Path="src/Monitor.Backend.Api/Monitor.Backend.Api.csproj" />
<Project Path="src/Monitor.Backend.Application/Monitor.Backend.Application.csproj" />
<Project Path="src/Monitor.Backend.Gateway/Monitor.Backend.Gateway.csproj" />
<Project Path="src/Monitor.Backend.Storage.Sqlite/Monitor.Backend.Storage.Sqlite.csproj" />
<Project Path="src/Monitor.Backend/Monitor.Backend.csproj" />
<Project Path="src/MoShiro.Backend.Analysis/MoShiro.Backend.Analysis.csproj" />
<Project Path="src/MoShiro.Backend.Api/MoShiro.Backend.Api.csproj" />
<Project Path="src/MoShiro.Backend.Application/MoShiro.Backend.Application.csproj" />
<Project Path="src/MoShiro.Backend.Gateway/MoShiro.Backend.Gateway.csproj" />
<Project Path="src/MoShiro.Backend.Storage.Sqlite/MoShiro.Backend.Storage.Sqlite.csproj" />
<Project Path="src/MoShiro.Backend/MoShiro.Backend.csproj" />
</Folder>
<Folder Name="/backend/tests/">
<Project Path="tests/Monitor.Backend.Analysis.Tests/Monitor.Backend.Analysis.Tests.csproj" />
<Project Path="tests/Monitor.Backend.Tests/Monitor.Backend.Tests.csproj" />
<Project Path="tests/Monitor.TestSupport/Monitor.TestSupport.csproj" />
<Project Path="tests/MoShiro.Backend.Analysis.Tests/MoShiro.Backend.Analysis.Tests.csproj" />
<Project Path="tests/MoShiro.Backend.Tests/MoShiro.Backend.Tests.csproj" />
<Project Path="tests/MoShiro.TestSupport/MoShiro.TestSupport.csproj" />
</Folder>
</Solution>
+13 -13
View File
@@ -1,24 +1,24 @@
# Backend
Backend 是 Monitor 的控制面和数据服务,负责 Agent 注册与连接、任务编排、原始测量数据持久化、指标分析,以及供前端使用的 REST API 和 SSE 实时数据流。
Backend 是 MoShiro 的控制面和数据服务,负责 Agent 注册与连接、任务编排、原始测量数据持久化、指标分析,以及供前端使用的 REST API 和 SSE 实时数据流。
## 组件
| 路径 | 职责 |
| --- | --- |
| `src/Monitor.Backend/` | 应用入口与依赖组合 |
| `src/Monitor.Backend.Api/` | Public、Admin API 和 OpenAPI |
| `src/Monitor.Backend.Gateway/` | Fleet RPC、节点状态与测量数据接收 |
| `src/Monitor.Backend.Application/` | 业务用例、查询、任务和后台服务 |
| `src/Monitor.Backend.Analysis/` | 指标计算、累加器和服务模型 |
| `src/Monitor.Backend.Storage.Sqlite/` | 段文件与 SQLite 存储 |
| `src/MoShiro.Backend/` | 应用入口与依赖组合 |
| `src/MoShiro.Backend.Api/` | Public、Admin API 和 OpenAPI |
| `src/MoShiro.Backend.Gateway/` | Fleet RPC、节点状态与测量数据接收 |
| `src/MoShiro.Backend.Application/` | 业务用例、查询、任务和后台服务 |
| `src/MoShiro.Backend.Analysis/` | 指标计算、累加器和服务模型 |
| `src/MoShiro.Backend.Storage.Sqlite/` | 段文件与 SQLite 存储 |
| `fleet/` | 通用节点身份、注册和连接框架 |
| `shared/` | 测量引擎、协议和数据格式 |
| `tests/` | 集成测试和测试支持代码 |
## 环境要求
使用 .NET SDK 10。仓库根目录的 `backend/global.json` 指定所需 SDK 版本及滚动策略。
使用 .NET SDK 10。仓库根目录的 `global.json` 指定所需 SDK 版本及滚动策略。
## 构建与测试
@@ -26,14 +26,14 @@ Backend 是 Monitor 的控制面和数据服务,负责 Agent 注册与连接
```sh
cd backend
dotnet build Monitor.slnx -c Release
dotnet test --solution Monitor.slnx
dotnet build MoShiro.slnx -c Release
dotnet test --solution MoShiro.slnx
```
发布当前平台的 NativeAOT Backend:
```sh
dotnet publish src/Monitor.Backend -c Release -r <RID> -p:PublishAot=true
dotnet publish src/MoShiro.Backend -c Release -r <RID> -p:PublishAot=true
```
例如 `<RID>` 可为 `win-x64`、`linux-x64` 或 `linux-arm64`。NativeAOT 发布需要对应平台的 C/C++ 工具链。仓库根目录的 `scripts/publish.sh backend <RID>` 可构建单个平台的 Backend 可执行文件;容器发布由 Woodpecker tag 流水线完成,详见 [`../docs/deployment.md`](../docs/deployment.md)。
@@ -46,9 +46,9 @@ dotnet publish src/Monitor.Backend -c Release -r <RID> -p:PublishAot=true
```sh
cd backend
dotnet run --project src/Monitor.Backend -- --urls http://localhost:5000
dotnet run --project src/MoShiro.Backend -- --urls http://localhost:5000
```
首次启动时可用 `Monitor__InitialAdminUser` 和 `Monitor__InitialAdminPassword` 设置初始管理员。数据目录、公开地址、Cookie 和 CORS 等设置通过 `Monitor` 配置节传入;完整配置说明见 [`../docs/deployment.md`](../docs/deployment.md)。站点默认需要登录,可在管理端修改公开设置。
首次启动时可用 `MoShiro__InitialAdminUser` 和 `MoShiro__InitialAdminPassword` 设置初始管理员。数据目录、公开地址、Cookie 和 CORS 等设置通过 `MoShiro` 配置节传入;完整配置说明见 [`../docs/deployment.md`](../docs/deployment.md)。站点默认需要登录,可在管理端修改公开设置。
OpenAPI 文档由构建生成到 `openapi/v1.json`。API 设计和端点说明见 [`../docs/api.md`](../docs/api.md);项目总览见仓库根目录的 [`README.md`](../README.md)。
+4
View File
@@ -8,6 +8,10 @@
<EventSourceSupport>false</EventSourceSupport>
<DebuggerSupport>false</DebuggerSupport>
<HttpActivityPropagationSupport>false</HttpActivityPropagationSupport>
<Http3Support>false</Http3Support>
<MetricsSupport>false</MetricsSupport>
<UseSizeOptimizedLinq>true</UseSizeOptimizedLinq>
<JsonSerializerIsReflectionEnabledByDefault>false</JsonSerializerIsReflectionEnabledByDefault>
<ServerGarbageCollection>false</ServerGarbageCollection>
<ConcurrentGarbageCollection>false</ConcurrentGarbageCollection>
</PropertyGroup>
+13 -7
View File
@@ -1,4 +1,4 @@
# Monitor Agent
# MoShiro Agent
Agent 是部署在 VPS 或其他可达网络节点上的采集进程。它向 Backend 注册并保持连接,接受 Probe 的测量会话,执行线路任务中的 Agent 侧工作,再将两端数据合并后上报。
@@ -14,18 +14,18 @@ Probe ──TLS 控制 / UDP、TCP 测量──> Agent ──长连接与数据
## 构建
需要 .NET SDK 10。为应用 `backend/global.json` 的 SDK 设置,请从 `backend/` 目录构建:
需要 .NET SDK 10。为应用 根目录的 `global.json` 的 SDK 设置,请从 `backend/` 目录构建:
```sh
cd backend
dotnet build agent/src/Monitor.Agent -c Release
dotnet build agent/src/MoShiro.Agent -c Release
dotnet build fleet/tools/fleet-enroll -c Release
```
发布 NativeAOT 程序需要对应平台的 C/C++ 工具链;`<RID>` 应与构建环境和目标平台匹配:
```sh
dotnet publish agent/src/Monitor.Agent -c Release -r <RID> -p:PublishAot=true
dotnet publish agent/src/MoShiro.Agent -c Release -r <RID> -p:PublishAot=true
dotnet publish fleet/tools/fleet-enroll -c Release -r <RID> -p:PublishAot=true
```
@@ -36,7 +36,7 @@ dotnet publish fleet/tools/fleet-enroll -c Release -r <RID> -p:PublishAot=true
先在管理端创建加入令牌,然后在 Agent 主机上运行一次性注册工具:
```sh
fleet-enroll --server https://<backend> --join <token> --name <agent-name> --identity-dir /etc/monitor
fleet-enroll --server https://<backend> --join <token> --name <agent-name> --identity-dir /etc/moshiro
```
注册工具会在身份目录写入节点身份和证书。准备本地 `agent.json`,其中 `fleet.server` 指向 Backend,`fleet.identity_dir` 指向上述目录;`listen`、`limits` 和 `reporting` 可按节点环境调整。配置文件不会由 Backend 修改。限制目标网段和测量预算时,应按实际网络与机器能力配置;`allowed_target_cidrs` 默认为空,不允许 Agent 执行目标探测任务。
@@ -44,10 +44,14 @@ fleet-enroll --server https://<backend> --join <token> --name <agent-name> --ide
启动 Agent:
```sh
monitor-agent --config /etc/monitor/agent.json
moshiro-agent --config /etc/moshiro/agent.json
```
不传 `--config` 时默认读取 `/etc/monitor/agent.json`。Agent 默认监听控制端口 `7443/tcp`,数据端口 `7443/udp` 和 `7444/tcp`;防火墙和面板中的节点端口配置需与本地配置一致。`reporting.host_stats` 可关闭主机状态上报;启用时只上报资源使用等汇总数据,不包含主机名、进程或 IP。
容器节点的运行方式(podman + systemd)和宿主机更新器见 [`../../docs/deployment.md`](../../docs/deployment.md)。也可使用 OCI 镜像 `registry.oeo.one/moshiro/moshiro-agent:<version>`,自动选择 Linux amd64 或 arm64 镜像。镜像同时包含 `fleet-enroll`;可通过 `--entrypoint /usr/local/bin/fleet-enroll` 执行首次注册。容器默认读取命名 volume 中的 `/etc/moshiro/agent.json` 与身份文件;建议使用宿主网络,Linux ICMP 优先使用普通用户 ping socket,内核 ping_group_range 需允许映射后的 GID。镜像以 UID/GID 1654 运行,新命名 volume 由引擎按镜像目录初始化权限。
不传 `--config` 时默认读取 `/etc/moshiro/agent.json`。Agent 默认监听控制端口 `7443/tcp`,数据端口 `7443/udp` 和 `7444/tcp`;防火墙和面板中的节点端口配置需与本地配置一致。`reporting.host_stats` 可关闭主机状态上报;启用时只上报资源使用等汇总数据,不包含主机名、进程或 IP。
IPv6 线路需要 Agent 实际绑定 IPv6。新配置的默认 `listen.bind` 为 `::`,覆盖整个双栈接口;已有配置如果仍写 `0.0.0.0`,可改为 `::` 后重启。Compose 与 systemd 容器部署使用宿主网络,诊断反映宿主 IPv6;自定义容器网络时,Agent 同时报告容器网络以及可读取到的宿主 `/proc` IPv6 状态。诊断不上传 IP 或接口名称。
## 设计说明
@@ -55,3 +59,5 @@ monitor-agent --config /etc/monitor/agent.json
- 数据面会校验会话认证、重放窗口和速率限制;无效数据包会静默丢弃。
- 测量引擎和协议由 `backend/shared/` 提供;节点身份、注册和连接由 `backend/fleet/` 提供。
- 详细安全边界与拓扑见 [`../../docs/architecture.md`](../../docs/architecture.md),协议说明见 [`../../docs/protocol.md`](../../docs/protocol.md),部署见 [`../../docs/deployment.md`](../../docs/deployment.md)。
Docker 与 rootless/rootful Podman 共用 `packaging/compose/agent/compose.yaml`,配置、身份和缓存放在命名 volume 中,无需宿主配置目录、keep-id 或手动 chown。首次导入和自动更新见部署文档。
@@ -4,17 +4,20 @@ using Fleet.Core;
using Fleet.Hosting;
using Fleet.Link;
using Fleet.Rpc;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>Agent 的组装点:数据面、点对点监听、与 Backend 的链路都在这里接起来。没有全局状态,测试可以在同一进程里启动多个。</summary>
public sealed class AgentApp : IDisposable
{
public const string Version = "0.1.0";
public const string Version = MoShiro.Protocol.BuildInfo.Version;
private readonly AgentConfig _config;
private readonly bool _ipv6Supported;
private readonly RestartController _restart;
private readonly RestartController? _ownedRestart;
private readonly LoadedNodeIdentity _identity;
private readonly TimeProvider _time;
private readonly INodeLog _log;
@@ -25,6 +28,7 @@ public sealed class AgentApp : IDisposable
private readonly TaskCompletionSource _ready = new(TaskCreationOptions.RunContinuationsAsynchronously);
private readonly AgentAdmission _admission;
private readonly Outbox _outbox = new(64 << 20);
private readonly UpdateBridge _update;
private readonly BackendUplink _uplink;
private readonly UdpMux _udp;
private readonly TcpAcceptor _tcp;
@@ -34,8 +38,12 @@ public sealed class AgentApp : IDisposable
private readonly TestRunner _tests;
private readonly FleetLink _link;
public AgentApp(AgentConfig config, LoadedNodeIdentity identity, TimeProvider time, INodeLog log)
/// <param name="dataDirectory">配置目录(容器里的 <c>/etc/moshiro</c>):宿主机更新器和节点在这里交换清单、请求和状态。</param>
public AgentApp(AgentConfig config, LoadedNodeIdentity identity, string dataDirectory, TimeProvider time, INodeLog log, RestartController? restart = null)
{
_ownedRestart = restart is null ? new RestartController() : null;
_restart = restart ?? _ownedRestart!;
_update = new UpdateBridge(dataDirectory, new ConfigStore(dataDirectory).Write, log);
_config = config;
_identity = identity;
_time = time;
@@ -44,6 +52,12 @@ public sealed class AgentApp : IDisposable
_uplink = new BackendUplink(time);
IPAddress bind = IPAddress.Parse(config.Listen.Bind);
if (bind.Equals(IPAddress.IPv6Any) && !System.Net.Sockets.Socket.OSSupportsIPv6)
{
bind = IPAddress.Any;
}
_ipv6Supported = bind.AddressFamily == System.Net.Sockets.AddressFamily.InterNetworkV6;
_admission = new AgentAdmission(_streams);
_udp = new UdpMux(_sessions, _admission, _clock, _counters.Data, log, bind, config.Listen.UdpPorts);
_tcp = new TcpAcceptor(bind, config.Listen.TcpPorts, identity.Certificate, _sessions, _counters.Data);
@@ -90,7 +104,11 @@ public sealed class AgentApp : IDisposable
/// <summary>数据面和控制端口都开始监听之后完成(配置里的端口为 0 时,这里能读到实际端口)。</summary>
public Task Ready => _ready.Task;
public void Dispose() => _udp.Dispose();
public void Dispose()
{
_udp.Dispose();
_ownedRestart?.Dispose();
}
internal StreamRegistry Streams => _streams;
@@ -104,6 +122,9 @@ public sealed class AgentApp : IDisposable
public FleetLink Link => _link;
/// <summary>对等面(Probe 连进来的会话)的路由器,测试用。</summary>
public RpcRouter<PeerContext> PeerRouter => _hub.Router;
/// <summary>期望状态里是否已经有这个 Probe 的公钥(测试用:握手会按它校验)。</summary>
public bool IsPeerKnown(string probeId, string publicKeyBase64) =>
_tasks.LookupPeerKey(probeId) is { } key && Convert.ToBase64String(key) == publicKeyBase64;
@@ -178,43 +199,66 @@ public sealed class AgentApp : IDisposable
private FleetLink BuildLink()
{
var router = new RpcRouter<LinkContext>();
var router = new RpcRouter<LinkContext>(RpcSide.Node, MoShiroMethodTable.Table, NodeRoles.Agent);
HashSet<string>? allowed = _config.Policy.AllowMethods is { } list ? [.. list] : null;
bool Allow(string method) => allowed is null || allowed.Contains(method);
if (Allow(MonitorMethods.TestRun))
if (Allow(MoShiroMethods.TestRun))
{
router.On(
MonitorMethods.TestRun,
MonitorJsonContext.Default.TestRunParams,
MonitorJsonContext.Default.TestRunResult,
MoShiroMethods.TestRun,
MoShiroJsonContext.Default.TestRunParams,
MoShiroJsonContext.Default.TestRunResult,
(_, run, _) => ValueTask.FromResult(_tests.Start(run)));
}
if (Allow(MonitorMethods.TestCancel))
if (Allow(MoShiroMethods.TestCancel))
{
router.On(
MonitorMethods.TestCancel,
MonitorJsonContext.Default.TestCancelParams,
MonitorJsonContext.Default.TestRunResult,
MoShiroMethods.TestCancel,
MoShiroJsonContext.Default.TestCancelParams,
MoShiroJsonContext.Default.TestRunResult,
(_, cancel, _) => ValueTask.FromResult(Result<TestRunResult>.Ok(new TestRunResult { Accepted = _tests.Cancel(cancel.TestId) })));
}
if (Allow(MonitorMethods.ProbeDisconnect))
if (Allow(MoShiroMethods.UpdateOffer))
{
router.On(
MonitorMethods.ProbeDisconnect,
MonitorJsonContext.Default.ProbeDisconnectParams,
MonitorJsonContext.Default.TestRunResult,
async (_, request, _) => Result<TestRunResult>.Ok(new TestRunResult { Accepted = await _hub.DisconnectAsync(request.ProbeId.Value) }));
router.OnNotify(MoShiroMethods.UpdateOffer, MoShiroJsonContext.Default.UpdateOfferParams, (_, offer, _) =>
{
_update.WriteOffer(offer);
return ValueTask.CompletedTask;
});
}
if (Allow(MonitorMethods.ProbeAgentsChanged))
if (Allow(MoShiroMethods.Schedule))
{
router.OnNotify(
MonitorMethods.ProbeAgentsChanged,
MonitorJsonContext.Default.ProbeAgentsResult,
(_, list, _) => _hub.RelayAgentsAsync(list));
router.OnNotify(MoShiroMethods.Schedule, MoShiroJsonContext.Default.ScheduleParams, (_, schedule, _) =>
{
if (!_restart.SetRemote(schedule.RestartAt))
{
_log.Warn("ignoring an invalid restart schedule from the panel");
}
return ValueTask.CompletedTask;
});
}
if (Allow(MoShiroMethods.UpdateApply))
{
router.OnNotify(MoShiroMethods.UpdateApply, MoShiroJsonContext.Default.UpdateOfferParams, (_, _, _) =>
{
_update.RequestApply();
return ValueTask.CompletedTask;
});
}
if (Allow(MoShiroMethods.ProbeDisconnect))
{
router.On(
MoShiroMethods.ProbeDisconnect,
MoShiroJsonContext.Default.ProbeDisconnectParams,
MoShiroJsonContext.Default.TestRunResult,
async (_, request, _) => Result<TestRunResult>.Ok(new TestRunResult { Accepted = await _hub.DisconnectAsync(request.ProbeId.Value) }));
}
var options = new FleetLinkOptions
@@ -223,11 +267,11 @@ public sealed class AgentApp : IDisposable
Identity = _identity.Identity,
NodeId = _identity.NodeId,
NodeVersion = Version,
Capabilities = [MonitorCapabilities.Base, MonitorCapabilities.KindsPrefix + string.Join(',', BlockCatalog.AgentKinds)],
Capabilities = [MoShiroCapabilities.Base, MoShiroCapabilities.KindsPrefix + string.Join(',', BlockCatalog.AgentKinds)],
HelloMeta = HelloMeta,
HeartbeatExtra = HeartbeatExtra,
};
var link = new FleetLink(options, router, [new StateHandler<TaskSet>(TaskSet.StateType, MonitorJsonContext.Default.TaskSet, _tasks.ApplyAsync)], _outbox, _time, _log);
var link = new FleetLink(options, router, [new StateHandler<TaskSet>(TaskSet.StateType, MoShiroJsonContext.Default.TaskSet, _tasks.ApplyAsync)], _outbox, _time, _log);
// 链路重新连上之后补报:在线的 Probe 都告诉 Backend(Backend 重启或链路抖动期间错过了 probe_connected)
link.OnConnected = async (_, _) =>
@@ -244,6 +288,9 @@ public sealed class AgentApp : IDisposable
{
var meta = new AgentHelloMeta
{
Ipv6 = Ipv6Diagnostics.Sample(_ipv6Supported, Environment.GetEnvironmentVariable("MOSHIRO_NETWORK_SCOPE"), Environment.GetEnvironmentVariable("HOST_PROC")),
Ipv6Supported = _ipv6Supported,
Platform = MoShiro.Measurement.PlatformDetector.Detect(),
TlsSpki = _identity.File.TlsSpki,
ProbeTcpPort = ControlPort,
ProbeUdpPorts = [.. UdpPorts],
@@ -255,7 +302,7 @@ public sealed class AgentApp : IDisposable
MaxConcurrentTests = _config.Limits.MaxConcurrentTests,
},
};
return JsonSerializer.SerializeToElement(meta, MonitorJsonContext.Default.AgentHelloMeta);
return JsonSerializer.SerializeToElement(meta, MoShiroJsonContext.Default.AgentHelloMeta);
}
private long _lastIn;
@@ -270,6 +317,7 @@ public sealed class AgentApp : IDisposable
long packetsOut = _counters.PacketsOut;
var extra = new AgentHeartbeatExtra
{
Ipv6 = Ipv6Diagnostics.Sample(_ipv6Supported, Environment.GetEnvironmentVariable("MOSHIRO_NETWORK_SCOPE"), Environment.GetEnvironmentVariable("HOST_PROC")),
Probes = [.. _hub.ConnectedProbes()],
Sessions = _sessions.All.Count,
PpsIn = _lastAtMs == 0 ? 0 : (packetsIn - _lastIn) / seconds,
@@ -277,10 +325,11 @@ public sealed class AgentApp : IDisposable
DroppedInvalid = _counters.DroppedInvalid,
DroppedRateLimited = _counters.DroppedRateLimited,
QueueBytes = _outbox.PendingBytes,
Update = _update.ReadStatus(),
};
_lastIn = packetsIn;
_lastOut = packetsOut;
_lastAtMs = nowMs;
return JsonSerializer.SerializeToElement(extra, MonitorJsonContext.Default.AgentHeartbeatExtra);
return JsonSerializer.SerializeToElement(extra, MoShiroJsonContext.Default.AgentHeartbeatExtra);
}
}
@@ -1,6 +1,6 @@
using System.Text.Json.Serialization;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>
/// <c>agent.json</c>:节点本地配置,面板无法修改。其中 <c>limits</c> 是本地硬上限:期望任务集超出这些范围时,
@@ -22,6 +22,10 @@ public sealed class AgentConfig
[JsonPropertyName("reporting")]
public ReportingSection Reporting { get; set; } = new();
/// <summary>每天定时重启一次,例如 <c>04:00+08:00</c>(不写偏移按 UTC);不写就不重启。环境变量 <c>MOSHIRO_RESTART_AT</c> 优先。</summary>
[JsonPropertyName("restart_at")]
public string? RestartAt { get; set; }
}
public sealed class ReportingSection
@@ -40,13 +44,13 @@ public sealed class FleetSection
public string Server { get; set; } = "";
[JsonPropertyName("identity_dir")]
public string IdentityDir { get; set; } = "/etc/monitor";
public string IdentityDir { get; set; } = "/etc/moshiro";
}
public sealed class ListenSection
{
[JsonPropertyName("bind")]
public string Bind { get; set; } = "0.0.0.0";
public string Bind { get; set; } = "::";
/// <summary>Probe 控制通道(TLS,点对点会话)。</summary>
[JsonPropertyName("control_port")]
@@ -1,9 +1,9 @@
using Fleet.Core;
using Fleet.Link;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>
/// Probe 的列块到达之后:找到对应的流 → 与 Agent 在数据面记录的列合并 → 盖章(块头、时钟锚点)→ 编码 → 放进发件箱(通道 16)。
@@ -44,6 +44,7 @@ public sealed class ChunkAssembler
_ => null,
};
header.ProbeId = new ProbeId(probe.ProbeId);
header.IpVersion = probe.IpVersion; // 地址族以认证过的控制会话为准,不接受列块自报值
KindSpec? spec = stream is null ? null : Kinds.Find(stream.Kind);
if (stream is null || spec is null || spec.Id != header.Kind)
{
@@ -70,6 +71,6 @@ public sealed class ChunkAssembler
/// <summary>Agent 自己采集的块(目标任务)直接发出。</summary>
public void Emit(RawChunk chunk)
{
_outbox.Enqueue(MonitorChannels.RawChunk, chunk.Header.ChunkId.Value, RawChunkCodec.Encode(chunk));
_outbox.Enqueue(MoShiroChannels.RawChunk, chunk.Header.ChunkId.Value, RawChunkCodec.Encode(chunk));
}
}
@@ -4,10 +4,10 @@ using System.Net.Sockets;
using Fleet.Core;
using Fleet.Link;
using Fleet.Rpc;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>进程内计数(只有计数,没有主机信息):用于心跳和面板上的节点状态。数据面的包计数由共享的 <see cref="DataPlaneCounters"/> 提供。</summary>
public sealed class AgentCounters
@@ -56,6 +56,10 @@ public sealed class ProbeSession : DataSession
public string ProbeId => Info.PeerId;
public string IpVersion => Info.Remote is IPEndPoint endpoint && endpoint.Address.AddressFamily == AddressFamily.InterNetworkV6 && !endpoint.Address.IsIPv4MappedToIPv6 ? IpVersions.V6 : IpVersions.V4;
public string PathKey => ProbeId + "~" + IpVersion;
/// <summary>Probe 的程序版本(握手里自报,只用于展示)。</summary>
public string? Version => Info.Version;
@@ -72,7 +76,7 @@ public sealed class ProbeSession : DataSession
get
{
SocketAddress? source = Volatile.Read(ref _udpSource);
return source is null ? null : (IPEndPoint)new IPEndPoint(IPAddress.Any, 0).Create(source);
return source is null ? null : (IPEndPoint)new IPEndPoint(source.Family == AddressFamily.InterNetworkV6 ? IPAddress.IPv6Any : IPAddress.Any, 0).Create(source);
}
}
@@ -125,7 +129,7 @@ public sealed class ProbeSession : DataSession
public void ForgetStream(uint streamId) => _windows.TryRemove(streamId, out _);
}
/// <summary>当前连着的 Probe 会话:每个 Probe 一个;同一个 Probe 重连时新会话取代它自己的旧会话。</summary>
/// <summary>每个 Probe 的 IPv4、IPv6 会话独立;重连只替换同地址族的旧会话。</summary>
public sealed class SessionRegistry : IDataSessions
{
private readonly object _gate = new();
@@ -143,11 +147,11 @@ public sealed class SessionRegistry : IDataSessions
}
}
public ProbeSession? Get(string probeId)
public ProbeSession? Get(string probeId, string ipVersion = IpVersions.V4)
{
lock (_gate)
{
return _byProbe.GetValueOrDefault(probeId);
return _byProbe.GetValueOrDefault(probeId + "~" + ipVersion);
}
}
@@ -156,13 +160,13 @@ public sealed class SessionRegistry : IDataSessions
{
lock (_gate)
{
_byProbe.TryGetValue(session.ProbeId, out ProbeSession? old);
_byProbe.TryGetValue(session.PathKey, out ProbeSession? old);
if (old is not null)
{
_bySession.Remove(old.SessionId);
}
_byProbe[session.ProbeId] = session;
_byProbe[session.PathKey] = session;
_bySession[session.SessionId] = session;
return old;
}
@@ -172,9 +176,9 @@ public sealed class SessionRegistry : IDataSessions
{
lock (_gate)
{
if (_byProbe.TryGetValue(session.ProbeId, out ProbeSession? current) && ReferenceEquals(current, session))
if (_byProbe.TryGetValue(session.PathKey, out ProbeSession? current) && ReferenceEquals(current, session))
{
_byProbe.Remove(session.ProbeId);
_byProbe.Remove(session.PathKey);
}
if (_bySession.TryGetValue(session.SessionId, out ProbeSession? bySession) && ReferenceEquals(bySession, session))
@@ -1,28 +1,35 @@
using System.Globalization;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>
/// 主机资源占用的只读快照:CPU 占用、内存、根分区、网卡累计流量与瞬时网速、系统运行时长。
/// 只有汇总的数字:<b>不采集 CPU 型号、主机名、网卡名和任何 IP 地址</b>,所以面板和展示页想泄漏也泄漏不出去。
/// Linux 读 <c>/proc</c>;读不到的项是 null(例如 Windows 上的 CPU)。不执行任何命令,不写文件。
/// Linux 读 <c>/proc</c> 和 <c>/sys</c>;容器通过 HOST_PROC、HOST_SYS、HOST_ROOT 指定宿主路径。
/// 读不到的项是 null(例如 Windows 上的 CPU)。不执行任何命令,不写文件。
/// </summary>
public sealed class HostSampler
{
private static readonly string[] s_virtualPrefixes = ["lo", "veth", "docker", "br-", "virbr", "cni", "flannel", "cali", "tun", "tap"];
private static readonly string[] s_virtualPrefixes = ["lo", "veth", "docker", "podman", "br-", "virbr", "cni", "flannel", "cali", "tun", "tap"];
private readonly TimeProvider _time;
private readonly Func<string, string?> _read;
private readonly Func<(long Total, long Used)?> _disk;
private readonly string _procRoot;
private readonly string _sysRoot;
private (long Total, long Idle)? _previousCpu;
private (long At, long Rx, long Tx)? _previousNet;
public HostSampler(TimeProvider time, Func<string, string?>? read = null, Func<(long Total, long Used)?>? disk = null)
public HostSampler(TimeProvider time, Func<string, string?>? read = null, Func<(long Total, long Used)?>? disk = null,
string? procRoot = null, string? sysRoot = null, string? diskRoot = null)
{
_time = time;
_read = read ?? ReadProc;
_disk = disk ?? ReadRootDisk;
_procRoot = Root(procRoot, "HOST_PROC", "/proc");
_sysRoot = Root(sysRoot, "HOST_SYS", "/sys");
string root = Root(diskRoot, "HOST_ROOT", OperatingSystem.IsWindows() ? Path.GetPathRoot(Environment.SystemDirectory) ?? "C:\\" : "/");
_disk = disk ?? (() => ReadRootDisk(root));
}
public HostSnapshot Sample()
@@ -30,7 +37,7 @@ public sealed class HostSampler
long now = _time.GetTimestamp();
var snapshot = new HostSnapshot { AtUtcMs = _time.GetUtcNow().ToUnixTimeMilliseconds() };
if (_read("/proc/stat") is { } stat && ParseCpu(stat) is { } cpu)
if (_read(Path.Combine(_procRoot, "stat")) is { } stat && ParseCpu(stat) is { } cpu)
{
if (_previousCpu is { } before && cpu.Total > before.Total)
{
@@ -41,7 +48,7 @@ public sealed class HostSampler
_previousCpu = cpu;
}
if (_read("/proc/meminfo") is { } meminfo && ParseMemory(meminfo) is { } memory)
if (_read(Path.Combine(_procRoot, "meminfo")) is { } meminfo && ParseMemory(meminfo) is { } memory)
{
snapshot.MemTotal = memory.Total;
snapshot.MemUsed = memory.Used;
@@ -53,10 +60,11 @@ public sealed class HostSampler
snapshot.DiskUsed = disk.Used;
}
if (_read("/proc/net/dev") is { } dev && ParseNetwork(dev) is { } net)
if (_read(Path.Combine(_procRoot, "net/dev")) is { } dev && ParseNetwork(dev) is { } net)
{
snapshot.NetRxTotal = net.Rx;
snapshot.NetTxTotal = net.Tx;
snapshot.NetLinkMbps = ReadLinkSpeed(dev);
if (_previousNet is { } before && net.Rx >= before.Rx && net.Tx >= before.Tx)
{
double seconds = Math.Max(0.001, _time.GetElapsedTime(before.At, now).TotalSeconds);
@@ -67,7 +75,11 @@ public sealed class HostSampler
_previousNet = (now, net.Rx, net.Tx);
}
snapshot.UptimeSeconds = _read("/proc/uptime") is { } uptime && ParseUptime(uptime) is { } up ? up : Environment.TickCount64 / 1000;
snapshot.UptimeSeconds = _read(Path.Combine(_procRoot, "uptime")) is { } uptime && ParseUptime(uptime) is { } up ? up : null;
if (!OperatingSystem.IsLinux() && snapshot.UptimeSeconds is null)
{
snapshot.UptimeSeconds = Environment.TickCount64 / 1000;
}
return snapshot;
}
@@ -165,11 +177,54 @@ public sealed class HostSampler
}
}
private static (long Total, long Used)? ReadRootDisk()
private static string Root(string? value, string variable, string fallback)
{
value ??= Environment.GetEnvironmentVariable(variable);
return string.IsNullOrWhiteSpace(value) ? fallback : value;
}
private long? ReadLinkSpeed(string dev)
{
long total = 0;
bool any = false;
foreach (string line in dev.Split('\n'))
{
int colon = line.IndexOf(':', StringComparison.Ordinal);
if (colon < 0)
{
continue;
}
string name = line[..colon].Trim();
if (s_virtualPrefixes.Any(p => name.StartsWith(p, StringComparison.Ordinal)))
{
continue;
}
string? speed = _read(Path.Combine(_sysRoot, "class/net", name, "speed"));
if (long.TryParse(speed?.Trim(), CultureInfo.InvariantCulture, out long mbps) && mbps > 0)
{
total += mbps;
any = true;
}
else
{
return null;
}
}
return any ? total : null;
}
private static (long Total, long Used)? ReadRootDisk(string root)
{
try
{
string root = OperatingSystem.IsWindows() ? Path.GetPathRoot(Environment.SystemDirectory) ?? "C:\\" : "/";
if (!Directory.Exists(root))
{
return null;
}
var drive = new DriveInfo(root);
return drive.IsReady ? (drive.TotalSize, drive.TotalSize - drive.AvailableFreeSpace) : null;
}
@@ -0,0 +1,26 @@
<Project>
<Import Project="Sdk.props" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.props" Sdk="StuDev.AotAnywhere"
Condition="'$(UseAotAnywhere)' == 'true'" />
<PropertyGroup>
<OutputType>Exe</OutputType>
<AssemblyName>moshiro-agent</AssemblyName>
<RootNamespace>MoShiro.Agent</RootNamespace>
<IsAotCompatible>true</IsAotCompatible>
<PublishAot>true</PublishAot>
<PublishTrimmed>true</PublishTrimmed>
<TrimMode>full</TrimMode>
<StripSymbols>true</StripSymbols>
<AllowUnsafeBlocks>true</AllowUnsafeBlocks>
<MetadataUpdaterSupport>false</MetadataUpdaterSupport>
</PropertyGroup>
<ItemGroup>
<InternalsVisibleTo Include="MoShiro.Backend.Tests" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Link\Fleet.Link.csproj" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Hosting\Fleet.Hosting.csproj" />
<ProjectReference Include="..\..\..\shared\MoShiro.Measurement\MoShiro.Measurement.csproj" />
</ItemGroup>
<Import Project="Sdk.targets" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.targets" Sdk="StuDev.AotAnywhere"
Condition="'$(UseAotAnywhere)' == 'true'" />
</Project>
@@ -2,10 +2,10 @@ using System.Text.Json;
using Fleet.Core;
using Fleet.Link;
using Fleet.Rpc;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>Agent → Backend 的上行:事件和测试状态是尽力而为的通知(链路断开时丢弃;列块走发件箱,不会丢)。</summary>
public sealed class BackendUplink
@@ -31,7 +31,7 @@ public sealed class BackendUplink
}
var parameters = new AgentEventParams { Kind = kind, AtUtcMs = _time.GetUtcNow().ToUnixTimeMilliseconds(), Detail = detail, Probe = probe };
await SafeAsync(() => session.NotifyAsync(MonitorMethods.Event, parameters, MonitorJsonContext.Default.AgentEventParams, CancellationToken.None));
await SafeAsync(() => session.NotifyAsync(MoShiroMethods.Event, parameters, MoShiroJsonContext.Default.AgentEventParams, CancellationToken.None));
}
public async ValueTask SendHostAsync(HostSnapshot snapshot)
@@ -42,7 +42,7 @@ public sealed class BackendUplink
return;
}
await SafeAsync(() => session.NotifyAsync(MonitorMethods.Host, snapshot, MonitorJsonContext.Default.HostSnapshot, CancellationToken.None));
await SafeAsync(() => session.NotifyAsync(MoShiroMethods.Host, snapshot, MoShiroJsonContext.Default.HostSnapshot, CancellationToken.None));
}
public async ValueTask SendTestStatusAsync(TestId test, string status, int? phase = null, string? errorCode = null)
@@ -54,19 +54,7 @@ public sealed class BackendUplink
}
var parameters = new TestStatusParams { TestId = test, Status = status, Phase = phase, ErrorCode = errorCode };
await SafeAsync(() => session.NotifyAsync(MonitorMethods.TestStatus, parameters, MonitorJsonContext.Default.TestStatusParams, CancellationToken.None));
}
/// <summary>代 Probe 向 Backend 取 Agent 列表(Probe 不认识 Backend)。链路断开时返回 fleet_node_offline。</summary>
public async ValueTask<Result<ProbeAgentsResult>> RequestProbeAgentsAsync(ProbeAgentsParams request, CancellationToken ct)
{
RpcSession<LinkContext>? session = _link?.CurrentSession;
if (session is null)
{
return Result<ProbeAgentsResult>.Fail(FleetErrors.NodeOffline, "the agent has no link to the backend");
}
return await session.RequestAsync(MonitorMethods.ProbeAgents, request, MonitorJsonContext.Default.ProbeAgentsParams, MonitorJsonContext.Default.ProbeAgentsResult, ct);
await SafeAsync(() => session.NotifyAsync(MoShiroMethods.TestStatus, parameters, MoShiroJsonContext.Default.TestStatusParams, CancellationToken.None));
}
private static async ValueTask SafeAsync(Func<ValueTask> send)
@@ -102,9 +90,12 @@ public sealed class PeerHub
private readonly TimeProvider _time;
private readonly INodeLog _log;
/// <summary>对等面入站方法的路由器(只读,测试用)。</summary>
public RpcRouter<PeerContext> Router => _router;
public PeerHub(SessionRegistry sessions, MonoClock clock, ChunkAssembler chunks, TaskManager tasks, BackendUplink uplink, TimeProvider time, INodeLog log)
{
_router = BuildRouter(uplink);
_router = BuildRouter();
_sessions = sessions;
_clock = clock;
_chunks = chunks;
@@ -127,7 +118,7 @@ public sealed class PeerHub
var order = new Queue<Guid>();
rpc.OnBinary = async (frame, token) =>
{
if (!OutboxFrame.TryParse(frame, out byte channel, out Guid id, out ReadOnlyMemory<byte> payload) || channel != MonitorChannels.ColumnBlock)
if (!OutboxFrame.TryParse(frame, out byte channel, out Guid id, out ReadOnlyMemory<byte> payload) || channel != MoShiroChannels.ColumnBlock)
{
return;
}
@@ -188,7 +179,7 @@ public sealed class PeerHub
await _tasks.CloseSessionAsync(probe);
_log.Info("probe disconnected", ("probe", info.PeerId));
OnProbeLost?.Invoke(probe);
await _uplink.EmitEventAsync(AgentEventKinds.ProbeDisconnected, info.PeerId);
await _uplink.EmitEventAsync(AgentEventKinds.ProbeDisconnected, info.PeerId, Describe(probe));
}
}
@@ -220,7 +211,7 @@ public sealed class PeerHub
/// <summary>当前连着的 Probe(心跳和 Backend 重连之后的补报都用它)。</summary>
public IReadOnlyList<ConnectedProbe> ConnectedProbes() => [.. _sessions.All.Select(Describe)];
private static ConnectedProbe Describe(ProbeSession p) => new() { Id = new ProbeId(p.ProbeId), Version = p.Version, Remote = p.RemoteAddress, SinceUtcMs = p.ConnectedAtUtcMs };
private static ConnectedProbe Describe(ProbeSession p) => new() { Id = new ProbeId(p.ProbeId), Version = p.Version, Remote = p.RemoteAddress, SinceUtcMs = p.ConnectedAtUtcMs, IpVersion = p.IpVersion };
public IReadOnlyList<string> ConnectedProbeIds() => [.. _sessions.All.Select(p => p.ProbeId)];
@@ -230,13 +221,17 @@ public sealed class PeerHub
/// <summary>Backend 要求断开某个 Probe(它会按退避重连,除非已经被吊销)。没有这个会话时返回 false。</summary>
public async ValueTask<bool> DisconnectAsync(string probeId)
{
if (_sessions.Get(probeId)?.Rpc is not { } rpc)
bool disconnected = false;
foreach (ProbeSession probe in _sessions.All)
{
return false;
if (probe.ProbeId == probeId && probe.Rpc is { } rpc)
{
await rpc.CloseAsync("disconnected", CancellationToken.None);
disconnected = true;
}
}
await rpc.CloseAsync("disconnected", CancellationToken.None);
return true;
return disconnected;
}
/// <summary>名单(期望状态里的 peers)里已经没有的 Probe:立刻关闭它的会话(吊销生效)。</summary>
@@ -252,27 +247,6 @@ public sealed class PeerHub
}
}
/// <summary>Backend 的 Agent 列表变化时转交给所有 Probe(Probe 据此调整它连接的 Agent)。</summary>
public async ValueTask RelayAgentsAsync(ProbeAgentsResult list)
{
foreach (ProbeSession probe in _sessions.All)
{
if (probe.Rpc is not { } rpc)
{
continue;
}
try
{
await rpc.NotifyAsync(MonitorMethods.ProbeAgentsChanged, list, MonitorJsonContext.Default.ProbeAgentsResult, CancellationToken.None);
}
catch (Exception ex) when (ex is ObjectDisposedException or InvalidOperationException or IOException)
{
// 会话刚好结束
}
}
}
private async ValueTask PushAsync(ProbeSession probe, LineTaskSet set, CancellationToken ct)
{
if (probe.Rpc is null)
@@ -284,7 +258,7 @@ public sealed class PeerHub
{
Type = LineTaskSet.StateType,
Version = set.Version,
Doc = JsonSerializer.SerializeToElement(set, MonitorJsonContext.Default.LineTaskSet),
Doc = JsonSerializer.SerializeToElement(set, MoShiroJsonContext.Default.LineTaskSet),
};
try
{
@@ -300,16 +274,11 @@ public sealed class PeerHub
}
}
private static RpcRouter<PeerContext> BuildRouter(BackendUplink uplink) =>
new RpcRouter<PeerContext>()
private static RpcRouter<PeerContext> BuildRouter() =>
new RpcRouter<PeerContext>(RpcSide.Listener, MoShiroMethodTable.Table)
.On(
MonitorMethods.ProbeAgents,
MonitorJsonContext.Default.ProbeAgentsParams,
MonitorJsonContext.Default.ProbeAgentsResult,
(_, request, ct) => uplink.RequestProbeAgentsAsync(request, ct))
.On(
MonitorMethods.Clock,
MonitorJsonContext.Default.ClockParams,
MonitorJsonContext.Default.ClockResult,
MoShiroMethods.Clock,
MoShiroJsonContext.Default.ClockParams,
MoShiroJsonContext.Default.ClockResult,
(context, _, _) => ValueTask.FromResult(Result<ClockResult>.Ok(new ClockResult { AgentMonoUs = context.Clock.NowUs() })));
}
@@ -1,6 +1,6 @@
using Fleet.Core;
using Fleet.Hosting;
using Monitor.Agent;
using MoShiro.Agent;
TimeProvider time = TimeProvider.System;
INodeLog log = new ConsoleNodeLog(time);
@@ -11,7 +11,7 @@ if (PlatformCheck.VerifyCrypto() is { } problem)
return 1;
}
string configPath = "/etc/monitor/agent.json";
string configPath = "/etc/moshiro/agent.json";
for (int i = 0; i < args.Length - 1; i++)
{
if (args[i] == "--config")
@@ -22,9 +22,11 @@ for (int i = 0; i < args.Length - 1; i++)
AgentConfig config;
LoadedNodeIdentity identity;
RestartController restart;
try
{
config = NodeConfig.Load(configPath, AgentConfigContext.Default.AgentConfig);
restart = new RestartController(RestartSchedule.Resolve(config.RestartAt));
identity = LoadedNodeIdentity.Load(config.Fleet.IdentityDir);
}
catch (Exception ex) when (ex is IOException or InvalidDataException or InvalidOperationException or System.Text.Json.JsonException or System.Security.Cryptography.CryptographicException)
@@ -33,10 +35,11 @@ catch (Exception ex) when (ex is IOException or InvalidDataException or InvalidO
return 1;
}
using var restartScope = restart;
using (identity)
{
using var app = new AgentApp(config, identity, time, log);
await ServiceShell.RunAsync(app.RunAsync, log);
using var app = new AgentApp(config, identity, Path.GetDirectoryName(Path.GetFullPath(configPath))!, time, log, restart);
await ServiceShell.RunAsync(app.RunAsync, log, restart, time);
}
return 0;
@@ -1,10 +1,10 @@
using System.Net;
using System.Text.Json;
using Fleet.Core;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>
/// 目标任务(Agent 自己探测一个地址,没有对端):用和线路任务同一套积木(<see cref="BlockCatalog"/> 里的拨号方,例如 icmp_echo、connect),
@@ -80,7 +80,7 @@ public sealed class TargetRunners
Log = _log,
Session = () => null,
PeerAddress = address,
Fragments = new LocalSink(this, spec),
Fragments = new LocalSink(this, spec, address.AddressFamily == System.Net.Sockets.AddressFamily.InterNetworkV6 ? IpVersions.V6 : IpVersions.V4),
};
IBlock block = BlockCatalog.Find(spec.Kind)!.Dialer(env, new BlockSpec(decision.Kind!, decision.Resolved!, 0, port, new FragmentScope(spec.Id, null, null)));
using var cts = CancellationTokenSource.CreateLinkedTokenSource(ct);
@@ -132,11 +132,13 @@ public sealed class TargetRunners
{
private readonly TargetRunners _owner;
private readonly TaskSpec _spec;
private readonly string _ipVersion;
public LocalSink(TargetRunners owner, TaskSpec spec)
public LocalSink(TargetRunners owner, TaskSpec spec, string ipVersion)
{
_owner = owner;
_spec = spec;
_ipVersion = ipVersion;
}
public void Ship(string kind, FragmentScope scope, IReadOnlyList<RawStream> streams, IReadOnlyDictionary<string, JsonElement>? parameters, long startUtcMs, int durationMs, int port)
@@ -149,6 +151,7 @@ public sealed class TargetRunners
TaskId = scope.Task,
AgentId = _owner._agent,
Target = _spec.Target,
IpVersion = _ipVersion,
Params = parameters is null ? [] : new Dictionary<string, JsonElement>(parameters),
Port = port,
StartUtcMs = startUtcMs,
@@ -1,10 +1,10 @@
using System.Text.Json;
using Fleet.Core;
using Fleet.Rpc;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>
/// 期望任务集(<c>task_set</c>)的调和器:每个任务先经过 <see cref="BudgetGuard"/>(参数校验 + 本地硬上限)。
@@ -1,15 +1,15 @@
using System.Collections.Concurrent;
using Fleet.Core;
using Fleet.Rpc;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Agent;
namespace MoShiro.Agent;
/// <summary>
/// 按需测试的编排(<c>monitor.test.run</c>):Agent 是时间的主人。每个阶段:
/// 登记数据面的流 → <c>monitor.test.phase</c>(Probe 准备好自己一侧的角色)→ 运行(单向数据流由 Agent 发包,回显类等待 Probe 发包)
/// → <c>monitor.test.phase_end</c>(Probe 收尾,回复它交出的列块个数)→ 等所有列块合并上传 → <c>phase_done</c>。
/// 按需测试的编排(<c>moshiro.test.run</c>):Agent 是时间的主人。每个阶段:
/// 登记数据面的流 → <c>moshiro.test.phase</c>(Probe 准备好自己一侧的角色)→ 运行(单向数据流由 Agent 发包,回显类等待 Probe 发包)
/// → <c>moshiro.test.phase_end</c>(Probe 收尾,回复它交出的列块个数)→ 等所有列块合并上传 → <c>phase_done</c>。
/// 同一时间只能运行 <c>max_concurrent_tests</c> 个测试;Probe 掉线或被取消时测试中止,已经上传的阶段数据保留。
/// </summary>
public sealed class TestRunner
@@ -43,17 +43,22 @@ public sealed class TestRunner
public int Running => _running.Count;
/// <summary>校验并接受一个测试(立即返回),实际运行在后台进行,状态通过 <c>monitor.test.status</c> 回报。</summary>
/// <summary>校验并接受一个测试(立即返回),实际运行在后台进行,状态通过 <c>moshiro.test.status</c> 回报。</summary>
public Result<TestRunResult> Start(TestRunParams run)
{
if (_sessions.Get(run.ProbeId.Value) is not { } probe)
if (!IpVersions.IsValid(run.IpVersion))
{
return Result<TestRunResult>.Fail(MonitorErrors.ProbeOffline, $"probe {run.ProbeId.Value} is not connected to this agent");
return Result<TestRunResult>.Fail(MoShiroErrors.ValidationFailed, "ip_version must be ipv4 or ipv6");
}
if (_sessions.Get(run.ProbeId.Value, run.IpVersion) is not { } probe)
{
return Result<TestRunResult>.Fail(MoShiroErrors.ProbeOffline, $"probe {run.ProbeId.Value} is not connected to this agent");
}
if (run.Phases.Count == 0)
{
return Result<TestRunResult>.Fail(MonitorErrors.ValidationFailed, "a test needs at least one phase");
return Result<TestRunResult>.Fail(MoShiroErrors.ValidationFailed, "a test needs at least one phase");
}
var resolved = new List<(TestPhase Phase, TaskDecision Decision)>();
@@ -64,7 +69,7 @@ public sealed class TestRunner
if (!decision.Accepted)
{
return Result<TestRunResult>.Fail(
decision.Reason == RejectReasons.Unsupported ? FleetErrors.Unsupported : MonitorErrors.BudgetExceeded,
decision.Reason == RejectReasons.Unsupported ? FleetErrors.Unsupported : MoShiroErrors.BudgetExceeded,
$"phase {phase.Kind} refused: {decision.Reason}");
}
@@ -84,7 +89,7 @@ public sealed class TestRunner
if (_running.Count >= _maxConcurrent || !_running.TryAdd(run.TestId.Value, running))
{
running.Dispose();
return Result<TestRunResult>.Fail(MonitorErrors.TestBusy, "this agent is already running a test");
return Result<TestRunResult>.Fail(MoShiroErrors.TestBusy, "this agent is already running a test");
}
_ = Task.Run(() => RunAsync(run.TestId, resolved, running), CancellationToken.None);
@@ -154,7 +159,7 @@ public sealed class TestRunner
catch (OperationCanceledException)
{
finalStatus = running.Lost ? TestStatuses.Failed : TestStatuses.Aborted;
errorCode = running.Lost ? MonitorErrors.ProbeOffline.Value : null;
errorCode = running.Lost ? MoShiroErrors.ProbeOffline.Value : null;
await TryAbortProbeAsync(running.Probe, test);
}
catch (TestFailure failure)
@@ -183,7 +188,7 @@ public sealed class TestRunner
_streams.Add(stream);
try
{
RpcSession<PeerContext> rpc = probe.Rpc ?? throw new TestFailure(MonitorErrors.ProbeOffline.Value);
RpcSession<PeerContext> rpc = probe.Rpc ?? throw new TestFailure(MoShiroErrors.ProbeOffline.Value);
var start = new TestPhaseStartParams
{
TestId = test,
@@ -194,7 +199,7 @@ public sealed class TestRunner
DurationSeconds = durationSeconds,
Port = stream.Port,
};
Result<TestPhaseStartResult> accepted = await rpc.RequestAsync(MonitorMethods.TestPhase, start, MonitorJsonContext.Default.TestPhaseStartParams, MonitorJsonContext.Default.TestPhaseStartResult, ct);
Result<TestPhaseStartResult> accepted = await rpc.RequestAsync(MoShiroMethods.TestPhase, start, MoShiroJsonContext.Default.TestPhaseStartParams, MoShiroJsonContext.Default.TestPhaseStartResult, ct);
if (!accepted.TryGet(out TestPhaseStartResult? started, out FleetError? error) || !started.Accepted)
{
throw new TestFailure(error?.Code.Value ?? FleetErrors.Unsupported.Value);
@@ -209,7 +214,7 @@ public sealed class TestRunner
}
catch (BlockFaultException)
{
throw new TestFailure(MonitorErrors.ProbeOffline.Value);
throw new TestFailure(MoShiroErrors.ProbeOffline.Value);
}
}
else
@@ -219,7 +224,7 @@ public sealed class TestRunner
long sent = stream.Running.Block.Produced;
var end = new TestPhaseEndParams { TestId = test, Phase = index, StreamId = stream.Id, SentCount = sent };
Result<TestPhaseEndResult> ended = await rpc.RequestAsync(MonitorMethods.TestPhaseEnd, end, MonitorJsonContext.Default.TestPhaseEndParams, MonitorJsonContext.Default.TestPhaseEndResult, ct, s_phaseSlack);
Result<TestPhaseEndResult> ended = await rpc.RequestAsync(MoShiroMethods.TestPhaseEnd, end, MoShiroJsonContext.Default.TestPhaseEndParams, MoShiroJsonContext.Default.TestPhaseEndResult, ct, s_phaseSlack);
if (!ended.TryGet(out TestPhaseEndResult? finished, out FleetError? endError))
{
throw new TestFailure(endError.Code.Value);
@@ -254,7 +259,7 @@ public sealed class TestRunner
try
{
await rpc.NotifyAsync(MonitorMethods.TestAbort, new TestAbortParams { TestId = test }, MonitorJsonContext.Default.TestAbortParams, CancellationToken.None);
await rpc.NotifyAsync(MoShiroMethods.TestAbort, new TestAbortParams { TestId = test }, MoShiroJsonContext.Default.TestAbortParams, CancellationToken.None);
}
catch (Exception ex) when (ex is ObjectDisposedException or InvalidOperationException or IOException)
{
@@ -1,20 +0,0 @@
<Project>
<Import Project="Sdk.props" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.props" Sdk="StuDev.AotAnywhere"
Condition="'$(RuntimeIdentifier)' == 'linux-x64' Or '$(RuntimeIdentifier)' == 'linux-arm64' Or '$(RuntimeIdentifier)' == 'linux-arm' Or '$(RuntimeIdentifier)' == 'linux-musl-x64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm' Or '$(RuntimeIdentifier)' == 'osx-x64' Or '$(RuntimeIdentifier)' == 'osx-arm64' Or '$(RuntimeIdentifier)' == 'win-x64' Or '$(RuntimeIdentifier)' == 'win-arm64'" />
<PropertyGroup>
<OutputType>Exe</OutputType>
<AssemblyName>monitor-agent</AssemblyName>
<RootNamespace>Monitor.Agent</RootNamespace>
<IsAotCompatible>true</IsAotCompatible>
</PropertyGroup>
<ItemGroup>
<InternalsVisibleTo Include="Monitor.Backend.Tests" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Link\Fleet.Link.csproj" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Hosting\Fleet.Hosting.csproj" />
<ProjectReference Include="..\..\..\shared\Monitor.Measurement\Monitor.Measurement.csproj" />
</ItemGroup>
<Import Project="Sdk.targets" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.targets" Sdk="StuDev.AotAnywhere"
Condition="'$(RuntimeIdentifier)' == 'linux-x64' Or '$(RuntimeIdentifier)' == 'linux-arm64' Or '$(RuntimeIdentifier)' == 'linux-arm' Or '$(RuntimeIdentifier)' == 'linux-musl-x64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm' Or '$(RuntimeIdentifier)' == 'osx-x64' Or '$(RuntimeIdentifier)' == 'osx-arm64' Or '$(RuntimeIdentifier)' == 'win-x64' Or '$(RuntimeIdentifier)' == 'win-arm64'" />
</Project>
+104
View File
@@ -0,0 +1,104 @@
# Fleet:通用节点框架
「很多节点连到一个面板」这类产品都要解决同一批问题:节点怎么接入、怎么证明身份、怎么保持连接、怎么可靠上报、怎么接收配置。Fleet 把这些做成与业务无关的库,MoShiro 是它的第一个使用者;产品只写业务。Fleet 不引用任何 `MoShiro.*`,也不出现产品的名字、端口和协议前缀,由 `ArchitectureTests` 在构建时检查。
```text
┌──────────────────── 产品层(MoShiro:Protocol · Measurement · Agent · Probe · Backend)────────────────────┐
└────────────────────────────────────────────┬─────────────────────────────────────────────────────────────┘
│ 只依赖 Fleet
┌───────────────────────────── Fleet ────────┴─────────────────────────────────────────────────────────────┐
│ 节点侧(只依赖 BCL,NativeAOT) 服务端侧(ASP.NET Core) │
│ Fleet.Core Result、退避、有界队列、ID、令牌哈希 Fleet.Server.Abstractions IFleetStore 等接口 │
│ Fleet.Identity 密钥、签名、证书、KeyStore Fleet.Server.Enrollment 加入令牌、注册、审批 │
│ Fleet.Rpc JSON-RPC 信封、路由、二进制帧 Fleet.Server.Link wss 网关、在线表、状态下发 │
│ Fleet.Link 会话、认证、重连、发件箱、状态同步、点对点 │
│ Fleet.Hosting 模块宿主、配置、日志、服务外壳 工具:fleet-enroll(一次性注册) │
│ Fleet.Enrollment.Client 注册流程(给 fleet-enroll 用) │
└────────────────────────────────────────────────────────────────────────────────────────────────────────┘
```
## 设计原则
1. **与业务无关**:产品通过扩展点接入:元数据类型、RPC 方法、发件箱消息类型、期望状态文档类型。
2. **安全默认值**:Fleet 永远不提供 exec、shell、文件读写、远程下发代码这类方法;节点只响应本地注册过的方法(本地策略 `allow_methods` 还能收窄);调用对方不支持的方法在本地直接返回 `fleet_unsupported`。
3. **权限分离**:注册(`fleet-enroll`,一次性)和常驻业务进程是两个进程;常驻进程不含注册代码,也不使用 `HttpClient`。
4. **NativeAOT**:节点侧只依赖 BCL,泛型扩展点通过产品传入的 `JsonTypeInfo<T>` 序列化,不使用反射。
5. **更新不在 Fleet 里**:节点进程没有自更新能力。容器节点的签名更新由宿主机上的更新器完成,见 [`docs/deployment.md`](../../docs/deployment.md);纯二进制节点的更新是待办,见 [`TODO.md`](../../TODO.md)。
## 身份
| 对象 | 算法 | 保存位置 |
| --- | --- | --- |
| 节点身份密钥 | ECDSA P-256(BCL 自带) | `<identity_dir>/node.key`(0600);服务端只保存公钥 |
| 点对点 TLS 证书 | 自签名 ECDSA P-256 | `<identity_dir>/tls.pfx`;对端按 SPKI SHA-256 指纹校验 |
| 签名串 | `"fleet|v1|<purpose>|" + 各字段` | 带用途前缀,防止同一个签名在不同场景被复用 |
用途:`auth`(连接服务端)、`peer`(点对点)、`enroll`(注册时的持有证明)。产品可以追加自己的用途,不能占用 `fleet` 保留的。
## 注册
```text
管理员 ── 创建加入令牌 {labels, max_uses, expires_at, auto_approve} ──> 服务端
<── 令牌明文(只返回一次)+ 安装命令
节点:fleet-enroll --server https://panel --join jt_xxx --name jp-1 --identity-dir /etc/myapp
├─ 生成身份密钥和 TLS 证书
├─ POST 注册端点 { join_token, name, pubkey, tls_spki, proof = 签名("enroll", 令牌哈希 + pubkey), node_version, meta }
│ 服务端:校验令牌 → uses+1 → 创建节点(active 或 pending)→ 触发产品的 IEnrollmentHook
└─ 写入 <identity_dir>/node.json { node_id, server, … },然后退出
```
- 同名节点已存在时,新注册进入 `pending`,管理员选择「接管」(旧记录换成新公钥,历史数据保留)或「作为新节点」。
- 节点状态:`pending`(能连接,但不下发状态、不接收上报)、`active`、`disabled`(拒绝连接)。
## RPC
- JSON-RPC 2.0;方法名 `领域.动作`,`fleet.*` 保留给框架,产品用自己的前缀(MoShiro 是 `moshiro.*`)。
- 二进制帧 `u8 channel | 正文`:`0`–`15` 框架保留,产品从 `16` 开始。
- 错误:`error.data.code` 为 snake_case,框架的以 `fleet_` 开头,产品可以扩展。
- **方法必须声明**:每个方法在声明表(`MethodTable`)里写明发起方、请求还是通知、哪些角色可以参与;不存在双向方法,同一个方法名在一个平面里只有一个发起方。`RpcRouter` 在注册、分发和发送三处核对,违规回 `fleet_protocol_violation`,超过上限断开连接。详见 [`docs/security.md`](../../docs/security.md) 第 8 节。
- 与传输无关:同一套路由跑在 wss 上,也跑在 TLS 帧(`u32 长度 | u8 类型 | 正文`)上。
| 保留方法 | 方向 | 说明 |
| --- | --- | --- |
| `fleet.challenge` / `fleet.auth` | 服务端→节点 / 节点→服务端 | 签名认证(`auth` 用途,nonce 30 秒内有效) |
| `fleet.hello` | 节点→服务端 | 版本、能力、本地策略摘要、当前时间 → 返回时钟偏差;产品的元数据也放在这里 |
| `fleet.heartbeat` | 节点→服务端 | 每 10 秒一次,产品可以附加计数 |
| `fleet.state.sync` | 服务端→节点 | `{version, doc}` → `{applied_version, rejected[]}` |
| `fleet.outbox.ack` | 服务端→节点 | `{ids[], duplicates[]}` |
| `fleet.revoked` | 服务端→节点 | 节点被禁用或接管,随后断开 |
## 连接
```text
connect(wss) → fleet.challenge → fleet.auth → fleet.hello → 状态同步 → 正常运行
▲ │
└──── 指数退避重连(1 s → 60 s,带抖动;401 或 revoked 时停止重试)◀─ 断开
```
本地策略:`allow_methods`、`max_frame_bytes`、`max_inflight`。
**可靠发件箱**:产品 `EnqueueAsync` → 有界队列(按字节和条数;满时 `DropOldest` 或 `Reject`)→ 发送 → 服务端收件箱(`IInbox`)持久化后 `fleet.outbox.ack` → 移除。至少一次,按 `msg_id`(UUIDv7)去重;在途数据默认限 4 MiB;丢弃会累计缺口,通过 `OnGap` 通知产品。只在内存里,不落盘。
**期望状态同步**:服务端 `IStateProvider<TDoc>` 给每个节点算 `{version, doc}`,变化时 `NotifyChanged` 推送;节点 `StateHandler<TDoc>` 调和后返回 `{applied_version, rejected[]}`。下发的是**完整快照**,天然幂等,重连后服务端总会重发当前版本。`rejected[]` 带结构化原因(`unsupported`、`budget_exceeded`、`policy_denied`),面板可以直接展示。
## 点对点会话
```text
节点 A(发起方,如 Probe)── TLS(按 SPKI 校验证书)──> 节点 B(监听方,如 Agent)
<── fleet.peer.challenge ──
── fleet.peer.auth(peer 用途,A 的私钥签名)──>
B 用服务端下发的对端公钥列表验签(经 state.sync 下发),之后复用同一套 Rpc:路由、二进制通道、发件箱
```
握手用两个方法:`fleet.peer.challenge`(监听方→拨号方,通知)和 `fleet.peer.auth`(拨号方→监听方,请求);握手完成后,两侧各自只能发起声明表里属于自己这一侧的方法。
对端公钥和允许连接的节点由产品通过期望状态下发给监听方,Fleet 只管认证和会话,拓扑由产品决定;UDP、原始 TCP 数据面由产品自己实现,Fleet 只提供会话密钥派生和首帧 MAC 的辅助函数。
## 宿主与服务端抽象
- `ModuleHost`:按依赖顺序启动模块、相反顺序停止,隔离故障,重启时退避。
- `NodeConfig`:源生成 JSON 读取,环境变量覆盖,启动时校验。
- `ConfigStore`:常驻进程里**唯一**允许写文件的地方,只能写配置目录,写入原子替换、权限 0600;父目录不存在会创建。
- `ServiceShell`:SIGTERM、Ctrl+C,systemd `sd_notify`、OpenWrt procd(前台运行)和 Windows 服务。
- 服务端接口:`IFleetStore`(节点、令牌、公钥、状态版本)、`IStateProvider<TDoc>`、`IInbox`、`INodeEventSink`、`IEnrollmentHook`。MoShiro 把 `IFleetStore` 放进自己的 `moshiro.db`,和业务数据由同一个单写入者写入。
@@ -13,6 +13,8 @@ public static class FleetErrors
public static readonly ErrorCode Unauthorized = new("fleet_unauthorized");
public static readonly ErrorCode Forbidden = new("fleet_forbidden");
public static readonly ErrorCode Unsupported = new("fleet_unsupported");
/// <summary>方法没有声明、方向不对、类型不对或角色不允许:对方违反了协议。</summary>
public static readonly ErrorCode ProtocolViolation = new("fleet_protocol_violation");
public static readonly ErrorCode InvalidParams = new("fleet_invalid_params");
public static readonly ErrorCode NotFound = new("fleet_not_found");
public static readonly ErrorCode Internal = new("fleet_internal");
+1 -1
View File
@@ -3,7 +3,7 @@ using System.Text.Json.Serialization;
namespace Fleet.Core;
/// <summary>节点 ID(带类型前缀的 ULID)。Monitor 中的 AgentId 就是 NodeId。</summary>
/// <summary>节点 ID(带类型前缀的 ULID)。MoShiro 中的 AgentId 就是 NodeId。</summary>
[JsonConverter(typeof(NodeIdJsonConverter))]
public readonly record struct NodeId(string Value)
{
+37 -2
View File
@@ -126,7 +126,8 @@ public static class PlatformCheck
/// <summary>处理 Ctrl+C 和 SIGTERM:取消令牌,让各个模块有序退出。</summary>
public static class ServiceShell
{
public static async Task RunAsync(Func<CancellationToken, Task> run, INodeLog log)
/// <param name="restart">每天定时有序退出一次(见 <see cref="RestartController"/>),重新拉起由服务管理器负责。</param>
public static async Task RunAsync(Func<CancellationToken, Task> run, INodeLog log, RestartController? restart = null, TimeProvider? time = null)
{
using var cts = new CancellationTokenSource();
using PosixSignalRegistration term = PosixSignalRegistration.Create(PosixSignal.SIGTERM, context =>
@@ -156,6 +157,7 @@ public static class ServiceShell
cts.Cancel();
});
Task trimmer = TrimHeapAsync(cts.Token);
Task scheduled = restart is null ? Task.CompletedTask : WaitForRestartAsync(restart, time ?? TimeProvider.System, log, cts);
try
{
await run(cts.Token);
@@ -168,6 +170,38 @@ public static class ServiceShell
{
await cts.CancelAsync();
await trimmer;
await scheduled;
}
}
private static async Task WaitForRestartAsync(RestartController restart, TimeProvider time, INodeLog log, CancellationTokenSource cts)
{
while (!cts.IsCancellationRequested)
{
// 先取令牌、再取设置:中间被改动的话令牌会被取消,下一轮重新排
CancellationToken changed = restart.ChangeToken;
RestartSchedule? schedule = restart.Current;
using CancellationTokenSource wait = CancellationTokenSource.CreateLinkedTokenSource(cts.Token, changed);
try
{
if (schedule is null)
{
await Task.Delay(Timeout.Infinite, wait.Token);
continue;
}
DateTimeOffset next = schedule.NextAfter(time.GetUtcNow() + RestartSchedule.MinimumUptime);
log.Info("restart scheduled", ("at", next.ToString("O", System.Globalization.CultureInfo.InvariantCulture)), ("source", restart.IsLocal ? "local" : "panel"));
await Task.Delay(next - time.GetUtcNow(), time, wait.Token);
}
catch (OperationCanceledException)
{
continue; // 进程在退出,或者设置被改了:循环条件和下一轮会处理
}
log.Info("stopping", ("result", "scheduled_restart"));
await cts.CancelAsync();
return;
}
}
@@ -243,7 +277,8 @@ public sealed class ConfigStore
throw new InvalidOperationException("config files must stay inside the config directory");
}
Directory.CreateDirectory(_directory);
// 相对路径里的子目录(例如 update/)也要建:新建的命名卷里只有镜像自带的目录
Directory.CreateDirectory(Path.GetDirectoryName(path)!);
string temp = path + ".tmp";
#pragma warning disable RS0030 // 纯采集边界的唯一豁免点:ConfigStore 只写配置目录,先写临时文件再原子替换
File.WriteAllBytes(temp, content);
@@ -0,0 +1,73 @@
namespace Fleet.Hosting;
/// <summary>
/// 当前生效的定时重启:节点本地的设置优先;本地没有写时用面板下发的,面板改了之后立即重新排下一次,不需要重启进程。
/// </summary>
public sealed class RestartController : IDisposable
{
private readonly RestartSchedule? _local;
private readonly object _gate = new();
private RestartSchedule? _remote;
private CancellationTokenSource _changed = new();
public RestartController(RestartSchedule? local = null)
{
_local = local;
}
/// <summary>本地有设置时为 true:面板下发的不生效。</summary>
public bool IsLocal => _local is not null;
public RestartSchedule? Current
{
get
{
lock (_gate)
{
return _local ?? _remote;
}
}
}
/// <summary>当前设置被改动时会被取消的令牌;先取令牌、再取 <see cref="Current"/>,就不会漏掉中间的改动。</summary>
public CancellationToken ChangeToken
{
get
{
lock (_gate)
{
return _changed.Token;
}
}
}
/// <summary>面板下发的设置;为空表示面板关闭了定时重启。返回 false 表示文本不合法(节点忽略它)。</summary>
public bool SetRemote(string? text)
{
RestartSchedule? schedule = null;
if (!string.IsNullOrWhiteSpace(text) && !RestartSchedule.TryParse(text, out schedule))
{
return false;
}
CancellationTokenSource previous;
lock (_gate)
{
_remote = schedule;
previous = _changed;
_changed = new CancellationTokenSource();
}
previous.Cancel();
previous.Dispose();
return true;
}
public void Dispose()
{
lock (_gate)
{
_changed.Dispose();
}
}
}
@@ -0,0 +1,105 @@
using System.Globalization;
namespace Fleet.Hosting;
/// <summary>
/// 每天固定时刻有序退出一次,由 systemd、容器引擎或 procd 的「总是重启」拉起。写法 <c>HH:mm</c>(按 UTC)、
/// <c>HH:mm±hh:mm</c> 或 <c>HH:mm</c> 加 <c>Z</c>:容器里没有时区数据库,所以偏移要写明,例如 <c>04:00+08:00</c>。
/// 进程自己退出,不需要任何执行外部命令的能力。
/// </summary>
public sealed class RestartSchedule
{
/// <summary>环境变量,设置了就覆盖配置文件里的值。</summary>
public const string EnvironmentVariable = "MOSHIRO_RESTART_AT";
/// <summary>刚启动的一段时间内不会触发:重启之后不会因为时刻还没过去而再重启一次,也避免重启风暴。</summary>
public static readonly TimeSpan MinimumUptime = TimeSpan.FromMinutes(10);
private readonly TimeSpan _time;
private readonly TimeSpan _offset;
private RestartSchedule(TimeSpan time, TimeSpan offset)
{
_time = time;
_offset = offset;
}
public static bool TryParse(string? text, out RestartSchedule? schedule)
{
schedule = null;
string value = text?.Trim() ?? "";
// HH:mm,后面可以跟 Z 或 ±hh:mm
if (value.Length < 5 || value[2] != ':' ||
!TryTwoDigits(value, 0, out int hour) || !TryTwoDigits(value, 3, out int minute) || hour > 23 || minute > 59)
{
return false;
}
TimeSpan offset = TimeSpan.Zero;
string rest = value[5..];
if (rest.Length > 0 && rest != "Z")
{
if (rest.Length != 6 || rest[0] is not ('+' or '-') || rest[3] != ':' ||
!TryTwoDigits(rest, 1, out int offsetHours) || !TryTwoDigits(rest, 4, out int offsetMinutes) || offsetMinutes > 59)
{
return false;
}
offset = new TimeSpan(offsetHours, offsetMinutes, 0);
if (offset > TimeSpan.FromHours(14))
{
return false;
}
if (rest[0] == '-')
{
offset = -offset;
}
}
schedule = new RestartSchedule(new TimeSpan(hour, minute, 0), offset);
return true;
}
private static bool TryTwoDigits(string text, int start, out int value)
{
value = 0;
if (!char.IsAsciiDigit(text[start]) || !char.IsAsciiDigit(text[start + 1]))
{
return false;
}
value = (text[start] - '0') * 10 + (text[start + 1] - '0');
return true;
}
/// <summary>
/// 环境变量优先,其次配置文件的值;都没有就不定时重启。写错了抛 <see cref="InvalidOperationException"/>,
/// 启动时就能看到,而不是悄悄不生效。
/// </summary>
public static RestartSchedule? Resolve(string? configured)
{
string? text = Environment.GetEnvironmentVariable(EnvironmentVariable);
if (string.IsNullOrWhiteSpace(text))
{
text = configured;
}
if (string.IsNullOrWhiteSpace(text))
{
return null;
}
return TryParse(text, out RestartSchedule? schedule)
? schedule
: throw new InvalidOperationException($"restart_at must look like 04:00, 04:00Z or 04:00+08:00 (got '{text}')");
}
/// <summary>严格晚于 <paramref name="now"/> 的下一个触发时刻。</summary>
public DateTimeOffset NextAfter(DateTimeOffset now)
{
DateTimeOffset local = now.ToOffset(_offset);
DateTimeOffset candidate = new DateTimeOffset(local.Year, local.Month, local.Day, 0, 0, 0, _offset) + _time;
return candidate <= local ? candidate.AddDays(1) : candidate;
}
}
+4 -1
View File
@@ -49,7 +49,7 @@ public sealed class FleetLinkOptions
public string NodeVersion { get; init; } = "0.0.0";
/// <summary>框架之外的能力,例如 <c>monitor.v1</c> 与 <c>monitor.kinds:udp_echo,tcp_echo</c>。</summary>
/// <summary>框架之外的能力,例如 <c>moshiro.v1</c> 与 <c>moshiro.kinds:udp_echo,tcp_echo</c>。</summary>
public IReadOnlyList<string> Capabilities { get; init; } = [];
/// <summary>附加到 <c>fleet.hello</c> 的产品元数据。</summary>
@@ -105,6 +105,9 @@ public sealed class FleetLink
public RpcSession<LinkContext>? CurrentSession => _session;
/// <summary>入站方法的路由器(只读,测试用它核对声明表和实际注册的处理函数)。</summary>
public RpcRouter<LinkContext> Router => _router;
/// <summary>被服务端撤销或禁用:不会再重连。</summary>
public bool Revoked => _stopped;
+7 -1
View File
@@ -57,6 +57,11 @@ public sealed class PeerListener
public async Task RunAsync(CancellationToken ct)
{
var listener = new TcpListener(_options.Listen);
if (_options.Listen.Address.Equals(IPAddress.IPv6Any))
{
listener.Server.DualMode = true;
}
listener.Start();
Port = ((IPEndPoint)listener.LocalEndpoint).Port;
Started.TrySetResult();
@@ -219,7 +224,8 @@ public static class PeerConnector
public static async Task<(PeerSessionInfo Info, TlsFrameTransport Transport)> ConnectAsync(
string host, int port, byte[] expectedSpkiSha256, NodeIdentity identity, string peerId, string? version, int maxFrameBytes, TimeProvider time, CancellationToken ct)
{
var client = new TcpClient { NoDelay = true };
TcpClient client = IPAddress.TryParse(host, out IPAddress? address) ? new TcpClient(address.AddressFamily) : new TcpClient();
client.NoDelay = true;
PeerSockets.EnableKeepAlive(client.Client);
try
{
+26 -1
View File
@@ -15,6 +15,31 @@ public static class FleetMethods
public const string Revoked = "fleet.revoked";
}
/// <summary>
/// 框架自己的方法声明,两个平面各一份。认证和握手阶段由连接代码直接处理的(<c>Inline</c>)不经过路由器。
/// 同一个方法名在两个平面里可以有不同的发起方:<c>fleet.state.sync</c> 在节点面是服务端发给节点,在对等面是 Agent 发给 Probe。
/// </summary>
public static class FleetMethodTable
{
public static readonly MethodTable Table = new(
[
// 节点面
MethodSpec.Notification(FleetMethods.Challenge, RpcSide.Server).AsInline(),
MethodSpec.Request(FleetMethods.Auth, RpcSide.Node).AsInline(),
MethodSpec.Request(FleetMethods.Hello, RpcSide.Node).AsInline(),
MethodSpec.Notification(FleetMethods.Heartbeat, RpcSide.Node).AsInline(),
MethodSpec.Request(FleetMethods.StateSync, RpcSide.Server),
MethodSpec.Notification(FleetMethods.OutboxAck, RpcSide.Server),
MethodSpec.Notification(FleetMethods.Revoked, RpcSide.Server),
// 对等面
MethodSpec.Notification(PeerMethods.Challenge, RpcSide.Listener).AsInline(),
MethodSpec.Request(PeerMethods.Auth, RpcSide.Dialer).AsInline(),
MethodSpec.Request(FleetMethods.StateSync, RpcSide.Listener),
MethodSpec.Notification(FleetMethods.OutboxAck, RpcSide.Listener),
]);
}
/// <summary>二进制通道:0–15 保留给框架,产品从 16 开始。</summary>
public static class FleetChannels
{
@@ -58,7 +83,7 @@ public sealed class HelloParams
[JsonPropertyName("now_utc_ms")]
public long NowUtcMs { get; set; }
/// <summary>产品自定义的元数据(例如 Monitor 的 tls_spki、probe_ports、local_limits 摘要)。</summary>
/// <summary>产品自定义的元数据(例如 MoShiro 的 tls_spki、probe_ports、local_limits 摘要)。</summary>
[JsonPropertyName("meta")]
public JsonElement? Meta { get; set; }
}
+120
View File
@@ -0,0 +1,120 @@
namespace Fleet.Rpc;
/// <summary>连接的哪一侧。节点面:服务端和节点;对等面:拨号方(Probe)和监听方(Agent)。</summary>
public enum RpcSide
{
Server,
Node,
Dialer,
Listener,
}
public enum RpcPlane
{
Node,
Peer,
}
public enum MethodKind
{
Request,
Notification,
}
public static class RpcSides
{
public static RpcPlane PlaneOf(RpcSide side) => side is RpcSide.Server or RpcSide.Node ? RpcPlane.Node : RpcPlane.Peer;
public static RpcSide Opposite(RpcSide side) => side switch
{
RpcSide.Server => RpcSide.Node,
RpcSide.Node => RpcSide.Server,
RpcSide.Dialer => RpcSide.Listener,
_ => RpcSide.Dialer,
};
}
/// <summary>
/// 一个方法的声明:<b>只有 <see cref="From"/> 能发起它</b>(不存在双向方法),类型是请求还是通知,哪些节点角色可以参与。
/// <see cref="Inline"/> 表示它在认证或握手阶段由连接代码直接处理,不经过路由器。
/// </summary>
public sealed record MethodSpec(string Name, RpcSide From, MethodKind Kind, IReadOnlyList<string>? Roles = null, bool Inline = false)
{
public RpcPlane Plane => RpcSides.PlaneOf(From);
/// <summary>接收方。</summary>
public RpcSide To => RpcSides.Opposite(From);
public bool AllowsRole(string? role) => Roles is null || (role is not null && Roles.Contains(role));
public static MethodSpec Request(string name, RpcSide from, params string[] roles) => new(name, from, MethodKind.Request, roles.Length == 0 ? null : roles);
public static MethodSpec Notification(string name, RpcSide from, params string[] roles) => new(name, from, MethodKind.Notification, roles.Length == 0 ? null : roles);
public MethodSpec AsInline() => this with { Inline = true };
}
/// <summary>二进制通道的声明:谁能发、哪些角色可以发。通道 0–15 保留给框架,产品从 16 开始。</summary>
public sealed record ChannelSpec(byte Channel, RpcSide From, IReadOnlyList<string>? Roles = null)
{
public RpcPlane Plane => RpcSides.PlaneOf(From);
public bool AllowsRole(string? role) => Roles is null || (role is not null && Roles.Contains(role));
}
/// <summary>
/// 协议方法和二进制通道的声明表,单一真源。路由器用它在注册时拒绝方向、类型、角色不符的处理函数,
/// 在收发时拒绝没有声明或不该由对方发起的方法。同一个方法名在不同平面里可以有不同的发起方,所以按(平面,名字)查找。
/// </summary>
public sealed class MethodTable
{
private readonly Dictionary<(RpcPlane, string), MethodSpec> _methods = [];
private readonly Dictionary<(RpcPlane, byte), ChannelSpec> _channels = [];
public MethodTable(IEnumerable<MethodSpec> methods, IEnumerable<ChannelSpec>? channels = null)
{
foreach (MethodSpec spec in methods)
{
if (!_methods.TryAdd((spec.Plane, spec.Name), spec))
{
throw new ArgumentException($"method {spec.Name} is declared twice in the {spec.Plane} plane");
}
}
foreach (ChannelSpec spec in channels ?? [])
{
if (!_channels.TryAdd((spec.Plane, spec.Channel), spec))
{
throw new ArgumentException($"channel {spec.Channel} is declared twice in the {spec.Plane} plane");
}
}
}
public IEnumerable<MethodSpec> Methods => _methods.Values;
public IEnumerable<ChannelSpec> Channels => _channels.Values;
public MethodSpec? Find(RpcPlane plane, string method) => _methods.GetValueOrDefault((plane, method));
public ChannelSpec? FindChannel(RpcPlane plane, byte channel) => _channels.GetValueOrDefault((plane, channel));
/// <summary>合并多张表(框架的加产品的)。</summary>
public static MethodTable Combine(params MethodTable[] tables) =>
new(tables.SelectMany(t => t.Methods), tables.SelectMany(t => t.Channels));
/// <summary>
/// 对账:<paramref name="receiver"/> 这一侧、<paramref name="role"/> 这个角色应该处理、但没有注册处理函数的方法
/// (认证阶段直接处理的 <see cref="MethodSpec.Inline"/> 方法除外)。<paramref name="role"/> 为空表示不限角色(服务端要处理所有角色发来的)。
/// </summary>
public IReadOnlyList<string> MissingHandlers(RpcSide receiver, string? role, IEnumerable<string> registered)
{
HashSet<string> have = [.. registered];
return
[
.. Methods
.Where(m => m.To == receiver && !m.Inline && (role is null || m.AllowsRole(role)) && !have.Contains(m.Name))
.Select(m => m.Name)
.Order(StringComparer.Ordinal),
];
}
}
+128 -8
View File
@@ -4,7 +4,17 @@ using Fleet.Core;
namespace Fleet.Rpc;
/// <summary>方法路由,与传输方式无关。<typeparamref name="TContext"/> 由传输层提供(例如调用方节点)。</summary>
/// <summary>对端违反了协议:方法没有声明、方向不对、请求和通知用反了、或者角色不允许。</summary>
public readonly record struct RpcViolation(string Method, string Reason);
/// <summary>一次分发的结果:要回的响应(通知为空),以及对端有没有违规。</summary>
public readonly record struct RpcDispatch(RpcMessage? Response, RpcViolation? Violation);
/// <summary>
/// 方法路由,与传输方式无关,也是协议方向的<b>唯一卡口</b>:节点和服务端、Probe 和 Agent 的入站分发都经过这里。
/// <typeparamref name="TContext"/> 由传输层提供(例如调用方节点)。路由器知道自己在连接的哪一侧(<see cref="Side"/>)和声明表,
/// 注册处理函数时核对方向、类型和角色,分发时拒绝没有声明或不该由对方发起的方法,发送时核对自己有没有资格发起。
/// </summary>
public sealed class RpcRouter<TContext>
{
private delegate ValueTask<Result<JsonElement?>> Route(TContext context, JsonElement? parameters, CancellationToken ct);
@@ -13,12 +23,28 @@ public sealed class RpcRouter<TContext>
private readonly Dictionary<string, Route> _routes = new(StringComparer.Ordinal);
private readonly Action<string, Exception>? _onUnhandled;
private readonly string? _localRole;
private readonly Func<TContext, string?>? _callerRole;
public RpcRouter(Action<string, Exception>? onUnhandled = null)
/// <param name="side">本路由器在连接的哪一侧。</param>
/// <param name="table">方法声明表。</param>
/// <param name="localRole">本节点的角色(节点侧和对等面用):声明限定了角色的方法,角色不符就不能注册。</param>
/// <param name="callerRole">从调用上下文取出对端的角色(服务端用):声明限定了角色的方法,角色不符就拒绝。</param>
public RpcRouter(RpcSide side, MethodTable table, string? localRole = null, Func<TContext, string?>? callerRole = null, Action<string, Exception>? onUnhandled = null)
{
Side = side;
Table = table;
_localRole = localRole;
_callerRole = callerRole;
_onUnhandled = onUnhandled;
}
public RpcSide Side { get; }
public MethodTable Table { get; }
private RpcPlane Plane => RpcSides.PlaneOf(Side);
public bool Has(string method) => _routes.ContainsKey(method);
public IReadOnlyCollection<string> Methods => _routes.Keys;
@@ -29,6 +55,7 @@ public sealed class RpcRouter<TContext>
JsonTypeInfo<TResult> resultInfo,
Func<TContext, TParams, CancellationToken, ValueTask<Result<TResult>>> handler)
{
Declare(method, MethodKind.Request);
_routes[method] = async (context, raw, ct) =>
{
if (!TryDeserialize(raw, paramsInfo, out TParams? p))
@@ -49,6 +76,7 @@ public sealed class RpcRouter<TContext>
JsonTypeInfo<TParams> paramsInfo,
Func<TContext, TParams, CancellationToken, ValueTask> handler)
{
Declare(method, MethodKind.Notification);
_routes[method] = async (context, raw, ct) =>
{
if (!TryDeserialize(raw, paramsInfo, out TParams? p))
@@ -66,12 +94,26 @@ public sealed class RpcRouter<TContext>
/// 处理一条请求或通知。请求返回响应;通知返回 null(失败时只丢弃)。
/// 处理函数里的异常属于程序缺陷:在这个边界捕获,回复 internal 错误,并交给 <c>onUnhandled</c> 记录。
/// </summary>
public async ValueTask<RpcMessage?> DispatchAsync(TContext context, RpcMessage message, CancellationToken ct)
public async ValueTask<RpcMessage?> DispatchAsync(TContext context, RpcMessage message, CancellationToken ct) =>
(await DispatchCheckedAsync(context, message, ct)).Response;
/// <summary>
/// 先核对声明(方向、类型、角色),再分发。违规的消息不会到达处理函数,返回的结果里带 <see cref="RpcViolation"/>,
/// 由会话决定怎么办(计数、超过上限就断开连接)。声明过、方向对、但这一侧没有处理函数的方法不算违规,只是不支持。
/// </summary>
public async ValueTask<RpcDispatch> DispatchCheckedAsync(TContext context, RpcMessage message, CancellationToken ct)
{
string method = message.Method ?? "";
if (Check(context, message) is { } violation)
{
return new RpcDispatch(
message.Id is null ? null : RpcMessage.Failure(message.Id, new FleetError(FleetErrors.ProtocolViolation, violation.Reason)),
violation);
}
if (!_routes.TryGetValue(method, out Route? route))
{
return message.Id is null ? null : RpcMessage.Failure(message.Id, new FleetError(FleetErrors.Unsupported, $"unknown method {method}"));
return new RpcDispatch(message.Id is null ? null : RpcMessage.Failure(message.Id, new FleetError(FleetErrors.Unsupported, $"unknown method {method}")), null);
}
Result<JsonElement?> result;
@@ -91,12 +133,90 @@ public sealed class RpcRouter<TContext>
if (message.Id is null)
{
return null;
return new RpcDispatch(null, null);
}
return result.TryGet(out JsonElement? value, out FleetError? error)
? new RpcMessage { Id = message.Id, Result = value ?? s_emptyObject }
: RpcMessage.Failure(message.Id, error);
return new RpcDispatch(
result.TryGet(out JsonElement? value, out FleetError? error)
? new RpcMessage { Id = message.Id, Result = value ?? s_emptyObject }
: RpcMessage.Failure(message.Id, error),
null);
}
/// <summary>入站检查:这条消息是不是对方有资格发起的、类型对不对、对方的角色允不允许。</summary>
public RpcViolation? Check(TContext context, RpcMessage message)
{
string method = message.Method ?? "";
MethodSpec? spec = Table.Find(Plane, method);
if (spec is null)
{
return new RpcViolation(method, $"method {method} is not declared");
}
if (spec.To != Side)
{
return new RpcViolation(method, $"{method} cannot be initiated by the {RpcSides.Opposite(Side)} side");
}
MethodKind kind = message.Id is null ? MethodKind.Notification : MethodKind.Request;
if (spec.Kind != kind)
{
return new RpcViolation(method, $"{method} is a {spec.Kind.ToString().ToLowerInvariant()}");
}
if (_callerRole is not null && !spec.AllowsRole(_callerRole(context)))
{
return new RpcViolation(method, $"{method} is not allowed for this node role");
}
return null;
}
/// <summary>
/// 发送前检查:本侧有没有资格发起这个方法、类型对不对、对端的角色允不允许(服务端发给节点时传入节点的角色)。
/// 违反是程序缺陷,抛 <see cref="InvalidOperationException"/>。
/// </summary>
public void CheckSend(string method, MethodKind kind, string? peerRole = null)
{
MethodSpec? spec = Table.Find(Plane, method);
if (spec is null)
{
throw new InvalidOperationException($"method {method} is not declared");
}
if (spec.From != Side)
{
throw new InvalidOperationException($"{method} cannot be initiated by the {Side} side");
}
if (spec.Kind != kind)
{
throw new InvalidOperationException($"{method} is a {spec.Kind.ToString().ToLowerInvariant()}");
}
if (peerRole is not null && !spec.AllowsRole(peerRole))
{
throw new InvalidOperationException($"{method} cannot be sent to a {peerRole}");
}
}
private void Declare(string method, MethodKind kind)
{
MethodSpec spec = Table.Find(Plane, method) ?? throw new InvalidOperationException($"method {method} is not declared in the method table");
if (spec.To != Side)
{
throw new InvalidOperationException($"{method} is initiated by the {spec.From} side and cannot be handled on the {Side} side");
}
if (spec.Kind != kind)
{
throw new InvalidOperationException($"{method} is declared as a {spec.Kind.ToString().ToLowerInvariant()}");
}
if (_localRole is not null && !spec.AllowsRole(_localRole))
{
throw new InvalidOperationException($"{method} is not for the {_localRole} role");
}
}
private static bool TryDeserialize<T>(JsonElement? raw, JsonTypeInfo<T> info, [System.Diagnostics.CodeAnalysis.NotNullWhen(true)] out T? value)
+27 -6
View File
@@ -34,6 +34,9 @@ public sealed class RpcSessionOptions
/// <summary>待处理的入站请求上限;超过时接收循环等待(背压)。</summary>
public int InboundQueue { get; set; } = 256;
/// <summary>对端违反协议(方法没有声明、方向、类型或角色不对)的次数上限;超过就断开会话。</summary>
public int MaxViolations { get; set; } = 20;
}
/// <summary>
@@ -53,6 +56,8 @@ public sealed class RpcSession<TContext> : IAsyncDisposable
private readonly CancellationTokenSource _stop = new();
private long _nextId;
private long _lastReceive;
private int _violations;
private bool _violated;
public RpcSession(IRpcTransport transport, RpcRouter<TContext> router, TContext context, TimeProvider time, RpcSessionOptions? options = null)
{
@@ -81,7 +86,7 @@ public sealed class RpcSession<TContext> : IAsyncDisposable
while (!linked.IsCancellationRequested)
{
RpcFrame? frame = await _transport.ReceiveAsync(linked.Token);
if (frame is null)
if (frame is null || Volatile.Read(ref _violated))
{
return;
}
@@ -184,8 +189,16 @@ public sealed class RpcSession<TContext> : IAsyncDisposable
public ValueTask NotifyAsync<TParams>(string method, TParams parameters, JsonTypeInfo<TParams> paramsInfo, CancellationToken ct) =>
SendAsync(RpcMessage.Notification(method, parameters, paramsInfo), ct);
public ValueTask SendAsync(RpcMessage message, CancellationToken ct) =>
_transport.SendAsync(new RpcFrame(FrameKind.Text, JsonSerializer.SerializeToUtf8Bytes(message, FleetRpcContext.Default.RpcMessage)), ct);
public ValueTask SendAsync(RpcMessage message, CancellationToken ct)
{
if (message.Method is not null)
{
// 请求和通知只能发本侧有资格发起的方法(响应没有方法名,不检查)
_router.CheckSend(message.Method, message.Id is null ? MethodKind.Notification : MethodKind.Request);
}
return _transport.SendAsync(new RpcFrame(FrameKind.Text, JsonSerializer.SerializeToUtf8Bytes(message, FleetRpcContext.Default.RpcMessage)), ct);
}
public ValueTask SendBinaryAsync(ReadOnlyMemory<byte> frame, CancellationToken ct) =>
_transport.SendAsync(new RpcFrame(FrameKind.Binary, frame), ct);
@@ -205,10 +218,18 @@ public sealed class RpcSession<TContext> : IAsyncDisposable
{
await foreach (RpcMessage message in _inbound.Reader.ReadAllAsync(ct))
{
RpcMessage? response = await _router.DispatchAsync(_context, message, ct);
if (response is not null)
RpcDispatch outcome = await _router.DispatchCheckedAsync(_context, message, ct);
if (outcome.Response is not null)
{
await SendAsync(response, ct);
await SendAsync(outcome.Response, ct);
}
if (outcome.Violation is not null && ++_violations > _options.MaxViolations)
{
// 只关闭传输,不取消接收:接收循环在下一次读到帧或对端确认关闭时正常结束
Volatile.Write(ref _violated, true);
await _transport.CloseAsync("protocol violations", CancellationToken.None);
return;
}
}
}
@@ -2,7 +2,7 @@ namespace Fleet.Server;
public sealed class FleetServerOptions
{
/// <summary>节点 ID 的类型前缀,Monitor 使用 <c>agt</c>。</summary>
/// <summary>节点 ID 的类型前缀,MoShiro 使用 <c>agt</c>。</summary>
public string NodeIdPrefix { get; set; } = "node";
public int MaxFrameBytes { get; set; } = 16 << 20;
@@ -12,4 +12,7 @@ public sealed class FleetServerOptions
public TimeSpan HeartbeatTimeout { get; set; } = TimeSpan.FromSeconds(45);
public TimeSpan RequestTimeout { get; set; } = TimeSpan.FromSeconds(15);
/// <summary>节点违反协议(方法没有声明、方向、类型或角色不对)的次数上限;超过就断开这条连接。</summary>
public int MaxViolations { get; set; } = 20;
}
@@ -2,7 +2,7 @@ using Fleet.Core;
namespace Fleet.Server;
/// <summary>Fleet 服务端需要的存储。产品可以自己实现(Monitor 把它放进 monitor.db)。</summary>
/// <summary>Fleet 服务端需要的存储。产品可以自己实现(MoShiro 把它放进 moshiro.db)。</summary>
public interface IFleetStore
{
ValueTask<JoinTokenRecord?> FindJoinTokenAsync(ReadOnlyMemory<byte> tokenSha256, CancellationToken ct);
@@ -33,6 +33,7 @@ public sealed class FleetLinkHost : IFleetConnections, IStateNotifier
Inboxes = inboxes.ToDictionary(i => i.Channel);
Sinks = [.. sinks];
Router = router.Inner;
RoleOf = router.RoleOf;
}
internal IFleetStore Store { get; }
@@ -51,6 +52,8 @@ public sealed class FleetLinkHost : IFleetConnections, IStateNotifier
internal RpcRouter<NodeCallContext> Router { get; }
internal Func<NodeRecord, string?>? RoleOf { get; }
public IReadOnlyCollection<NodeId> OnlineNodes => [.. _connections.Keys];
public bool IsOnline(NodeId node) => _connections.ContainsKey(node);
@@ -161,10 +164,18 @@ public sealed class FleetLinkHost : IFleetConnections, IStateNotifier
/// <summary>产品注册的 RPC 方法路由。</summary>
public sealed class FleetRouter
{
public FleetRouter(RpcRouter<NodeCallContext> inner)
public FleetRouter(RpcRouter<NodeCallContext> inner, Func<NodeRecord, string?>? roleOf)
{
Inner = inner;
RoleOf = roleOf;
}
internal RpcRouter<NodeCallContext> Inner { get; }
internal Func<NodeRecord, string?>? RoleOf { get; }
/// <summary>已注册的入站方法名(只读,测试用它核对声明表)。</summary>
public IReadOnlyCollection<string> Methods => Inner.Methods;
public MethodTable Table => Inner.Table;
}
@@ -19,6 +19,9 @@ internal static partial class FleetLog
[LoggerMessage(Level = LogLevel.Warning, Message = "state push failed node={Node} type={Type} err={Err}")]
public static partial void StatePushFailed(ILogger logger, string node, string type, string err);
[LoggerMessage(Level = LogLevel.Warning, Message = "protocol violation node={Node} method={Method} reason={Reason}")]
public static partial void ProtocolViolation(ILogger logger, string node, string method, string reason);
[LoggerMessage(Level = LogLevel.Error, Message = "unhandled error in handler method={Method}")]
public static partial void HandlerFailed(ILogger logger, Exception exception, string method);
@@ -12,6 +12,8 @@ namespace Fleet.Server;
public sealed class FleetServerBuilder
{
private readonly List<Action<RpcRouter<NodeCallContext>, IServiceProvider>> _routeConfigurators = [];
private MethodTable? _methods;
private Func<NodeRecord, string?>? _roleOf;
internal FleetServerBuilder(IServiceCollection services)
{
@@ -22,6 +24,21 @@ public sealed class FleetServerBuilder
internal IReadOnlyList<Action<RpcRouter<NodeCallContext>, IServiceProvider>> RouteConfigurators => _routeConfigurators;
internal MethodTable Methods => _methods ?? throw new InvalidOperationException("call WithMethods(...) when adding the Fleet server: the method table is required");
internal Func<NodeRecord, string?>? RoleOf => _roleOf;
/// <summary>
/// 方法声明表(框架的加产品的,见 <see cref="MethodTable.Combine"/>)和节点的角色解析(例如按节点 ID 前缀区分 Agent 和 Probe)。必须设置:
/// 路由器据此核对每个处理函数的方向、类型和角色,并拒绝没有声明或不该由节点发起的方法。
/// </summary>
public FleetServerBuilder WithMethods(MethodTable methods, Func<NodeRecord, string?> roleOf)
{
_methods = methods;
_roleOf = roleOf;
return this;
}
public FleetServerBuilder AddState<[DynamicallyAccessedMembers(DynamicallyAccessedMemberTypes.PublicConstructors)] TProvider>()
where TProvider : class, IStateProvider
{
@@ -50,7 +67,7 @@ public sealed class FleetServerBuilder
return this;
}
/// <summary>注册产品的 RPC 方法(<c>monitor.*</c> 等)。处理函数通过 <see cref="IServiceProvider"/> 在调用时解析服务,避免构造期的循环依赖。</summary>
/// <summary>注册产品的 RPC 方法(<c>moshiro.*</c> 等)。处理函数通过 <see cref="IServiceProvider"/> 在调用时解析服务,避免构造期的循环依赖。</summary>
public FleetServerBuilder MapMethods(Action<RpcRouter<NodeCallContext>, IServiceProvider> configure)
{
_routeConfigurators.Add(configure);
@@ -71,13 +88,18 @@ public static class FleetServerExtensions
services.AddSingleton(provider =>
{
ILogger logger = provider.GetRequiredService<ILoggerFactory>().CreateLogger("Fleet.Rpc");
var router = new RpcRouter<NodeCallContext>((method, ex) => FleetLog.HandlerFailed(logger, ex, method));
Func<NodeRecord, string?>? roleOf = builder.RoleOf;
var router = new RpcRouter<NodeCallContext>(
RpcSide.Server,
builder.Methods,
callerRole: roleOf is null ? null : context => roleOf(context.Node),
onUnhandled: (method, ex) => FleetLog.HandlerFailed(logger, ex, method));
foreach (Action<RpcRouter<NodeCallContext>, IServiceProvider> configurator in builder.RouteConfigurators)
{
configurator(router, provider);
}
return new FleetRouter(router);
return new FleetRouter(router, roleOf);
});
services.AddSingleton(provider => new FleetLinkHost(
provider.GetRequiredService<IFleetStore>(),
@@ -0,0 +1,98 @@
using System.Net.WebSockets;
using System.Security.Cryptography;
using System.Text.Json;
using Fleet.Core;
using Fleet.Identity;
using Fleet.Rpc;
namespace Fleet.Server;
/// <summary>认证:挑战、签名校验、拒绝。</summary>
internal sealed partial class NodeConnection
{
private async Task<bool> AuthenticateAsync(CancellationToken ct)
{
string nonce = Convert.ToBase64String(RandomNumberGenerator.GetBytes(32));
long expiresAtMs = _host.Time.GetUtcNow().Add(_host.Options.ChallengeTtl).ToUnixTimeMilliseconds();
await SendAsync(
RpcMessage.Notification(FleetMethods.Challenge, new ChallengeParams { Nonce = nonce, ExpiresAtMs = expiresAtMs }, FleetRpcContext.Default.ChallengeParams),
ct);
using var timeout = new CancellationTokenSource(s_authTimeout, _host.Time);
using CancellationTokenSource linked = CancellationTokenSource.CreateLinkedTokenSource(ct, timeout.Token);
Frame? frame;
try
{
frame = await ReceiveFrameAsync(linked.Token);
}
catch (OperationCanceledException) when (!ct.IsCancellationRequested)
{
await RejectAsync(null, "auth timeout", ct);
return false;
}
if (frame is not { IsText: true } || !TryParse(frame.Data, out RpcMessage? message) || message.Method != FleetMethods.Auth || message.Id is null)
{
await RejectAsync(null, "expected fleet.auth", ct);
return false;
}
AuthParams? auth = null;
try
{
auth = message.Params?.Deserialize(FleetRpcContext.Default.AuthParams);
}
catch (JsonException)
{
// 按缺失处理
}
NodeRecord? node = auth is null ? null : await _host.Store.GetNodeAsync(new NodeId(auth.NodeId), ct);
bool signatureOk = node is not null
&& _host.Time.GetUtcNow().ToUnixTimeMilliseconds() <= expiresAtMs
&& TryDecode(auth!.Signature, out byte[] signature)
&& NodeIdentity.Verify(node.PublicKey, signature, SignaturePurpose.Auth, auth.NodeId, nonce);
if (!signatureOk)
{
// 节点不存在与签名错误返回同样的错误,不泄露节点是否存在。
await RejectAsync(message.Id, "authentication failed", ct);
return false;
}
_node = node;
if (node!.Status == NodeStatus.Disabled)
{
await SendAsync(RpcMessage.Failure(message.Id, new FleetError(FleetErrors.NodeDisabled, "node is disabled")), ct);
await RevokeAsync("disabled", ct);
return false;
}
_session = new NodeSessionInfo(node.Id, _host.Time.GetUtcNow().ToUnixTimeMilliseconds(), node.NodeVersion, [], null);
await SendAsync(
RpcMessage.Success(message.Id.Value, new AuthResult { Status = StatusName(node.Status) }, FleetRpcContext.Default.AuthResult),
ct);
return true;
}
private async Task RejectAsync(long? id, string reason, CancellationToken ct)
{
FleetLog.AuthRejected(_host.Logger, reason);
_closeReason ??= reason;
if (id is not null)
{
await SendAsync(RpcMessage.Failure(id, new FleetError(FleetErrors.Unauthorized, "authentication failed")), ct);
}
if (_socket.State == WebSocketState.Open)
{
try
{
await _socket.CloseOutputAsync(WebSocketCloseStatus.PolicyViolation, "unauthorized", ct);
}
catch (WebSocketException)
{
// 对端已经断开
}
}
}
}
@@ -0,0 +1,240 @@
using System.Text.Json;
using Fleet.Core;
using Fleet.Rpc;
namespace Fleet.Server;
/// <summary>接收:读帧、分发消息、hello、心跳、二进制通道(收件箱)。</summary>
internal sealed partial class NodeConnection
{
private async Task ReceiveLoopAsync(CancellationToken ct)
{
while (!ct.IsCancellationRequested)
{
using var idle = new CancellationTokenSource(_host.Options.HeartbeatTimeout, _host.Time);
using CancellationTokenSource linked = CancellationTokenSource.CreateLinkedTokenSource(ct, idle.Token);
Frame? frame;
try
{
frame = await ReceiveFrameAsync(linked.Token);
}
catch (OperationCanceledException) when (idle.IsCancellationRequested && !ct.IsCancellationRequested)
{
_closeReason ??= "heartbeat timeout";
return;
}
if (frame is null)
{
_closeReason ??= "closed by peer";
return;
}
if (frame.IsText)
{
if (TryParse(frame.Data, out RpcMessage? message))
{
await HandleMessageAsync(message, ct);
}
}
else
{
await HandleBinaryAsync(frame.Data, ct);
}
}
}
private async Task HandleMessageAsync(RpcMessage message, CancellationToken ct)
{
if (message.IsResponse)
{
if (_pending.TryGetValue(message.Id!.Value, out TaskCompletionSource<RpcMessage>? waiter))
{
waiter.TrySetResult(message);
}
return;
}
switch (message.Method)
{
case FleetMethods.Hello:
await HandleHelloAsync(message, ct);
return;
case FleetMethods.Heartbeat:
await HandleHeartbeatAsync(message, ct);
return;
}
if (_node!.Status != NodeStatus.Active)
{
if (message.Id is not null)
{
await SendAsync(RpcMessage.Failure(message.Id, new FleetError(FleetErrors.NodePending, "node is not approved yet")), ct);
}
return;
}
var context = new NodeCallContext(_node, _session!, _host);
RpcDispatch outcome;
try
{
outcome = await _host.Router.DispatchCheckedAsync(context, message, ct);
}
catch (OperationCanceledException) when (ct.IsCancellationRequested)
{
throw;
}
if (outcome.Response is not null)
{
await SendAsync(outcome.Response, ct);
}
await NoteViolationAsync(outcome.Violation);
}
/// <summary>记录节点的违规;超过上限就断开这条连接。</summary>
private async Task NoteViolationAsync(RpcViolation? violation)
{
if (violation is not { } v)
{
return;
}
FleetLog.ProtocolViolation(_host.Logger, _node!.Id.Value, v.Method, v.Reason);
if (++_violations > _host.Options.MaxViolations)
{
await CloseAsync("protocol violations");
}
}
private async Task HandleHelloAsync(RpcMessage message, CancellationToken ct)
{
HelloParams? hello = null;
try
{
hello = message.Params?.Deserialize(FleetRpcContext.Default.HelloParams);
}
catch (JsonException)
{
// 按缺失处理
}
if (hello is null)
{
if (message.Id is not null)
{
await SendAsync(RpcMessage.Failure(message.Id, new FleetError(FleetErrors.InvalidParams, "invalid hello")), ct);
}
return;
}
long nowMs = _host.Time.GetUtcNow().ToUnixTimeMilliseconds();
_session = _session! with
{
NodeVersion = hello.NodeVersion,
Capabilities = hello.Capabilities,
HelloMetaJson = hello.Meta?.GetRawText(),
};
_node = _node! with { NodeVersion = hello.NodeVersion };
_helloDone = true;
await UpdatePresenceAsync(online: true, ct);
await FireAsync(NodeEventKind.Hello, ct, hello);
if (message.Id is not null)
{
await SendAsync(
RpcMessage.Success(
message.Id.Value,
new HelloResult { ServerTimeMs = nowMs, ClockOffsetMs = hello.NowUtcMs - nowMs },
FleetRpcContext.Default.HelloResult),
ct);
}
SignalStateChanged();
}
private async Task HandleHeartbeatAsync(RpcMessage message, CancellationToken ct)
{
JsonElement? extra = null;
try
{
extra = message.Params?.Deserialize(FleetRpcContext.Default.HeartbeatParams)?.Extra;
}
catch (JsonException)
{
// 忽略格式不正确的附加字段
}
long now = _host.Time.GetTimestamp();
if (_host.Time.GetElapsedTime(Interlocked.Read(ref _lastPresenceTicks), now) >= s_presenceRefresh)
{
await UpdatePresenceAsync(online: true, ct);
}
await FireAsync(NodeEventKind.Heartbeat, ct, hello: null, extra);
if (message.Id is not null)
{
await SendAsync(new RpcMessage { Id = message.Id, Result = JsonSerializer.SerializeToElement(new EmptyObject(), FleetEmptyObject.Info) }, ct);
}
}
private async Task HandleBinaryAsync(byte[] data, CancellationToken ct)
{
if (_node!.Status != NodeStatus.Active || !OutboxFrame.TryParse(data, out byte channel, out Guid msgId, out ReadOnlyMemory<byte> payload))
{
return;
}
// 通道也要声明:节点只能往声明过、并且允许它的角色使用的通道发
ChannelSpec? spec = _host.Router.Table.FindChannel(RpcPlane.Node, channel);
if (spec is null || spec.From != RpcSide.Node || !spec.AllowsRole(_host.RoleOf?.Invoke(_node)))
{
await NoteViolationAsync(new RpcViolation($"channel {channel}", $"channel {channel} is not declared for this node"));
return;
}
if (!_host.Inboxes.TryGetValue(channel, out IInbox? inbox))
{
return;
}
InboxOutcome outcome;
try
{
outcome = await inbox.HandleAsync(_node, msgId, payload, ct);
}
catch (OperationCanceledException) when (ct.IsCancellationRequested)
{
throw;
}
catch (Exception ex)
{
// 没有持久化就不确认,节点会重发。
FleetLog.InboxFailed(_host.Logger, ex, _node.Id.Value, channel);
return;
}
if (outcome == InboxOutcome.Rejected)
{
return;
}
var ack = new OutboxAckParams();
string id = msgId.ToString();
if (outcome == InboxOutcome.Duplicate)
{
ack.Duplicates.Add(id);
}
else
{
ack.Ids.Add(id);
}
await SendAsync(RpcMessage.Notification(FleetMethods.OutboxAck, ack, FleetRpcContext.Default.OutboxAckParams), ct);
}
}
@@ -0,0 +1,120 @@
using System.Net.WebSockets;
using System.Text.Json;
using Fleet.Core;
using Fleet.Rpc;
namespace Fleet.Server;
/// <summary>发给节点:关闭、吊销、发送一条消息、发起请求并等响应。</summary>
internal sealed partial class NodeConnection
{
/// <summary>
/// 服务端发起的关闭:发送 Close 帧,等对端确认之后接收循环自然结束。
/// 注意不能直接取消接收的令牌——取消 <c>ReceiveAsync</c> 会中止整个 WebSocket,对端拿不到关闭原因;取消只作为超时兜底。
/// </summary>
public async ValueTask CloseAsync(string reason)
{
_closeReason ??= reason;
try
{
await _sendLock.WaitAsync(CancellationToken.None);
}
catch (ObjectDisposedException)
{
return;
}
try
{
if (_socket.State == WebSocketState.Open)
{
await _socket.CloseOutputAsync(WebSocketCloseStatus.NormalClosure, reason, CancellationToken.None);
}
}
catch (WebSocketException)
{
// 对端已经断开
}
finally
{
_sendLock.Release();
}
_lifetime.CancelAfter(s_closeTimeout);
}
public async ValueTask RevokeAsync(string reason, CancellationToken ct)
{
await SendAsync(RpcMessage.Notification(FleetMethods.Revoked, new RevokedParams { Reason = reason }, FleetRpcContext.Default.RevokedParams), ct);
await CloseAsync(reason);
}
public async ValueTask<bool> SendAsync(RpcMessage message, CancellationToken ct)
{
if (message.Method is not null)
{
// 服务端只能发自己有资格发起的方法,并且对端的角色要允许(响应没有方法名,不检查)
string? peerRole = _node is null ? null : _host.RoleOf?.Invoke(_node);
_host.Router.CheckSend(message.Method, message.Id is null ? MethodKind.Notification : MethodKind.Request, peerRole);
}
byte[] bytes = JsonSerializer.SerializeToUtf8Bytes(message, FleetRpcContext.Default.RpcMessage);
try
{
await _sendLock.WaitAsync(ct);
}
catch (ObjectDisposedException)
{
return false;
}
try
{
if (_socket.State != WebSocketState.Open)
{
return false;
}
await _socket.SendAsync(bytes, WebSocketMessageType.Text, endOfMessage: true, ct);
return true;
}
catch (WebSocketException)
{
return false;
}
finally
{
_sendLock.Release();
}
}
public async ValueTask<Result<JsonElement>> RequestAsync(string method, JsonElement parameters, CancellationToken ct)
{
long id = Interlocked.Increment(ref _nextRequestId);
var waiter = new TaskCompletionSource<RpcMessage>(TaskCreationOptions.RunContinuationsAsynchronously);
_pending[id] = waiter;
try
{
var request = new RpcMessage { Id = id, Method = method, Params = parameters };
if (!await SendAsync(request, ct))
{
return Result<JsonElement>.Fail(FleetErrors.NodeOffline, "node is not connected");
}
using var timeout = new CancellationTokenSource(_host.Options.RequestTimeout, _host.Time);
using CancellationTokenSource linked = CancellationTokenSource.CreateLinkedTokenSource(ct, timeout.Token);
RpcMessage response = await waiter.Task.WaitAsync(linked.Token);
return response.Error is not null
? Result<JsonElement>.Fail(response.Error.ToFleetError())
: Result<JsonElement>.Ok(response.Result ?? default);
}
catch (OperationCanceledException) when (!ct.IsCancellationRequested)
{
return Result<JsonElement>.Fail(FleetErrors.Timeout, $"{method} timed out");
}
finally
{
_pending.TryRemove(id, out _);
}
}
}
@@ -0,0 +1,123 @@
using System.Text.Json;
using Fleet.Core;
using Fleet.Rpc;
namespace Fleet.Server;
/// <summary>状态:期望状态推送、在线状态、节点事件。</summary>
internal sealed partial class NodeConnection
{
public void SignalStateChanged() => _stateSignal.Writer.TryWrite(true);
private async Task StatePushLoopAsync(CancellationToken ct)
{
while (await _stateSignal.Reader.WaitToReadAsync(ct))
{
_stateSignal.Reader.TryRead(out _);
bool complete = await PushStateAsync(ct);
if (!complete)
{
await Task.Delay(s_statePushRetry, _host.Time, ct);
SignalStateChanged();
}
}
}
private async Task<bool> PushStateAsync(CancellationToken ct)
{
NodeRecord? fresh = await _host.Store.GetNodeAsync(_node!.Id, ct);
if (fresh is null)
{
return true;
}
_node = fresh with { NodeVersion = _node.NodeVersion ?? fresh.NodeVersion };
if (fresh.Status == NodeStatus.Disabled)
{
await RevokeAsync("disabled", ct);
return true;
}
if (fresh.Status != NodeStatus.Active || !_helloDone)
{
// pending 不下发状态;还没有 hello 的连接等 hello 之后再推送。
return true;
}
bool complete = true;
foreach (IStateProvider provider in _host.StateProviders)
{
StateDocument? document = await provider.GetAsync(_node, ct);
if (document is null || (_lastPushedVersion.TryGetValue(provider.Type, out long last) && last == document.Version))
{
continue;
}
Result<JsonElement> raw = await RequestAsync(
FleetMethods.StateSync,
JsonSerializer.SerializeToElement(
new StateSyncParams { Type = provider.Type, Version = document.Version, Doc = document.Doc },
FleetRpcContext.Default.StateSyncParams),
ct);
if (!raw.TryGet(out JsonElement value, out FleetError? error))
{
FleetLog.StatePushFailed(_host.Logger, _node.Id.Value, provider.Type, error.Code.Value);
complete = false;
continue;
}
StateSyncResult? result = null;
try
{
result = value.Deserialize(FleetRpcContext.Default.StateSyncResult);
}
catch (JsonException)
{
// 按缺失处理
}
if (result is null)
{
complete = false;
continue;
}
_lastPushedVersion[provider.Type] = document.Version;
await provider.OnAppliedAsync(_node, document.Version, result, ct);
}
return complete;
}
private async Task UpdatePresenceAsync(bool online, CancellationToken ct)
{
Interlocked.Exchange(ref _lastPresenceTicks, _host.Time.GetTimestamp());
await _host.Store.UpdatePresenceAsync(
_node!.Id,
new Presence(online, _host.Time.GetUtcNow().ToUnixTimeMilliseconds(), _session?.NodeVersion, _session?.Capabilities ?? [], _session?.HelloMetaJson),
ct);
}
private async Task FireAsync(NodeEventKind kind, CancellationToken ct, HelloParams? hello = null, JsonElement? extra = null)
{
var nodeEvent = new NodeEvent(kind, _node!, _host.Time.GetUtcNow().ToUnixTimeMilliseconds(), hello, extra);
foreach (INodeEventSink sink in _host.Sinks)
{
try
{
await sink.OnEventAsync(nodeEvent, ct);
}
catch (OperationCanceledException) when (ct.IsCancellationRequested)
{
throw;
}
catch (Exception ex)
{
// 产品的事件处理失败不能影响连接。
FleetLog.SinkFailed(_host.Logger, ex, kind.ToString(), _node!.Id.Value);
}
}
}
}
@@ -0,0 +1,85 @@
using System.Buffers;
using System.Net.WebSockets;
using System.Text.Json;
using Fleet.Rpc;
namespace Fleet.Server;
/// <summary>线上的帧:WebSocket 收帧、解析、状态名。</summary>
internal sealed partial class NodeConnection
{
private async ValueTask<Frame?> ReceiveFrameAsync(CancellationToken ct)
{
var writer = new ArrayBufferWriter<byte>(1024);
while (true)
{
Memory<byte> buffer = writer.GetMemory(4096);
ValueWebSocketReceiveResult result = await _socket.ReceiveAsync(buffer, ct);
if (result.MessageType == WebSocketMessageType.Close)
{
return null;
}
writer.Advance(result.Count);
if (writer.WrittenCount > _host.Options.MaxFrameBytes)
{
_closeReason ??= "frame too large";
await _socket.CloseOutputAsync(WebSocketCloseStatus.MessageTooBig, "frame too large", CancellationToken.None);
return null;
}
if (result.EndOfMessage)
{
return new Frame(result.MessageType == WebSocketMessageType.Text, writer.WrittenSpan.ToArray());
}
}
}
private static bool TryParse(byte[] data, [System.Diagnostics.CodeAnalysis.NotNullWhen(true)] out RpcMessage? message)
{
try
{
message = JsonSerializer.Deserialize(data, FleetRpcContext.Default.RpcMessage);
return message is not null;
}
catch (JsonException)
{
message = null;
return false;
}
}
private static bool TryDecode(string text, out byte[] data)
{
try
{
data = Convert.FromBase64String(text);
return true;
}
catch (FormatException)
{
data = [];
return false;
}
}
private static string StatusName(NodeStatus status) => status switch
{
NodeStatus.Active => "active",
NodeStatus.Pending => "pending",
_ => "disabled",
};
private sealed record Frame(bool IsText, byte[] Data);
}
[System.Text.Json.Serialization.JsonSerializable(typeof(EmptyObject))]
internal sealed partial class FleetEmptyObjectContext : System.Text.Json.Serialization.JsonSerializerContext;
internal sealed class EmptyObject;
internal static class FleetEmptyObject
{
public static System.Text.Json.Serialization.Metadata.JsonTypeInfo<EmptyObject> Info => FleetEmptyObjectContext.Default.EmptyObject;
}
@@ -1,11 +1,7 @@
using System.Buffers;
using System.Collections.Concurrent;
using System.Net.WebSockets;
using System.Security.Cryptography;
using System.Threading.Channels;
using System.Text.Json;
using Fleet.Core;
using Fleet.Identity;
using Fleet.Rpc;
namespace Fleet.Server;
@@ -14,7 +10,7 @@ namespace Fleet.Server;
/// 一个节点的 WebSocket 会话:挑战 → 签名认证 → hello → 状态推送 / 收件箱 / 产品 RPC。
/// 收到的消息按顺序处理,所以产品的处理函数不能等待“向同一个节点发起的请求”,否则会死锁。
/// </summary>
internal sealed class NodeConnection : IDisposable
internal sealed partial class NodeConnection : IDisposable
{
private static readonly TimeSpan s_authTimeout = TimeSpan.FromSeconds(15);
private static readonly TimeSpan s_presenceRefresh = TimeSpan.FromSeconds(60);
@@ -34,6 +30,7 @@ internal sealed class NodeConnection : IDisposable
private NodeSessionInfo? _session;
private string? _closeReason;
private bool _helloDone;
private int _violations;
public NodeConnection(FleetLinkHost host, WebSocket socket)
{
@@ -140,590 +137,4 @@ internal sealed class NodeConnection : IDisposable
_lifetime.Dispose();
_sendLock.Dispose();
}
public void SignalStateChanged() => _stateSignal.Writer.TryWrite(true);
/// <summary>
/// 服务端发起的关闭:发送 Close 帧,等对端确认之后接收循环自然结束。
/// 注意不能直接取消接收的令牌——取消 <c>ReceiveAsync</c> 会中止整个 WebSocket,对端拿不到关闭原因;取消只作为超时兜底。
/// </summary>
public async ValueTask CloseAsync(string reason)
{
_closeReason ??= reason;
try
{
await _sendLock.WaitAsync(CancellationToken.None);
}
catch (ObjectDisposedException)
{
return;
}
try
{
if (_socket.State == WebSocketState.Open)
{
await _socket.CloseOutputAsync(WebSocketCloseStatus.NormalClosure, reason, CancellationToken.None);
}
}
catch (WebSocketException)
{
// 对端已经断开
}
finally
{
_sendLock.Release();
}
_lifetime.CancelAfter(s_closeTimeout);
}
public async ValueTask RevokeAsync(string reason, CancellationToken ct)
{
await SendAsync(RpcMessage.Notification(FleetMethods.Revoked, new RevokedParams { Reason = reason }, FleetRpcContext.Default.RevokedParams), ct);
await CloseAsync(reason);
}
public async ValueTask<bool> SendAsync(RpcMessage message, CancellationToken ct)
{
byte[] bytes = JsonSerializer.SerializeToUtf8Bytes(message, FleetRpcContext.Default.RpcMessage);
try
{
await _sendLock.WaitAsync(ct);
}
catch (ObjectDisposedException)
{
return false;
}
try
{
if (_socket.State != WebSocketState.Open)
{
return false;
}
await _socket.SendAsync(bytes, WebSocketMessageType.Text, endOfMessage: true, ct);
return true;
}
catch (WebSocketException)
{
return false;
}
finally
{
_sendLock.Release();
}
}
public async ValueTask<Result<JsonElement>> RequestAsync(string method, JsonElement parameters, CancellationToken ct)
{
long id = Interlocked.Increment(ref _nextRequestId);
var waiter = new TaskCompletionSource<RpcMessage>(TaskCreationOptions.RunContinuationsAsynchronously);
_pending[id] = waiter;
try
{
var request = new RpcMessage { Id = id, Method = method, Params = parameters };
if (!await SendAsync(request, ct))
{
return Result<JsonElement>.Fail(FleetErrors.NodeOffline, "node is not connected");
}
using var timeout = new CancellationTokenSource(_host.Options.RequestTimeout, _host.Time);
using CancellationTokenSource linked = CancellationTokenSource.CreateLinkedTokenSource(ct, timeout.Token);
RpcMessage response = await waiter.Task.WaitAsync(linked.Token);
return response.Error is not null
? Result<JsonElement>.Fail(response.Error.ToFleetError())
: Result<JsonElement>.Ok(response.Result ?? default);
}
catch (OperationCanceledException) when (!ct.IsCancellationRequested)
{
return Result<JsonElement>.Fail(FleetErrors.Timeout, $"{method} timed out");
}
finally
{
_pending.TryRemove(id, out _);
}
}
// ───────────────────────── 认证 ─────────────────────────
private async Task<bool> AuthenticateAsync(CancellationToken ct)
{
string nonce = Convert.ToBase64String(RandomNumberGenerator.GetBytes(32));
long expiresAtMs = _host.Time.GetUtcNow().Add(_host.Options.ChallengeTtl).ToUnixTimeMilliseconds();
await SendAsync(
RpcMessage.Notification(FleetMethods.Challenge, new ChallengeParams { Nonce = nonce, ExpiresAtMs = expiresAtMs }, FleetRpcContext.Default.ChallengeParams),
ct);
using var timeout = new CancellationTokenSource(s_authTimeout, _host.Time);
using CancellationTokenSource linked = CancellationTokenSource.CreateLinkedTokenSource(ct, timeout.Token);
Frame? frame;
try
{
frame = await ReceiveFrameAsync(linked.Token);
}
catch (OperationCanceledException) when (!ct.IsCancellationRequested)
{
await RejectAsync(null, "auth timeout", ct);
return false;
}
if (frame is not { IsText: true } || !TryParse(frame.Data, out RpcMessage? message) || message.Method != FleetMethods.Auth || message.Id is null)
{
await RejectAsync(null, "expected fleet.auth", ct);
return false;
}
AuthParams? auth = null;
try
{
auth = message.Params?.Deserialize(FleetRpcContext.Default.AuthParams);
}
catch (JsonException)
{
// 按缺失处理
}
NodeRecord? node = auth is null ? null : await _host.Store.GetNodeAsync(new NodeId(auth.NodeId), ct);
bool signatureOk = node is not null
&& _host.Time.GetUtcNow().ToUnixTimeMilliseconds() <= expiresAtMs
&& TryDecode(auth!.Signature, out byte[] signature)
&& NodeIdentity.Verify(node.PublicKey, signature, SignaturePurpose.Auth, auth.NodeId, nonce);
if (!signatureOk)
{
// 节点不存在与签名错误返回同样的错误,不泄露节点是否存在。
await RejectAsync(message.Id, "authentication failed", ct);
return false;
}
_node = node;
if (node!.Status == NodeStatus.Disabled)
{
await SendAsync(RpcMessage.Failure(message.Id, new FleetError(FleetErrors.NodeDisabled, "node is disabled")), ct);
await RevokeAsync("disabled", ct);
return false;
}
_session = new NodeSessionInfo(node.Id, _host.Time.GetUtcNow().ToUnixTimeMilliseconds(), node.NodeVersion, [], null);
await SendAsync(
RpcMessage.Success(message.Id.Value, new AuthResult { Status = StatusName(node.Status) }, FleetRpcContext.Default.AuthResult),
ct);
return true;
}
private async Task RejectAsync(long? id, string reason, CancellationToken ct)
{
FleetLog.AuthRejected(_host.Logger, reason);
_closeReason ??= reason;
if (id is not null)
{
await SendAsync(RpcMessage.Failure(id, new FleetError(FleetErrors.Unauthorized, "authentication failed")), ct);
}
if (_socket.State == WebSocketState.Open)
{
try
{
await _socket.CloseOutputAsync(WebSocketCloseStatus.PolicyViolation, "unauthorized", ct);
}
catch (WebSocketException)
{
// 对端已经断开
}
}
}
// ───────────────────────── 接收循环 ─────────────────────────
private async Task ReceiveLoopAsync(CancellationToken ct)
{
while (!ct.IsCancellationRequested)
{
using var idle = new CancellationTokenSource(_host.Options.HeartbeatTimeout, _host.Time);
using CancellationTokenSource linked = CancellationTokenSource.CreateLinkedTokenSource(ct, idle.Token);
Frame? frame;
try
{
frame = await ReceiveFrameAsync(linked.Token);
}
catch (OperationCanceledException) when (idle.IsCancellationRequested && !ct.IsCancellationRequested)
{
_closeReason ??= "heartbeat timeout";
return;
}
if (frame is null)
{
_closeReason ??= "closed by peer";
return;
}
if (frame.IsText)
{
if (TryParse(frame.Data, out RpcMessage? message))
{
await HandleMessageAsync(message, ct);
}
}
else
{
await HandleBinaryAsync(frame.Data, ct);
}
}
}
private async Task HandleMessageAsync(RpcMessage message, CancellationToken ct)
{
if (message.IsResponse)
{
if (_pending.TryGetValue(message.Id!.Value, out TaskCompletionSource<RpcMessage>? waiter))
{
waiter.TrySetResult(message);
}
return;
}
switch (message.Method)
{
case FleetMethods.Hello:
await HandleHelloAsync(message, ct);
return;
case FleetMethods.Heartbeat:
await HandleHeartbeatAsync(message, ct);
return;
}
if (_node!.Status != NodeStatus.Active)
{
if (message.Id is not null)
{
await SendAsync(RpcMessage.Failure(message.Id, new FleetError(FleetErrors.NodePending, "node is not approved yet")), ct);
}
return;
}
var context = new NodeCallContext(_node, _session!, _host);
RpcMessage? response;
try
{
response = await _host.Router.DispatchAsync(context, message, ct);
}
catch (OperationCanceledException) when (ct.IsCancellationRequested)
{
throw;
}
if (response is not null)
{
await SendAsync(response, ct);
}
}
private async Task HandleHelloAsync(RpcMessage message, CancellationToken ct)
{
HelloParams? hello = null;
try
{
hello = message.Params?.Deserialize(FleetRpcContext.Default.HelloParams);
}
catch (JsonException)
{
// 按缺失处理
}
if (hello is null)
{
if (message.Id is not null)
{
await SendAsync(RpcMessage.Failure(message.Id, new FleetError(FleetErrors.InvalidParams, "invalid hello")), ct);
}
return;
}
long nowMs = _host.Time.GetUtcNow().ToUnixTimeMilliseconds();
_session = _session! with
{
NodeVersion = hello.NodeVersion,
Capabilities = hello.Capabilities,
HelloMetaJson = hello.Meta?.GetRawText(),
};
_node = _node! with { NodeVersion = hello.NodeVersion };
_helloDone = true;
await UpdatePresenceAsync(online: true, ct);
await FireAsync(NodeEventKind.Hello, ct, hello);
if (message.Id is not null)
{
await SendAsync(
RpcMessage.Success(
message.Id.Value,
new HelloResult { ServerTimeMs = nowMs, ClockOffsetMs = hello.NowUtcMs - nowMs },
FleetRpcContext.Default.HelloResult),
ct);
}
SignalStateChanged();
}
private async Task HandleHeartbeatAsync(RpcMessage message, CancellationToken ct)
{
JsonElement? extra = null;
try
{
extra = message.Params?.Deserialize(FleetRpcContext.Default.HeartbeatParams)?.Extra;
}
catch (JsonException)
{
// 忽略格式不正确的附加字段
}
long now = _host.Time.GetTimestamp();
if (_host.Time.GetElapsedTime(Interlocked.Read(ref _lastPresenceTicks), now) >= s_presenceRefresh)
{
await UpdatePresenceAsync(online: true, ct);
}
await FireAsync(NodeEventKind.Heartbeat, ct, hello: null, extra);
if (message.Id is not null)
{
await SendAsync(new RpcMessage { Id = message.Id, Result = JsonSerializer.SerializeToElement(new EmptyObject(), FleetEmptyObject.Info) }, ct);
}
}
private async Task HandleBinaryAsync(byte[] data, CancellationToken ct)
{
if (_node!.Status != NodeStatus.Active
|| !OutboxFrame.TryParse(data, out byte channel, out Guid msgId, out ReadOnlyMemory<byte> payload)
|| !_host.Inboxes.TryGetValue(channel, out IInbox? inbox))
{
return;
}
InboxOutcome outcome;
try
{
outcome = await inbox.HandleAsync(_node, msgId, payload, ct);
}
catch (OperationCanceledException) when (ct.IsCancellationRequested)
{
throw;
}
catch (Exception ex)
{
// 没有持久化就不确认,节点会重发。
FleetLog.InboxFailed(_host.Logger, ex, _node.Id.Value, channel);
return;
}
if (outcome == InboxOutcome.Rejected)
{
return;
}
var ack = new OutboxAckParams();
string id = msgId.ToString();
if (outcome == InboxOutcome.Duplicate)
{
ack.Duplicates.Add(id);
}
else
{
ack.Ids.Add(id);
}
await SendAsync(RpcMessage.Notification(FleetMethods.OutboxAck, ack, FleetRpcContext.Default.OutboxAckParams), ct);
}
// ───────────────────────── 状态推送 ─────────────────────────
private async Task StatePushLoopAsync(CancellationToken ct)
{
while (await _stateSignal.Reader.WaitToReadAsync(ct))
{
_stateSignal.Reader.TryRead(out _);
bool complete = await PushStateAsync(ct);
if (!complete)
{
await Task.Delay(s_statePushRetry, _host.Time, ct);
SignalStateChanged();
}
}
}
private async Task<bool> PushStateAsync(CancellationToken ct)
{
NodeRecord? fresh = await _host.Store.GetNodeAsync(_node!.Id, ct);
if (fresh is null)
{
return true;
}
_node = fresh with { NodeVersion = _node.NodeVersion ?? fresh.NodeVersion };
if (fresh.Status == NodeStatus.Disabled)
{
await RevokeAsync("disabled", ct);
return true;
}
if (fresh.Status != NodeStatus.Active || !_helloDone)
{
// pending 不下发状态;还没有 hello 的连接等 hello 之后再推送。
return true;
}
bool complete = true;
foreach (IStateProvider provider in _host.StateProviders)
{
StateDocument? document = await provider.GetAsync(_node, ct);
if (document is null || (_lastPushedVersion.TryGetValue(provider.Type, out long last) && last == document.Version))
{
continue;
}
Result<JsonElement> raw = await RequestAsync(
FleetMethods.StateSync,
JsonSerializer.SerializeToElement(
new StateSyncParams { Type = provider.Type, Version = document.Version, Doc = document.Doc },
FleetRpcContext.Default.StateSyncParams),
ct);
if (!raw.TryGet(out JsonElement value, out FleetError? error))
{
FleetLog.StatePushFailed(_host.Logger, _node.Id.Value, provider.Type, error.Code.Value);
complete = false;
continue;
}
StateSyncResult? result = null;
try
{
result = value.Deserialize(FleetRpcContext.Default.StateSyncResult);
}
catch (JsonException)
{
// 按缺失处理
}
if (result is null)
{
complete = false;
continue;
}
_lastPushedVersion[provider.Type] = document.Version;
await provider.OnAppliedAsync(_node, document.Version, result, ct);
}
return complete;
}
// ───────────────────────── 帮助方法 ─────────────────────────
private async Task UpdatePresenceAsync(bool online, CancellationToken ct)
{
Interlocked.Exchange(ref _lastPresenceTicks, _host.Time.GetTimestamp());
await _host.Store.UpdatePresenceAsync(
_node!.Id,
new Presence(online, _host.Time.GetUtcNow().ToUnixTimeMilliseconds(), _session?.NodeVersion, _session?.Capabilities ?? [], _session?.HelloMetaJson),
ct);
}
private async Task FireAsync(NodeEventKind kind, CancellationToken ct, HelloParams? hello = null, JsonElement? extra = null)
{
var nodeEvent = new NodeEvent(kind, _node!, _host.Time.GetUtcNow().ToUnixTimeMilliseconds(), hello, extra);
foreach (INodeEventSink sink in _host.Sinks)
{
try
{
await sink.OnEventAsync(nodeEvent, ct);
}
catch (OperationCanceledException) when (ct.IsCancellationRequested)
{
throw;
}
catch (Exception ex)
{
// 产品的事件处理失败不能影响连接。
FleetLog.SinkFailed(_host.Logger, ex, kind.ToString(), _node!.Id.Value);
}
}
}
private async ValueTask<Frame?> ReceiveFrameAsync(CancellationToken ct)
{
var writer = new ArrayBufferWriter<byte>(1024);
while (true)
{
Memory<byte> buffer = writer.GetMemory(4096);
ValueWebSocketReceiveResult result = await _socket.ReceiveAsync(buffer, ct);
if (result.MessageType == WebSocketMessageType.Close)
{
return null;
}
writer.Advance(result.Count);
if (writer.WrittenCount > _host.Options.MaxFrameBytes)
{
_closeReason ??= "frame too large";
await _socket.CloseOutputAsync(WebSocketCloseStatus.MessageTooBig, "frame too large", CancellationToken.None);
return null;
}
if (result.EndOfMessage)
{
return new Frame(result.MessageType == WebSocketMessageType.Text, writer.WrittenSpan.ToArray());
}
}
}
private static bool TryParse(byte[] data, [System.Diagnostics.CodeAnalysis.NotNullWhen(true)] out RpcMessage? message)
{
try
{
message = JsonSerializer.Deserialize(data, FleetRpcContext.Default.RpcMessage);
return message is not null;
}
catch (JsonException)
{
message = null;
return false;
}
}
private static bool TryDecode(string text, out byte[] data)
{
try
{
data = Convert.FromBase64String(text);
return true;
}
catch (FormatException)
{
data = [];
return false;
}
}
private static string StatusName(NodeStatus status) => status switch
{
NodeStatus.Active => "active",
NodeStatus.Pending => "pending",
_ => "disabled",
};
private sealed record Frame(bool IsText, byte[] Data);
}
[System.Text.Json.Serialization.JsonSerializable(typeof(EmptyObject))]
internal sealed partial class FleetEmptyObjectContext : System.Text.Json.Serialization.JsonSerializerContext;
internal sealed class EmptyObject;
internal static class FleetEmptyObject
{
public static System.Text.Json.Serialization.Metadata.JsonTypeInfo<EmptyObject> Info => FleetEmptyObjectContext.Default.EmptyObject;
}
@@ -9,6 +9,7 @@
<ProjectReference Include="..\..\src\Fleet.Identity\Fleet.Identity.csproj" />
<ProjectReference Include="..\..\src\Fleet.Rpc\Fleet.Rpc.csproj" />
<ProjectReference Include="..\..\src\Fleet.Link\Fleet.Link.csproj" />
<ProjectReference Include="..\..\src\Fleet.Hosting\Fleet.Hosting.csproj" />
</ItemGroup>
<ItemGroup>
<Using Include="Xunit" />
@@ -67,7 +67,7 @@ public sealed class IdentityAndRpcTests
[Fact]
public async Task Router_DispatchesTypedRequests_AndReportsStructuredErrors()
{
var router = new RpcRouter<string>()
var router = new RpcRouter<string>(RpcSide.Listener, TestTables.Of(TestTables.Req("test.add"), TestTables.Req("test.boom"), TestTables.Req("test.nope")))
.On("test.add", EchoContext.Default.AddParams, EchoContext.Default.AddResult,
(caller, p, _) => ValueTask.FromResult(p.A < 0 ? Result<AddResult>.Fail(FleetErrors.Forbidden, "negative") : Result<AddResult>.Ok(new AddResult { Sum = p.A + p.B, Caller = caller })))
.On<AddParams, AddResult>("test.boom", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) => throw new InvalidOperationException("secret detail"));
@@ -92,7 +92,7 @@ public sealed class IdentityAndRpcTests
public async Task Router_HandlerExceptions_AreContained_AndNeverLeakDetails()
{
string? logged = null;
var router = new RpcRouter<string>((method, ex) => logged = method + ":" + ex.Message)
var router = new RpcRouter<string>(RpcSide.Listener, TestTables.Of(TestTables.Req("test.boom")), onUnhandled: (method, ex) => logged = method + ":" + ex.Message)
.On<AddParams, AddResult>("test.boom", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) => throw new InvalidOperationException("secret detail"));
RpcMessage reply = (await router.DispatchAsync("a", Request(1, "test.boom", """{"a":1,"b":1}"""), CancellationToken.None))!;
@@ -105,7 +105,7 @@ public sealed class IdentityAndRpcTests
public async Task Router_Notifications_ProduceNoResponse_EvenOnFailure()
{
int calls = 0;
var router = new RpcRouter<string>().OnNotify("test.note", EchoContext.Default.AddParams, (_, _, _) =>
var router = new RpcRouter<string>(RpcSide.Listener, TestTables.Of(TestTables.Note("test.note"), TestTables.Note("test.unknown"))).OnNotify("test.note", EchoContext.Default.AddParams, (_, _, _) =>
{
calls++;
return ValueTask.CompletedTask;
+10 -8
View File
@@ -227,8 +227,9 @@ public sealed class LinkTests
{
(TlsFrameTransport left, TlsFrameTransport right) = await PairAsync();
RpcSession<string>? rightSession = null;
var leftRouter = new RpcRouter<string>().On("left.double", LinkJson.Default.Number, LinkJson.Default.Number, (_, p, _) => ValueTask.FromResult(Result<Number>.Ok(new Number { Value = p.Value * 2 })));
var rightRouter = new RpcRouter<string>().On(
MethodTable table = TestTables.Of(TestTables.Req("left.double", RpcSide.Listener), TestTables.Req("right.relay"), TestTables.Req("right.unhandled"));
var leftRouter = new RpcRouter<string>(RpcSide.Dialer, table).On("left.double", LinkJson.Default.Number, LinkJson.Default.Number, (_, p, _) => ValueTask.FromResult(Result<Number>.Ok(new Number { Value = p.Value * 2 })));
var rightRouter = new RpcRouter<string>(RpcSide.Listener, table).On(
"right.relay", LinkJson.Default.Number, LinkJson.Default.Number,
async (_, p, ct) =>
{
@@ -248,7 +249,7 @@ public sealed class LinkTests
Result<Number> result = await leftSession.RequestAsync("right.relay", new Number { Value = 21 }, LinkJson.Default.Number, LinkJson.Default.Number, default);
Assert.Equal(43, result.Value.Value);
Result<JsonElement> unknown = await leftSession.RequestAsync("nope", null, default);
Result<JsonElement> unknown = await leftSession.RequestAsync("right.unhandled", null, default);
Assert.Equal("fleet_unsupported", unknown.Error.Code.Value);
await leftSession.CloseAsync("bye", default);
@@ -262,13 +263,13 @@ public sealed class LinkTests
{
(TlsFrameTransport left, TlsFrameTransport right) = await PairAsync();
await using var silent = right; // 对端不读也不回
await using var session = new RpcSession<string>(left, new RpcRouter<string>(), "x", TimeProvider.System, new RpcSessionOptions { RequestTimeout = TimeSpan.FromMilliseconds(200) });
await using var session = new RpcSession<string>(left, new RpcRouter<string>(RpcSide.Dialer, TestTables.Of(TestTables.Req("test.any"))), "x", TimeProvider.System, new RpcSessionOptions { RequestTimeout = TimeSpan.FromMilliseconds(200) });
Task run = session.RunAsync(default);
Result<JsonElement> timedOut = await session.RequestAsync("any", null, default);
Result<JsonElement> timedOut = await session.RequestAsync("test.any", null, default);
Assert.Equal(FleetErrors.Timeout, timedOut.Error.Code);
Task<Result<JsonElement>> pending = session.RequestAsync("any", null, default, TimeSpan.FromSeconds(30)).AsTask();
Task<Result<JsonElement>> pending = session.RequestAsync("test.any", null, default, TimeSpan.FromSeconds(30)).AsTask();
await silent.CloseAsync("gone", default);
Assert.Equal(FleetErrors.Timeout, (await pending.WaitAsync(TimeSpan.FromSeconds(5))).Error.Code);
await run.WaitAsync(TimeSpan.FromSeconds(5));
@@ -372,7 +373,8 @@ public sealed class LinkTests
public async Task PeerSession_CarriesRpcBothWays_AfterTheHandshake()
{
using NodeIdentity probe = NodeIdentity.Generate();
var router = new RpcRouter<string>().On("agent.echo", LinkJson.Default.Number, LinkJson.Default.Number, (_, p, _) => ValueTask.FromResult(Result<Number>.Ok(new Number { Value = p.Value + 1 })));
MethodTable table = TestTables.Of(TestTables.Req("agent.echo"));
var router = new RpcRouter<string>(RpcSide.Listener, table).On("agent.echo", LinkJson.Default.Number, LinkJson.Default.Number, (_, p, _) => ValueTask.FromResult(Result<Number>.Ok(new Number { Value = p.Value + 1 })));
await using var agent = new PeerFixture(
_ => probe.PublicKeySpki,
async (info, transport, ct) =>
@@ -383,7 +385,7 @@ public sealed class LinkTests
await agent.Listener.Started.Task;
(_, TlsFrameTransport transport) = await PeerConnector.ConnectAsync("127.0.0.1", agent.Listener.Port, agent.Spki, probe, "prb_1", null, 1 << 20, TimeProvider.System, default);
await using var client = new RpcSession<string>(transport, new RpcRouter<string>(), "probe", TimeProvider.System);
await using var client = new RpcSession<string>(transport, new RpcRouter<string>(RpcSide.Dialer, table), "probe", TimeProvider.System);
Task run = client.RunAsync(default);
Result<Number> result = await client.RequestAsync("agent.echo", new Number { Value = 41 }, LinkJson.Default.Number, LinkJson.Default.Number, default);
Assert.Equal(42, result.Value.Value);
@@ -0,0 +1,200 @@
using System.Text.Json;
using System.Threading.Channels;
using Fleet.Core;
using Fleet.Rpc;
namespace Fleet.Tests;
/// <summary>协议方向声明:路由器是唯一卡口,注册、分发、发送三处都核对声明表。</summary>
public sealed class MethodDirectionTests
{
private static readonly MethodTable s_table = TestTables.Of(
MethodSpec.Request("t.ping", RpcSide.Node),
MethodSpec.Notification("t.note", RpcSide.Node, "agent"),
MethodSpec.Request("t.push", RpcSide.Server),
MethodSpec.Request("t.inline", RpcSide.Node).AsInline(),
// 同一个方法名在对等面里另有发起方:按(平面,名字)区分
MethodSpec.Request("t.ping", RpcSide.Listener));
private static readonly JsonElement s_empty = JsonDocument.Parse("{}").RootElement.Clone();
private static RpcMessage Request(string method) => new() { Id = 1, Method = method, Params = s_empty };
private static RpcMessage Notification(string method) => new() { Method = method, Params = s_empty };
private static RpcRouter<string> Server(string? localRole = null) => new(RpcSide.Server, s_table, localRole, callerRole: role => role);
[Fact]
public void Registration_RefusesUndeclaredMethods_WrongDirection_WrongKind_AndWrongRole()
{
RpcRouter<string> server = Server();
Assert.Throws<InvalidOperationException>(() => server.On("t.unknown", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) => throw new NotSupportedException()));
// t.push 是服务端发给节点的:服务端这一侧不能处理它
Assert.Throws<InvalidOperationException>(() => server.On("t.push", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) => throw new NotSupportedException()));
// t.ping 声明成请求:不能按通知注册
Assert.Throws<InvalidOperationException>(() => server.OnNotify("t.ping", EchoContext.Default.AddParams, (_, _, _) => ValueTask.CompletedTask));
// 节点这一侧:t.note 只给 agent 角色,probe 注册不了;t.push 是它该处理的
var probe = new RpcRouter<string>(RpcSide.Node, s_table, "probe");
probe.On("t.push", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) => throw new NotSupportedException());
Assert.Throws<InvalidOperationException>(() => new RpcRouter<string>(RpcSide.Server, s_table, "probe").OnNotify("t.note", EchoContext.Default.AddParams, (_, _, _) => ValueTask.CompletedTask));
server.On("t.ping", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) => ValueTask.FromResult(Result<AddResult>.Ok(new AddResult())));
}
[Fact]
public async Task Dispatch_RefusesWhatThePeerMayNotInitiate_AndReportsAViolation()
{
int calls = 0;
RpcRouter<string> server = Server()
.On("t.ping", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) =>
{
calls++;
return ValueTask.FromResult(Result<AddResult>.Ok(new AddResult()));
})
.OnNotify("t.note", EchoContext.Default.AddParams, (_, _, _) =>
{
calls++;
return ValueTask.CompletedTask;
});
// 正常
RpcDispatch ok = await server.DispatchCheckedAsync("agent", Request("t.ping"), default);
Assert.Null(ok.Violation);
Assert.NotNull(ok.Response!.Result);
// 没有声明的方法
RpcDispatch undeclared = await server.DispatchCheckedAsync("agent", Request("t.nope"), default);
Assert.NotNull(undeclared.Violation);
Assert.Equal("fleet_protocol_violation", undeclared.Response!.Error!.Data!.Code);
// 节点冒充服务端:t.push 只能由服务端发起
RpcDispatch backwards = await server.DispatchCheckedAsync("agent", Request("t.push"), default);
Assert.NotNull(backwards.Violation);
Assert.Equal("fleet_protocol_violation", backwards.Response!.Error!.Data!.Code);
// 请求和通知用反了
Assert.NotNull((await server.DispatchCheckedAsync("agent", Notification("t.ping"), default)).Violation);
RpcDispatch asRequest = await server.DispatchCheckedAsync("agent", Request("t.note"), default);
Assert.NotNull(asRequest.Violation);
Assert.NotNull(asRequest.Response);
// 违规的消息到不了处理函数;通知违规不回响应
Assert.Equal(1, calls);
RpcDispatch silentViolation = await server.DispatchCheckedAsync("agent", Notification("t.nope"), default);
Assert.NotNull(silentViolation.Violation);
Assert.Null(silentViolation.Response);
}
[Fact]
public async Task Dispatch_ADeclaredMethodWithoutAHandler_IsUnsupported_NotAViolation()
{
RpcDispatch outcome = await Server().DispatchCheckedAsync("agent", Request("t.ping"), default);
Assert.Null(outcome.Violation);
Assert.Equal("fleet_unsupported", outcome.Response!.Error!.Data!.Code);
}
[Fact]
public async Task Dispatch_ChecksTheCallersRole()
{
RpcRouter<string> server = Server().OnNotify("t.note", EchoContext.Default.AddParams, (_, _, _) => ValueTask.CompletedTask);
Assert.Null((await server.DispatchCheckedAsync("agent", Notification("t.note"), default)).Violation);
Assert.NotNull((await server.DispatchCheckedAsync("probe", Notification("t.note"), default)).Violation);
Assert.NotNull((await server.DispatchCheckedAsync(null!, Notification("t.note"), default)).Violation);
}
[Fact]
public void CheckSend_RefusesWhatThisSideMayNotInitiate()
{
RpcRouter<string> server = Server();
server.CheckSend("t.push", MethodKind.Request);
Assert.Throws<InvalidOperationException>(() => server.CheckSend("t.ping", MethodKind.Request)); // 那是节点发起的
Assert.Throws<InvalidOperationException>(() => server.CheckSend("t.push", MethodKind.Notification)); // 类型不对
Assert.Throws<InvalidOperationException>(() => server.CheckSend("t.unknown", MethodKind.Request));
// 对端的角色也要允许
var node = new RpcRouter<string>(RpcSide.Node, s_table, "agent");
node.CheckSend("t.note", MethodKind.Notification, "agent");
Assert.Throws<InvalidOperationException>(() => node.CheckSend("t.note", MethodKind.Notification, "probe"));
}
[Fact]
public void MethodTable_RefusesDuplicates_AndReportsMissingHandlers()
{
Assert.Throws<ArgumentException>(() => TestTables.Of(TestTables.Req("a"), TestTables.Note("a")));
Assert.Throws<ArgumentException>(() => new MethodTable([], [new ChannelSpec(16, RpcSide.Node), new ChannelSpec(16, RpcSide.Server)]));
// 服务端要处理 t.ping 和 t.note;t.inline 在握手阶段直接处理,不算缺
Assert.Equal(["t.note", "t.ping"], s_table.MissingHandlers(RpcSide.Server, null, []));
Assert.Equal(["t.ping"], s_table.MissingHandlers(RpcSide.Server, "probe", []));
Assert.Equal(["t.note"], s_table.MissingHandlers(RpcSide.Server, "agent", ["t.ping"]));
Assert.Empty(s_table.MissingHandlers(RpcSide.Server, "agent", ["t.ping", "t.note"]));
// 同名方法在对等面的接收方是拨号方?不:t.ping 在对等面由监听方发起,拨号方处理
Assert.Equal(["t.ping"], s_table.MissingHandlers(RpcSide.Dialer, null, []));
}
[Fact]
public async Task Session_ClosesTheConnection_AfterTooManyViolations()
{
(ChannelTransport a, ChannelTransport b) = ChannelTransport.Pair();
RpcRouter<string> server = Server().On("t.ping", EchoContext.Default.AddParams, EchoContext.Default.AddResult, (_, _, _) => ValueTask.FromResult(Result<AddResult>.Ok(new AddResult())));
await using var session = new RpcSession<string>(b, server, "agent", TimeProvider.System, new RpcSessionOptions { MaxViolations = 3 });
Task run = session.RunAsync(default);
for (int i = 1; i <= 6; i++)
{
await a.SendAsync(new RpcFrame(FrameKind.Text, JsonSerializer.SerializeToUtf8Bytes(new RpcMessage { Id = i, Method = "t.nope", Params = s_empty }, FleetRpcContext.Default.RpcMessage)), default);
}
// 违规的请求先收到失败响应,超过上限之后连接被关闭
int failures = 0;
while (await a.ReceiveAsync(default).AsTask().WaitAsync(TimeSpan.FromSeconds(5)) is { } frame)
{
if (JsonSerializer.Deserialize(frame.Data.Span, FleetRpcContext.Default.RpcMessage)?.Error?.Data?.Code == "fleet_protocol_violation")
{
failures++;
}
}
Assert.InRange(failures, 3, 6);
await a.CloseAsync("done", default); // 对端确认关闭之后,会话的接收循环正常结束
await run.WaitAsync(TimeSpan.FromSeconds(5));
}
/// <summary>进程内的一对传输:一端发的帧从另一端收到;关闭时对端收到 null。</summary>
private sealed class ChannelTransport : IRpcTransport
{
private readonly Channel<RpcFrame?> _in;
private readonly Channel<RpcFrame?> _out;
private ChannelTransport(Channel<RpcFrame?> input, Channel<RpcFrame?> output)
{
_in = input;
_out = output;
}
public static (ChannelTransport A, ChannelTransport B) Pair()
{
Channel<RpcFrame?> ab = Channel.CreateUnbounded<RpcFrame?>();
Channel<RpcFrame?> ba = Channel.CreateUnbounded<RpcFrame?>();
return (new ChannelTransport(ba, ab), new ChannelTransport(ab, ba));
}
public ValueTask SendAsync(RpcFrame frame, CancellationToken ct)
{
_out.Writer.TryWrite(frame);
return ValueTask.CompletedTask;
}
public async ValueTask<RpcFrame?> ReceiveAsync(CancellationToken ct) =>
await _in.Reader.WaitToReadAsync(ct) && _in.Reader.TryRead(out RpcFrame? frame) ? frame : null;
public ValueTask CloseAsync(string reason, CancellationToken ct)
{
_out.Writer.TryWrite(null);
return ValueTask.CompletedTask;
}
public ValueTask DisposeAsync() => ValueTask.CompletedTask;
}
}
@@ -0,0 +1,266 @@
using Fleet.Core;
using Fleet.Hosting;
namespace Fleet.Tests;
public sealed class RestartScheduleTests
{
private static RestartSchedule Parse(string text)
{
Assert.True(RestartSchedule.TryParse(text, out RestartSchedule? schedule), text);
return schedule!;
}
[Theory]
[InlineData("04:00")]
[InlineData("04:00Z")]
[InlineData("04:00+08:00")]
[InlineData("23:59-05:30")]
[InlineData(" 00:00 ")]
public void Parse_AcceptsTimeWithOptionalOffset(string text) => Parse(text);
[Theory]
[InlineData("")]
[InlineData("4:00")]
[InlineData("24:00")]
[InlineData("04:60")]
[InlineData("04:00+8:00")]
[InlineData("04:00+15:00")]
[InlineData("04:00 UTC")]
[InlineData("0400")]
[InlineData("04:00+08:60")]
public void Parse_RejectsEverythingElse(string text) => Assert.False(RestartSchedule.TryParse(text, out _));
[Fact]
public void NextAfter_IsStrictlyLater_AndRollsToTomorrow()
{
RestartSchedule utc = Parse("04:00");
Assert.Equal(new DateTimeOffset(2026, 10, 7, 4, 0, 0, TimeSpan.Zero), utc.NextAfter(new DateTimeOffset(2026, 10, 7, 3, 59, 59, TimeSpan.Zero)));
Assert.Equal(new DateTimeOffset(2026, 10, 8, 4, 0, 0, TimeSpan.Zero), utc.NextAfter(new DateTimeOffset(2026, 10, 7, 4, 0, 0, TimeSpan.Zero)));
Assert.Equal(new DateTimeOffset(2026, 10, 8, 4, 0, 0, TimeSpan.Zero), utc.NextAfter(new DateTimeOffset(2026, 10, 7, 20, 0, 0, TimeSpan.Zero)));
}
[Fact]
public void NextAfter_UsesTheWrittenOffset()
{
// 04:00+08:00 就是 UTC 前一天 20:00
RestartSchedule china = Parse("04:00+08:00");
Assert.Equal(new DateTimeOffset(2026, 10, 6, 20, 0, 0, TimeSpan.Zero), china.NextAfter(new DateTimeOffset(2026, 10, 6, 19, 0, 0, TimeSpan.Zero)));
Assert.Equal(new DateTimeOffset(2026, 10, 7, 20, 0, 0, TimeSpan.Zero), china.NextAfter(new DateTimeOffset(2026, 10, 6, 20, 0, 0, TimeSpan.Zero)));
}
[Fact]
public void NextAfter_WithMinimumUptime_DoesNotRestartAgainRightAfterARestart()
{
RestartSchedule schedule = Parse("04:00");
// 04:00:03 重启后启动:加上最短运行时间再算,下一次是明天,而不是几秒后
var started = new DateTimeOffset(2026, 10, 7, 4, 0, 3, TimeSpan.Zero);
Assert.Equal(new DateTimeOffset(2026, 10, 8, 4, 0, 0, TimeSpan.Zero), schedule.NextAfter(started + RestartSchedule.MinimumUptime));
// 03:55 启动:离 04:00 只有 5 分钟,不够最短运行时间,顺延到明天
var early = new DateTimeOffset(2026, 10, 7, 3, 55, 0, TimeSpan.Zero);
Assert.Equal(new DateTimeOffset(2026, 10, 8, 4, 0, 0, TimeSpan.Zero), schedule.NextAfter(early + RestartSchedule.MinimumUptime));
}
[Fact]
public void Resolve_ReportsABadValueInsteadOfIgnoringIt()
{
Assert.Null(RestartSchedule.Resolve(null));
Assert.Null(RestartSchedule.Resolve(" "));
InvalidOperationException error = Assert.Throws<InvalidOperationException>(() => RestartSchedule.Resolve("tomorrow"));
Assert.Contains("restart_at", error.Message);
}
[Fact]
public async Task ServiceShell_StopsAtTheScheduledTime()
{
var time = new ImmediateTimeProvider(new DateTimeOffset(2026, 10, 7, 12, 0, 0, TimeSpan.Zero));
var log = new ListLog();
bool stoppedByToken = false;
// 业务循环只在被取消时返回;定时重启到点以后令牌被取消,RunAsync 正常返回
await ServiceShell.RunAsync(async ct =>
{
try
{
await Task.Delay(Timeout.Infinite, ct);
}
catch (OperationCanceledException)
{
stoppedByToken = true;
throw;
}
}, log, new RestartController(Parse("04:00")), time).WaitAsync(TimeSpan.FromSeconds(10));
Assert.True(stoppedByToken);
Assert.Contains(log.Lines, l => l.Contains("scheduled_restart", StringComparison.Ordinal));
Assert.Contains(log.Lines, l => l.Contains("restart scheduled", StringComparison.Ordinal));
}
[Fact]
public async Task ServiceShell_PicksUpAScheduleThePanelSendsLater()
{
var time = new ImmediateTimeProvider(new DateTimeOffset(2026, 10, 7, 12, 0, 0, TimeSpan.Zero));
var log = new ListLog();
using var controller = new RestartController();
Task shell = ServiceShell.RunAsync(ct => Task.Delay(Timeout.Infinite, ct), log, controller, time);
await Task.Delay(300);
Assert.False(shell.IsCompleted, "no schedule yet: the node keeps running");
Assert.True(controller.SetRemote("04:30+08:00"));
await shell.WaitAsync(TimeSpan.FromSeconds(10));
Assert.Contains(log.Lines, l => l.Contains("scheduled_restart", StringComparison.Ordinal));
Assert.Contains(log.Lines, l => l.Contains("source=panel", StringComparison.Ordinal));
}
[Fact]
public async Task ServiceShell_ClearingTheScheduleCancelsTheRestart()
{
var time = new ManualTimeProvider(new DateTimeOffset(2026, 10, 7, 12, 0, 0, TimeSpan.Zero));
var log = new ListLog();
using var controller = new RestartController();
controller.SetRemote("04:30+08:00");
Task shell = ServiceShell.RunAsync(ct => Task.Delay(Timeout.Infinite, ct), log, controller, time);
await Task.Delay(300);
controller.SetRemote(null); // 面板关掉了
await Task.Delay(300);
Assert.False(shell.IsCompleted);
Assert.Contains(log.Lines, l => l.Contains("restart scheduled", StringComparison.Ordinal));
Assert.DoesNotContain(log.Lines, l => l.Contains("scheduled_restart", StringComparison.Ordinal));
// 清理:把进程停掉
Assert.True(controller.SetRemote("04:30+08:00"));
// 新的计时器是异步注册的:反复推进,直到进程退出
for (int i = 0; i < 100 && !shell.IsCompleted; i++)
{
time.Advance();
await Task.Delay(50);
}
await shell.WaitAsync(TimeSpan.FromSeconds(10));
}
[Fact]
public void Controller_LocalSettingWinsOverThePanel_AndBadTextIsRejected()
{
using var controller = new RestartController(Parse("04:00"));
Assert.True(controller.IsLocal);
Assert.True(controller.SetRemote("05:00"));
Assert.Equal(new DateTimeOffset(2026, 10, 7, 4, 0, 0, TimeSpan.Zero), controller.Current!.NextAfter(new DateTimeOffset(2026, 10, 7, 0, 0, 0, TimeSpan.Zero)));
using var remoteOnly = new RestartController();
Assert.Null(remoteOnly.Current);
Assert.False(remoteOnly.SetRemote("whenever"));
Assert.Null(remoteOnly.Current);
Assert.True(remoteOnly.SetRemote("05:00"));
Assert.NotNull(remoteOnly.Current);
Assert.True(remoteOnly.SetRemote(null));
Assert.Null(remoteOnly.Current);
}
/// <summary>计时器马上触发:用来验证「到点」之后的行为,不用真的等。</summary>
private sealed class ImmediateTimeProvider : TimeProvider
{
private readonly DateTimeOffset _now;
public ImmediateTimeProvider(DateTimeOffset now) => _now = now;
public override DateTimeOffset GetUtcNow() => _now;
public override ITimer CreateTimer(TimerCallback callback, object? state, TimeSpan dueTime, TimeSpan period)
{
// TrimHeapAsync 这类每 30 秒一次的循环也会走到这里;只让「很长」的延迟(定时重启,至少 10 分钟)马上触发
var timer = new ImmediateTimer(callback, state);
if (dueTime >= RestartSchedule.MinimumUptime)
{
ThreadPool.QueueUserWorkItem(_ => timer.Fire());
}
return timer;
}
}
/// <summary>计时器只在 <see cref="Advance"/> 时触发:用来验证「还没到点」时不会退出。</summary>
private sealed class ManualTimeProvider : TimeProvider
{
private readonly DateTimeOffset _now;
private readonly List<ImmediateTimer> _timers = [];
public ManualTimeProvider(DateTimeOffset now) => _now = now;
public override DateTimeOffset GetUtcNow() => _now;
public override ITimer CreateTimer(TimerCallback callback, object? state, TimeSpan dueTime, TimeSpan period)
{
var timer = new ImmediateTimer(callback, state);
lock (_timers)
{
_timers.Add(timer);
}
return timer;
}
public void Advance()
{
ImmediateTimer[] due;
lock (_timers)
{
due = [.. _timers];
_timers.Clear();
}
foreach (ImmediateTimer timer in due)
{
timer.Fire();
}
}
}
private sealed class ImmediateTimer : ITimer
{
private readonly TimerCallback _callback;
private readonly object? _state;
public ImmediateTimer(TimerCallback callback, object? state)
{
_callback = callback;
_state = state;
}
public void Fire() => _callback(_state);
public bool Change(TimeSpan dueTime, TimeSpan period) => true;
public void Dispose()
{
}
public ValueTask DisposeAsync() => ValueTask.CompletedTask;
}
private sealed class ListLog : INodeLog
{
private readonly List<string> _lines = [];
public IReadOnlyList<string> Lines
{
get
{
lock (_lines)
{
return [.. _lines];
}
}
}
public void Write(NodeLogLevel level, string message, params (string Key, object? Value)[] fields)
{
lock (_lines)
{
_lines.Add(message + " " + string.Join(' ', fields.Select(f => f.Key + "=" + f.Value)));
}
}
}
}
@@ -0,0 +1,13 @@
using Fleet.Rpc;
namespace Fleet.Tests;
/// <summary>测试里用的方法声明:除非另有说明,方法都由拨号方发给监听方。</summary>
internal static class TestTables
{
public static MethodTable Of(params MethodSpec[] specs) => new(specs);
public static MethodSpec Req(string name, RpcSide from = RpcSide.Dialer) => MethodSpec.Request(name, from);
public static MethodSpec Note(string name, RpcSide from = RpcSide.Dialer) => MethodSpec.Notification(name, from);
}
+1 -1
View File
@@ -2,7 +2,7 @@ using System.Text.Json.Nodes;
using Fleet.Core;
using Fleet.Enrollment.Client;
// fleet-enroll --server https://panel --join jt_xxx --name jp-1 --identity-dir /etc/monitor [--meta '{"probe_tcp_port":7443,...}'] [--version 1.0]
// fleet-enroll --server https://panel --join jt_xxx --name jp-1 --identity-dir /etc/moshiro [--meta '{"probe_tcp_port":7443,...}'] [--version 1.0]
// 一次性运行:生成身份、注册、写入身份目录,然后退出。
string? Arg(string name) => args.SkipWhile(a => a != name).Skip(1).FirstOrDefault();
@@ -1,17 +1,36 @@
<Project>
<Import Project="Sdk.props" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.props" Sdk="StuDev.AotAnywhere"
Condition="'$(RuntimeIdentifier)' == 'linux-x64' Or '$(RuntimeIdentifier)' == 'linux-arm64' Or '$(RuntimeIdentifier)' == 'linux-arm' Or '$(RuntimeIdentifier)' == 'linux-musl-x64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm' Or '$(RuntimeIdentifier)' == 'osx-x64' Or '$(RuntimeIdentifier)' == 'osx-arm64' Or '$(RuntimeIdentifier)' == 'win-x64' Or '$(RuntimeIdentifier)' == 'win-arm64'" />
Condition="'$(UseAotAnywhere)' == 'true'" />
<PropertyGroup>
<OutputType>Exe</OutputType>
<AssemblyName>fleet-enroll</AssemblyName>
<RootNamespace>FleetEnroll</RootNamespace>
<IsAotCompatible>true</IsAotCompatible>
<PublishAot>true</PublishAot>
<PublishTrimmed>true</PublishTrimmed>
<TrimMode>full</TrimMode>
<StripSymbols>true</StripSymbols>
<AllowUnsafeBlocks>true</AllowUnsafeBlocks>
<MetadataUpdaterSupport>false</MetadataUpdaterSupport>
<OptimizationPreference>Size</OptimizationPreference>
<UseSystemResourceKeys>true</UseSystemResourceKeys>
<StackTraceSupport>false</StackTraceSupport>
<EventSourceSupport>false</EventSourceSupport>
<DebuggerSupport>false</DebuggerSupport>
<HttpActivityPropagationSupport>false</HttpActivityPropagationSupport>
<Http3Support>false</Http3Support>
<MetricsSupport>false</MetricsSupport>
<UseSizeOptimizedLinq>true</UseSizeOptimizedLinq>
<JsonSerializerIsReflectionEnabledByDefault>false</JsonSerializerIsReflectionEnabledByDefault>
<IlcFoldIdenticalMethodBodies>true</IlcFoldIdenticalMethodBodies>
<ServerGarbageCollection>false</ServerGarbageCollection>
<ConcurrentGarbageCollection>false</ConcurrentGarbageCollection>
</PropertyGroup>
<ItemGroup>
<ProjectReference Include="..\..\src\Fleet.Enrollment.Client\Fleet.Enrollment.Client.csproj" />
</ItemGroup>
<Import Project="Sdk.targets" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.targets" Sdk="StuDev.AotAnywhere"
Condition="'$(RuntimeIdentifier)' == 'linux-x64' Or '$(RuntimeIdentifier)' == 'linux-arm64' Or '$(RuntimeIdentifier)' == 'linux-arm' Or '$(RuntimeIdentifier)' == 'linux-musl-x64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm' Or '$(RuntimeIdentifier)' == 'osx-x64' Or '$(RuntimeIdentifier)' == 'osx-arm64' Or '$(RuntimeIdentifier)' == 'win-x64' Or '$(RuntimeIdentifier)' == 'win-arm64'" />
Condition="'$(UseAotAnywhere)' == 'true'" />
</Project>
+1322 -30
View File
File diff suppressed because it is too large. Load diff
+4
View File
@@ -8,6 +8,10 @@
<EventSourceSupport>false</EventSourceSupport>
<DebuggerSupport>false</DebuggerSupport>
<HttpActivityPropagationSupport>false</HttpActivityPropagationSupport>
<Http3Support>false</Http3Support>
<MetricsSupport>false</MetricsSupport>
<UseSizeOptimizedLinq>true</UseSizeOptimizedLinq>
<JsonSerializerIsReflectionEnabledByDefault>false</JsonSerializerIsReflectionEnabledByDefault>
<ServerGarbageCollection>false</ServerGarbageCollection>
<ConcurrentGarbageCollection>false</ConcurrentGarbageCollection>
</PropertyGroup>
+21 -14
View File
@@ -1,46 +1,53 @@
# Monitor Probe
# MoShiro Probe
Probe 是网络测量的发起端,通常运行在家宽或其他需要观测的网络出口。它连接 Agent、执行线路任务,并把本端采集的原始数据交给 Agent 汇总上报。Probe 不直接连接 Backend,也不提供远程命令或文件管理功能。
Probe 是网络测量的发起端,通常运行在家宽或其他需要观测的网络出口。它连接 Backend(身份、Agent 列表、状态)和 Agent(测量),执行线路任务,并把本端采集的原始数据交给 Agent 汇总上报。Probe 不提供远程命令或文件管理功能。
## 工作流程
```text
Backend ──任务与节点列表──> Agent <──测量会话──> Probe
│ │
└──汇总并上报数据 <───┘
Backend ──任务──> Agent <──测量会话──> Probe
│ ▲ │ │
│ └──汇总并上报数据 <───┘ │
└─────Agent 列表、状态(Fleet 长连接)───────────┘
```
Probe 使用面板生成的 `probe.json` 读取自身身份、首次连接用的 Agent 地址和本地测试限制。成功连接后,Agent 会同步可用节点列表;Probe 将列表缓存到同目录的 `agents.json`。`probe.json` 含私钥,请限制文件访问权限,不要提交到版本控制。
Probe 使用面板生成的 `probe.json` 读取自身身份、面板地址(`server`)和本地测试限制。连上面板之后,面板下发要连接的 Agent 列表;Probe 将列表缓存到同目录的 `agents.json`,面板暂时连不上时靠它继续测量。`probe.json` 含私钥,请限制文件访问权限,不要提交到版本控制。
容器节点的运行方式(podman + systemd)和宿主机更新器见 [`../../docs/deployment.md`](../../docs/deployment.md);Probe 也可以用 OCI 镜像 `registry.oeo.one/moshiro/moshiro-probe:<version>`,容器默认读取 `/etc/moshiro/probe.json`,命名 volume 保存身份配置和缓存,镜像以 UID/GID 1654 运行。
## 构建
需要 .NET SDK 10。为应用 `backend/global.json` 的 SDK 设置,请从 `backend/` 目录构建:
需要 .NET SDK 10。为应用 根目录的 `global.json` 的 SDK 设置,请从 `backend/` 目录构建:
```sh
cd backend
dotnet build probe/src/Monitor.Probe -c Release
dotnet build probe/src/MoShiro.Probe -c Release
```
发布 NativeAOT 程序需要对应平台的 C/C++ 工具链;`<RID>` 应与构建环境和目标平台匹配:
```sh
dotnet publish probe/src/Monitor.Probe -c Release -r <RID> -p:PublishAot=true
dotnet publish probe/src/MoShiro.Probe -c Release -r <RID> -p:PublishAot=true
```
也可在仓库根目录运行 `bash scripts/publish.sh probe <RID>`,再使用 `scripts/package.sh probe <RID>` 生成部署包。支持平台和安装流程见 [`../../docs/deployment.md`](../../docs/deployment.md)。
## 配置与运行
在管理端创建 Probe 并下载一次性身份配置,将文件保存为目标机上的 `/etc/monitor/probe.json`。随后运行:
在管理端创建 Probe 并下载一次性身份配置,将文件保存为目标机上的 `/etc/moshiro/probe.json`。随后运行:
```sh
monitor-probe --config /etc/monitor/probe.json
moshiro-probe --config /etc/moshiro/probe.json
```
不传 `--config` 时默认读取 `/etc/monitor/probe.json`。配置中的 `limits` 是本地硬上限,Backend 无法远程放宽;路由器等低性能设备可调低测试码率和包速率,避免 Probe 自身成为测量瓶颈。部署时可使用仓库的 `packaging/` 文件和安装脚本。
不传 `--config` 时默认读取 `/etc/moshiro/probe.json`。配置中的 `limits` 是本地硬上限,Backend 无法远程放宽;路由器等低性能设备可调低测试码率和包速率,避免 Probe 自身成为测量瓶颈。部署时可使用仓库的 `packaging/` 文件和安装脚本。
也可使用 OCI 镜像 `registry.oeo.one/moshiro/moshiro-probe:<version>`,自动选择 Linux amd64 或 arm64 镜像。容器默认读取 `/etc/moshiro/probe.json`,并在同一目录维护 `agents.json` 缓存,配置与缓存存放在命名 volume 中。建议使用宿主网络;当前独立 ICMP 目标任务由 Agent 执行。
## 设计说明
- Probe 只执行编译进程序的测量模块。它不认识 Backend,不落盘测量数据,也不能执行命令或自行更新。
- 测量模块由 `backend/shared/Monitor.Measurement/` 与 Agent 共用;Probe 负责发起连接、调度任务和记录本端数据。
- Probe 只执行编译进程序的测量模块。它不落盘测量数据,也不能执行命令或自行更新(更新由宿主机上的更新器完成,Probe 只写清单和请求文件)。
- 测量模块由 `backend/shared/MoShiro.Measurement/` 与 Agent 共用;Probe 负责发起连接、调度任务和记录本端数据。
- 测量模型和协议细节见 [`../../docs/measurement.md`](../../docs/measurement.md) 与 [`../../docs/protocol.md`](../../docs/protocol.md);整体架构见 [`../../docs/architecture.md`](../../docs/architecture.md)。
Docker 与 rootless/rootful Podman 共用 `packaging/compose/probe/compose.yaml`,配置、身份和缓存放在命名 volume 中,无需宿主配置目录、keep-id 或手动 chown。首次导入和自动更新见部署文档。
@@ -6,16 +6,16 @@ using Fleet.Core;
using Fleet.Identity;
using Fleet.Link;
using Fleet.Rpc;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Probe;
namespace MoShiro.Probe;
/// <summary>所有连接共用的东西。</summary>
public sealed record ConnectionEnv(NodeIdentity Identity, string ProbeId, ProbeLimits Limits, AgentDirectory Directory, MonoClock Clock, TimeProvider Time, INodeLog Log);
/// <summary>
/// 与一个 Agent 的点对点会话:认证握手 → 同步时钟 → 取 Agent 列表 → 按 Agent 下发的 <c>line_task_set</c> 运行角色 →
/// 与一个 Agent 的点对点会话:认证握手 → 同步时钟 → 按 Agent 下发的 <c>line_task_set</c> 运行角色 →
/// 每 10 秒把已完成的行切成列块放进发件箱 → 响应按需测试的阶段。会话断开时所有角色停止(凭据属于会话),按退避重连。
/// 会话内的所有状态(套接字、发件箱、角色)随会话一起创建和释放。
/// </summary>
@@ -25,16 +25,19 @@ public sealed class AgentConnection
private static readonly TimeSpan s_clockInterval = TimeSpan.FromSeconds(30);
private static readonly RpcRouter<ConnectionSession> s_router = BuildRouter();
/// <summary>对等面入站方法的路由器(只读,测试用)。</summary>
public static RpcRouter<ConnectionSession> Router => s_router;
private readonly ConnectionEnv _env;
private volatile ConnectionSession? _session;
public AgentConnection(BootstrapAgent endpoint, ConnectionEnv env)
public AgentConnection(KnownAgent endpoint, ConnectionEnv env)
{
Endpoint = endpoint;
_env = env;
}
public BootstrapAgent Endpoint { get; }
public KnownAgent Endpoint { get; }
public bool IsConnected => _session is not null;
@@ -74,8 +77,20 @@ public sealed class AgentConnection
private async Task<string> RunOnceAsync(Backoff backoff, CancellationToken ct)
{
AddressFamily family = Endpoint.IpVersion == IpVersions.V6 ? AddressFamily.InterNetworkV6 : AddressFamily.InterNetwork;
using var resolveTimeout = new CancellationTokenSource(TimeSpan.FromSeconds(10), _env.Time);
using CancellationTokenSource resolveCt = CancellationTokenSource.CreateLinkedTokenSource(ct, resolveTimeout.Token);
IPAddress[] addresses = IPAddress.TryParse(Endpoint.Host, out IPAddress? literal)
? [literal]
: await Dns.GetHostAddressesAsync(Endpoint.Host, family, resolveCt.Token);
IPAddress? address = addresses.FirstOrDefault(a => a.AddressFamily == family && !a.IsIPv4MappedToIPv6);
if (address is null)
{
throw new SocketException((int)SocketError.HostNotFound);
}
(PeerSessionInfo info, TlsFrameTransport transport) = await PeerConnector.ConnectAsync(
Endpoint.Host, Endpoint.TcpPort, Convert.FromBase64String(Endpoint.Spki), _env.Identity, _env.ProbeId, ProbeApp.Version, 8 << 20, _env.Time, ct);
address.ToString(), Endpoint.TcpPort, Convert.FromBase64String(Endpoint.Spki), _env.Identity, _env.ProbeId, ProbeApp.Version, 8 << 20, _env.Time, ct);
var remote = (IPEndPoint)info.Remote!;
var data = new DataSession(info.Credentials.SessionIdValue, info.Credentials.Key);
using var udp = new UdpMux(new SingleSession(data), null, _env.Clock, new DataPlaneCounters(), _env.Log, remote.AddressFamily == AddressFamily.InterNetworkV6 ? IPAddress.IPv6Any : IPAddress.Any, []);
@@ -105,12 +120,11 @@ public sealed class AgentConnection
Task pump = PumpAsync(rpc, outbox, cts.Token);
try
{
// 先同步时钟(块头里要带偏移),再取 Agent 列表;Agent 随后推送线路任务集
// 先同步时钟(块头里要带偏移);Agent 随后推送线路任务集
await SyncClockAsync(rpc, clock, rounds: 5, cts.Token);
backoff.Reset();
_session = session;
_env.Log.Info("agent session up", ("agent", Endpoint.AgentId), ("offset_us", clock.OffsetUs));
await RefreshDirectoryAsync(rpc, cts.Token);
Task maintenance = MaintainAsync(rpc, session, clock, cts.Token);
Task finished = await Task.WhenAny(run, receive, pump, maintenance);
await cts.CancelAsync();
@@ -155,7 +169,7 @@ public sealed class AgentConnection
for (int i = 0; i < rounds; i++)
{
long sent = _env.Clock.NowUs();
Result<ClockResult> reply = await rpc.RequestAsync(MonitorMethods.Clock, new ClockParams { ProbeMonoUs = sent }, MonitorJsonContext.Default.ClockParams, MonitorJsonContext.Default.ClockResult, ct);
Result<ClockResult> reply = await rpc.RequestAsync(MoShiroMethods.Clock, new ClockParams { ProbeMonoUs = sent }, MoShiroJsonContext.Default.ClockParams, MoShiroJsonContext.Default.ClockResult, ct);
if (reply.TryGet(out ClockResult? value, out _))
{
clock.Add(sent, value.AgentMonoUs, _env.Clock.NowUs());
@@ -163,20 +177,6 @@ public sealed class AgentConnection
}
}
private async Task RefreshDirectoryAsync(RpcSession<ConnectionSession> rpc, CancellationToken ct)
{
Result<ProbeAgentsResult> list = await rpc.RequestAsync(
MonitorMethods.ProbeAgents, new ProbeAgentsParams { HaveVersion = _env.Directory.Version }, MonitorJsonContext.Default.ProbeAgentsParams, MonitorJsonContext.Default.ProbeAgentsResult, ct);
if (list.TryGet(out ProbeAgentsResult? agents, out FleetError? error))
{
_env.Directory.Apply(agents);
}
else
{
_env.Log.Warn("could not fetch the agent list", ("agent", Endpoint.AgentId), ("err", error.Code));
}
}
/// <summary>切块(每 10 秒)和时钟同步(每 30 秒)。</summary>
private async Task MaintainAsync(RpcSession<ConnectionSession> rpc, ConnectionSession session, ClockEstimator clock, CancellationToken ct)
{
@@ -194,7 +194,7 @@ public sealed class AgentConnection
}
private static RpcRouter<ConnectionSession> BuildRouter() =>
new RpcRouter<ConnectionSession>()
new RpcRouter<ConnectionSession>(RpcSide.Dialer, MoShiroMethodTable.Table)
.On(
FleetMethods.StateSync,
FleetRpcContext.Default.StateSyncParams,
@@ -204,7 +204,7 @@ public sealed class AgentConnection
var result = new StateSyncResult { AppliedVersion = sync.Version };
if (sync.Type == LineTaskSet.StateType)
{
LineTaskSet? set = sync.Doc.Deserialize(MonitorJsonContext.Default.LineTaskSet);
LineTaskSet? set = sync.Doc.Deserialize(MoShiroJsonContext.Default.LineTaskSet);
if (set is null)
{
return Result<StateSyncResult>.Fail(FleetErrors.InvalidParams, "line_task_set could not be parsed");
@@ -216,27 +216,19 @@ public sealed class AgentConnection
return Result<StateSyncResult>.Ok(result);
})
.On(
MonitorMethods.TestPhase,
MonitorJsonContext.Default.TestPhaseStartParams,
MonitorJsonContext.Default.TestPhaseStartResult,
MoShiroMethods.TestPhase,
MoShiroJsonContext.Default.TestPhaseStartParams,
MoShiroJsonContext.Default.TestPhaseStartResult,
(session, start, _) => ValueTask.FromResult(Result<TestPhaseStartResult>.Ok(new TestPhaseStartResult { Accepted = session.StartPhase(start) })))
.On(
MonitorMethods.TestPhaseEnd,
MonitorJsonContext.Default.TestPhaseEndParams,
MonitorJsonContext.Default.TestPhaseEndResult,
MoShiroMethods.TestPhaseEnd,
MoShiroJsonContext.Default.TestPhaseEndParams,
MoShiroJsonContext.Default.TestPhaseEndResult,
async (session, end, ct) => Result<TestPhaseEndResult>.Ok(new TestPhaseEndResult { Fragments = await session.EndPhaseAsync(end, ct) }))
.OnNotify(
MonitorMethods.TestAbort,
MonitorJsonContext.Default.TestAbortParams,
MoShiroMethods.TestAbort,
MoShiroJsonContext.Default.TestAbortParams,
async (session, abort, _) => await session.AbortAsync(abort.TestId))
.OnNotify(
MonitorMethods.ProbeAgentsChanged,
MonitorJsonContext.Default.ProbeAgentsResult,
(session, list, _) =>
{
session.Env.Directory.Apply(list);
return ValueTask.CompletedTask;
})
.OnNotify(
FleetMethods.OutboxAck,
FleetRpcContext.Default.OutboxAckParams,
@@ -1,13 +1,12 @@
using System.Text.Json;
using Fleet.Core;
using Fleet.Hosting;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Probe;
namespace MoShiro.Probe;
/// <summary>
/// Probe 要连接的 Agent 列表。Probe 不认识 Backend:列表先取自 <c>probe.json</c> 的 bootstrap(或上一次保存的 <c>agents.json</c>),
/// 连上任意一个 Agent 之后,由它代为向 Backend 查询(<c>monitor.probe.agents</c>),之后的变化由 Agent 推送(<c>agents_changed</c>)。
/// Probe 要连接的 Agent 列表:由面板作为期望状态(<c>probe_agents</c>)直接下发;面板连不上时用上一次保存的 <c>agents.json</c>。
/// 列表变化时用 <see cref="ConfigStore"/> 写回 <c>agents.json</c>(这是配置,不是测量数据;只在变化时写,保护路由器的闪存)。
/// </summary>
public sealed class AgentDirectory
@@ -17,29 +16,25 @@ public sealed class AgentDirectory
private readonly object _gate = new();
private readonly ConfigStore _store;
private readonly INodeLog _log;
private List<BootstrapAgent> _agents;
private List<KnownAgent> _agents = [];
private long _version;
public AgentDirectory(ConfigStore store, string configDirectory, ProbeFile file, INodeLog log)
public AgentDirectory(ConfigStore store, string configDirectory, INodeLog log)
{
_store = store;
_log = log;
_agents = [.. file.Bootstrap];
string cache = Path.Combine(configDirectory, FileName);
if (File.Exists(cache))
{
try
{
AgentsCache saved = NodeConfig.Load(cache, ProbeConfigContext.Default.AgentsCache);
if (saved.Agents.Count > 0)
{
_agents = saved.Agents;
_version = saved.Version;
}
_agents = saved.Agents;
_version = saved.Version;
}
catch (Exception ex) when (ex is IOException or InvalidDataException or JsonException or InvalidOperationException)
{
_log.Warn("agents.json is unreadable, using the bootstrap list", ("err", ex.Message));
_log.Warn("agents.json is unreadable, waiting for the panel", ("err", ex.Message));
}
}
}
@@ -58,7 +53,7 @@ public sealed class AgentDirectory
}
}
public IReadOnlyList<BootstrapAgent> Agents
public IReadOnlyList<KnownAgent> Agents
{
get
{
@@ -69,10 +64,22 @@ public sealed class AgentDirectory
}
}
/// <summary>应用 Agent 转来的列表。版本不比当前新的忽略(乱序到达的旧推送)。</summary>
public void Apply(ProbeAgentsResult result)
/// <summary>应用面板下发的列表。版本不比当前新的忽略(乱序到达的旧推送)。</summary>
public void Apply(ProbeAgentList result)
{
var next = result.Agents.Select(a => new BootstrapAgent { AgentId = a.AgentId.Value, Host = a.Host, TcpPort = a.TcpPort, Spki = a.Spki }).ToList();
var next = new List<KnownAgent>();
foreach (AgentEndpoint agent in result.Agents)
{
if (!string.IsNullOrWhiteSpace(agent.Host))
{
next.Add(new KnownAgent { AgentId = agent.AgentId.Value, Host = agent.Host, TcpPort = agent.TcpPort, Spki = agent.Spki });
}
if (!string.IsNullOrWhiteSpace(agent.Ipv6Host))
{
next.Add(new KnownAgent { AgentId = agent.AgentId.Value, Host = agent.Ipv6Host, TcpPort = agent.TcpPort, Spki = agent.Spki, IpVersion = IpVersions.V6 });
}
}
lock (_gate)
{
if (result.Version < _version || (result.Version == _version && Same(next, _agents)))
@@ -97,9 +104,9 @@ public sealed class AgentDirectory
Changed?.Invoke();
}
private static bool Same(List<BootstrapAgent> a, List<BootstrapAgent> b) =>
private static bool Same(List<KnownAgent> a, List<KnownAgent> b) =>
a.Count == b.Count && a.Zip(b).All(p => Equal(p.First, p.Second));
public static bool Equal(BootstrapAgent a, BootstrapAgent b) =>
a.AgentId == b.AgentId && a.Host == b.Host && a.TcpPort == b.TcpPort && a.Spki == b.Spki;
public static bool Equal(KnownAgent a, KnownAgent b) =>
a.AgentId == b.AgentId && a.IpVersion == b.IpVersion && a.Host == b.Host && a.TcpPort == b.TcpPort && a.Spki == b.Spki;
}
@@ -1,13 +1,13 @@
using System.Text.Json;
using Fleet.Core;
using Fleet.Link;
using Monitor.Measurement;
using Monitor.Protocol;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace Monitor.Probe;
namespace MoShiro.Probe;
/// <summary>
/// 估计 Agent 单调时钟 − Probe 单调时钟:每次 <c>monitor.clock</c> 往返取中点,在最近 8 次里取往返最短的一次
/// 估计 Agent 单调时钟 − Probe 单调时钟:每次 <c>moshiro.clock</c> 往返取中点,在最近 8 次里取往返最短的一次
/// (往返越短,对称性假设的误差越小)。Probe 的列块头里带着这个偏移,Backend 据此对齐两端的时间戳。
/// </summary>
public sealed class ClockEstimator
@@ -82,6 +82,6 @@ public sealed class FragmentShipper : IFragmentSink
}
}
_outbox.Enqueue(MonitorChannels.ColumnBlock, header.ChunkId.Value, RawChunkCodec.Encode(new RawChunk(header, streams)));
_outbox.Enqueue(MoShiroChannels.ColumnBlock, header.ChunkId.Value, RawChunkCodec.Encode(new RawChunk(header, streams)));
}
}
@@ -0,0 +1,26 @@
<Project>
<Import Project="Sdk.props" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.props" Sdk="StuDev.AotAnywhere"
Condition="'$(UseAotAnywhere)' == 'true'" />
<PropertyGroup>
<OutputType>Exe</OutputType>
<AssemblyName>moshiro-probe</AssemblyName>
<RootNamespace>MoShiro.Probe</RootNamespace>
<IsAotCompatible>true</IsAotCompatible>
<PublishAot>true</PublishAot>
<PublishTrimmed>true</PublishTrimmed>
<TrimMode>full</TrimMode>
<StripSymbols>true</StripSymbols>
<AllowUnsafeBlocks>true</AllowUnsafeBlocks>
<MetadataUpdaterSupport>false</MetadataUpdaterSupport>
</PropertyGroup>
<ItemGroup>
<InternalsVisibleTo Include="MoShiro.Backend.Tests" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Link\Fleet.Link.csproj" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Hosting\Fleet.Hosting.csproj" />
<ProjectReference Include="..\..\..\shared\MoShiro.Measurement\MoShiro.Measurement.csproj" />
</ItemGroup>
<Import Project="Sdk.targets" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.targets" Sdk="StuDev.AotAnywhere"
Condition="'$(UseAotAnywhere)' == 'true'" />
</Project>
+205
View File
@@ -0,0 +1,205 @@
using Fleet.Core;
using Fleet.Hosting;
using System.Text.Json;
using Fleet.Identity;
using Fleet.Link;
using Fleet.Rpc;
using MoShiro.Measurement;
using MoShiro.Protocol;
namespace MoShiro.Probe;
/// <summary>
/// Probe 的组装点:到面板的长连接(身份、Agent 列表、状态)+ 每个 Agent 一个测量连接(认证、重连、角色)。没有全局状态,测试可以在同一进程里启动多个。
/// Agent 列表变化时只调和连接:新增的建立,移除的断开,地址或证书指纹变化的重建,其余不受影响。
/// </summary>
public sealed class ProbeApp : IDisposable
{
public const string Version = MoShiro.Protocol.BuildInfo.Version;
private readonly ConnectionEnv _env;
private readonly Dictionary<string, Running> _connections = [];
private readonly SemaphoreSlim _changed = new(0);
private readonly INodeLog _log;
private readonly FleetLink _link;
public ProbeApp(ProbeFile file, string configDirectory, TimeProvider time, INodeLog log, RestartController? restart = null)
{
_log = log;
var directory = new AgentDirectory(new ConfigStore(configDirectory), configDirectory, log);
NodeIdentity identity = NodeIdentity.FromPkcs8(Convert.FromBase64String(file.PrivateKey));
_env = new ConnectionEnv(identity, file.ProbeId, file.Limits, directory, new MonoClock(time), time, log);
directory.Changed += () => _changed.Release();
var update = new UpdateBridge(configDirectory, new ConfigStore(configDirectory).Write, log);
var options = new FleetLinkOptions
{
Server = new Uri(file.Server),
Identity = identity,
NodeId = new NodeId(file.ProbeId),
NodeVersion = Version,
Capabilities = [MoShiroCapabilities.Base],
HeartbeatExtra = () => JsonSerializer.SerializeToElement(new ProbeHeartbeatExtra { Update = update.ReadStatus(), Ipv6 = Ipv6Diagnostics.Sample(null, Environment.GetEnvironmentVariable("MOSHIRO_NETWORK_SCOPE"), Environment.GetEnvironmentVariable("HOST_PROC")) }, MoShiroJsonContext.Default.ProbeHeartbeatExtra),
HelloMeta = () => JsonSerializer.SerializeToElement(new ProbeHelloMeta { Platform = PlatformDetector.Detect(), Ipv6 = Ipv6Diagnostics.Sample(null, Environment.GetEnvironmentVariable("MOSHIRO_NETWORK_SCOPE"), Environment.GetEnvironmentVariable("HOST_PROC")) }, MoShiroJsonContext.Default.ProbeHelloMeta),
};
var list = new StateHandler<ProbeAgentList>(
ProbeAgentList.StateType,
MoShiroJsonContext.Default.ProbeAgentList,
(version, doc, _) =>
{
directory.Apply(doc);
return ValueTask.FromResult(new StateSyncResult { AppliedVersion = version });
});
var router = new RpcRouter<LinkContext>(RpcSide.Node, MoShiroMethodTable.Table, NodeRoles.Probe)
.OnNotify(MoShiroMethods.UpdateOffer, MoShiroJsonContext.Default.UpdateOfferParams, (_, offer, _) =>
{
update.WriteOffer(offer);
return ValueTask.CompletedTask;
})
.OnNotify(MoShiroMethods.UpdateApply, MoShiroJsonContext.Default.UpdateOfferParams, (_, _, _) =>
{
update.RequestApply();
return ValueTask.CompletedTask;
})
.OnNotify(MoShiroMethods.Schedule, MoShiroJsonContext.Default.ScheduleParams, (_, schedule, _) =>
{
if (restart is not null && !restart.SetRemote(schedule.RestartAt))
{
log.Warn("ignoring an invalid restart schedule from the panel");
}
return ValueTask.CompletedTask;
});
_link = new FleetLink(options, router, [list], new Outbox(1 << 20), time, log);
}
/// <summary>到面板的连接(测试和状态查看用)。</summary>
public FleetLink Link => _link;
public AgentDirectory Directory => _env.Directory;
/// <summary>某个 Agent 的会话是否已经建立(测试和状态查看用)。</summary>
public bool IsConnected(string agentId, string ipVersion = IpVersions.V4)
{
lock (_connections)
{
return _connections.TryGetValue(agentId + "~" + ipVersion, out Running? running) && running.Connection.IsConnected;
}
}
public AgentConnection? Connection(string agentId, string ipVersion = IpVersions.V4)
{
lock (_connections)
{
return _connections.TryGetValue(agentId + "~" + ipVersion, out Running? running) ? running.Connection : null;
}
}
public async Task RunAsync(CancellationToken ct)
{
if (_env.Directory.Agents.Count == 0)
{
_log.Info("no agents known yet, waiting for the panel to send the list");
}
Task link = _link.RunAsync(ct);
try
{
while (!ct.IsCancellationRequested)
{
await ReconcileAsync(_env.Directory.Agents, ct);
await _changed.WaitAsync(ct);
}
}
catch (OperationCanceledException)
{
// 停止
}
finally
{
await ReconcileAsync([], CancellationToken.None);
await Task.WhenAny(link, Task.Delay(TimeSpan.FromSeconds(5), CancellationToken.None));
}
}
public void Dispose()
{
_env.Identity.Dispose();
_changed.Dispose();
}
private async Task ReconcileAsync(IReadOnlyList<KnownAgent> agents, CancellationToken ct)
{
var wanted = agents.ToDictionary(a => a.PathKey);
var stopped = new List<Running>();
lock (_connections)
{
foreach ((string id, Running running) in _connections.ToList())
{
if (!wanted.TryGetValue(id, out KnownAgent? endpoint) || !AgentDirectory.Equal(endpoint, running.Connection.Endpoint))
{
stopped.Add(running);
_connections.Remove(id);
}
}
}
foreach (Running running in stopped)
{
await running.StopAsync();
}
if (ct.IsCancellationRequested)
{
return;
}
lock (_connections)
{
foreach ((string id, KnownAgent endpoint) in wanted)
{
if (!_connections.ContainsKey(id))
{
_connections[id] = Running.Start(new AgentConnection(endpoint, _env));
}
}
}
}
private sealed class Running : IDisposable
{
private readonly CancellationTokenSource _cts = new();
private Task _run = Task.CompletedTask;
private Running(AgentConnection connection)
{
Connection = connection;
}
public AgentConnection Connection { get; }
public static Running Start(AgentConnection connection)
{
var running = new Running(connection);
running._run = Task.Run(() => connection.RunAsync(running._cts.Token), CancellationToken.None);
return running;
}
public async Task StopAsync()
{
await _cts.CancelAsync();
try
{
await _run.WaitAsync(TimeSpan.FromSeconds(10));
}
catch (Exception ex) when (ex is OperationCanceledException or TimeoutException)
{
// 没有及时退出:不再等
}
Dispose();
}
public void Dispose() => _cts.Dispose();
}
}
@@ -1,6 +1,7 @@
using System.Text.Json.Serialization;
using MoShiro.Protocol;
namespace Monitor.Probe;
namespace MoShiro.Probe;
/// <summary>
/// <c>probe.json</c>:面板 <c>POST /api/admin/probe/identity</c> 生成的文件(只能下载一次)。
@@ -15,16 +16,26 @@ public sealed class ProbeFile
[JsonPropertyName("private_key")]
public string PrivateKey { get; set; } = "";
/// <summary>首次连接用的 Agent;连上之后以 <c>agents.json</c>(最新的列表)为准。</summary>
[JsonPropertyName("bootstrap")]
public List<BootstrapAgent> Bootstrap { get; set; } = [];
/// <summary>面板的地址(http 或 https)。Probe 直接连它:身份、Agent 列表、更新都从这里来。</summary>
[JsonPropertyName("server")]
public string Server { get; set; } = "";
/// <summary>每天定时重启一次,例如 <c>04:00+08:00</c>(不写偏移按 UTC);不写就不重启。环境变量 <c>MOSHIRO_RESTART_AT</c> 优先;这里是本地配置,从面板重新下载 probe.json 会丢掉它,所以容器里推荐用环境变量。</summary>
[JsonPropertyName("restart_at")]
public string? RestartAt { get; set; }
[JsonPropertyName("limits")]
public ProbeLimits Limits { get; set; } = new();
}
public sealed class BootstrapAgent
public sealed class KnownAgent
{
[JsonPropertyName("ip_version")]
public string IpVersion { get; set; } = IpVersions.V4;
[JsonIgnore]
public string PathKey => AgentId + "~" + IpVersion;
[JsonPropertyName("agent_id")]
public string AgentId { get; set; } = "";
@@ -54,12 +65,12 @@ public sealed class ProbeLimits
[JsonSerializable(typeof(AgentsCache))]
public sealed partial class ProbeConfigContext : JsonSerializerContext;
/// <summary><c>agents.json</c>:最近一次从 Agent 同步到的列表(配置,不是测量数据)。</summary>
/// <summary><c>agents.json</c>:最近一次从面板同步到的 Agent 列表(配置,不是测量数据);面板连不上时 Probe 靠它继续测量。</summary>
public sealed class AgentsCache
{
[JsonPropertyName("version")]
public long Version { get; set; }
[JsonPropertyName("agents")]
public List<BootstrapAgent> Agents { get; set; } = [];
public List<KnownAgent> Agents { get; set; } = [];
}
@@ -1,6 +1,6 @@
using Fleet.Core;
using Fleet.Hosting;
using Monitor.Probe;
using MoShiro.Probe;
TimeProvider time = TimeProvider.System;
INodeLog log = new ConsoleNodeLog(time);
@@ -11,7 +11,7 @@ if (PlatformCheck.VerifyCrypto() is { } problem)
return 1;
}
string configPath = "/etc/monitor/probe.json";
string configPath = "/etc/moshiro/probe.json";
for (int i = 0; i < args.Length - 1; i++)
{
if (args[i] == "--config")
@@ -21,12 +21,14 @@ for (int i = 0; i < args.Length - 1; i++)
}
ProbeFile file;
RestartController restart;
try
{
file = NodeConfig.Load(configPath, ProbeConfigContext.Default.ProbeFile);
if (file.ProbeId.Length == 0 || file.PrivateKey.Length == 0)
restart = new RestartController(RestartSchedule.Resolve(file.RestartAt));
if (file.ProbeId.Length == 0 || file.PrivateKey.Length == 0 || !Uri.TryCreate(file.Server, UriKind.Absolute, out _))
{
throw new InvalidOperationException("probe.json needs probe_id and private_key (download it from the panel: POST /api/admin/probe/identity)");
throw new InvalidOperationException("probe.json needs probe_id, private_key and server (download it from the panel: Probes → Create)");
}
}
catch (Exception ex) when (ex is IOException or InvalidDataException or InvalidOperationException or System.Text.Json.JsonException)
@@ -35,11 +37,12 @@ catch (Exception ex) when (ex is IOException or InvalidDataException or InvalidO
return 1;
}
using var restartScope = restart;
string directory = Path.GetDirectoryName(Path.GetFullPath(configPath))!;
try
{
using var app = new ProbeApp(file, directory, time, log);
await ServiceShell.RunAsync(app.RunAsync, log);
using var app = new ProbeApp(file, directory, time, log, restart);
await ServiceShell.RunAsync(app.RunAsync, log, restart, time);
}
catch (Exception ex) when (ex is FormatException or System.Security.Cryptography.CryptographicException)
{
@@ -1,20 +0,0 @@
<Project>
<Import Project="Sdk.props" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.props" Sdk="StuDev.AotAnywhere"
Condition="'$(RuntimeIdentifier)' == 'linux-x64' Or '$(RuntimeIdentifier)' == 'linux-arm64' Or '$(RuntimeIdentifier)' == 'linux-arm' Or '$(RuntimeIdentifier)' == 'linux-musl-x64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm' Or '$(RuntimeIdentifier)' == 'osx-x64' Or '$(RuntimeIdentifier)' == 'osx-arm64' Or '$(RuntimeIdentifier)' == 'win-x64' Or '$(RuntimeIdentifier)' == 'win-arm64'" />
<PropertyGroup>
<OutputType>Exe</OutputType>
<AssemblyName>monitor-probe</AssemblyName>
<RootNamespace>Monitor.Probe</RootNamespace>
<IsAotCompatible>true</IsAotCompatible>
</PropertyGroup>
<ItemGroup>
<InternalsVisibleTo Include="Monitor.Backend.Tests" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Link\Fleet.Link.csproj" />
<ProjectReference Include="..\..\..\fleet\src\Fleet.Hosting\Fleet.Hosting.csproj" />
<ProjectReference Include="..\..\..\shared\Monitor.Measurement\Monitor.Measurement.csproj" />
</ItemGroup>
<Import Project="Sdk.targets" Sdk="Microsoft.NET.Sdk" />
<Import Project="Sdk.targets" Sdk="StuDev.AotAnywhere"
Condition="'$(RuntimeIdentifier)' == 'linux-x64' Or '$(RuntimeIdentifier)' == 'linux-arm64' Or '$(RuntimeIdentifier)' == 'linux-arm' Or '$(RuntimeIdentifier)' == 'linux-musl-x64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm64' Or '$(RuntimeIdentifier)' == 'linux-musl-arm' Or '$(RuntimeIdentifier)' == 'osx-x64' Or '$(RuntimeIdentifier)' == 'osx-arm64' Or '$(RuntimeIdentifier)' == 'win-x64' Or '$(RuntimeIdentifier)' == 'win-arm64'" />
</Project>
-154
View File
@@ -1,154 +0,0 @@
using Fleet.Core;
using Fleet.Hosting;
using Fleet.Identity;
using Monitor.Measurement;
namespace Monitor.Probe;
/// <summary>
/// Probe 的组装点:Agent 列表 → 每个 Agent 一个连接(认证、重连、角色)。没有全局状态,测试可以在同一进程里启动多个。
/// 列表变化时只调和连接:新增的建立,移除的断开,地址或证书指纹变化的重建,其余不受影响。
/// </summary>
public sealed class ProbeApp : IDisposable
{
public const string Version = "0.1.0";
private readonly ConnectionEnv _env;
private readonly Dictionary<string, Running> _connections = [];
private readonly SemaphoreSlim _changed = new(0);
private readonly INodeLog _log;
public ProbeApp(ProbeFile file, string configDirectory, TimeProvider time, INodeLog log)
{
_log = log;
var directory = new AgentDirectory(new ConfigStore(configDirectory), configDirectory, file, log);
NodeIdentity identity = NodeIdentity.FromPkcs8(Convert.FromBase64String(file.PrivateKey));
_env = new ConnectionEnv(identity, file.ProbeId, file.Limits, directory, new MonoClock(time), time, log);
directory.Changed += () => _changed.Release();
}
public AgentDirectory Directory => _env.Directory;
/// <summary>某个 Agent 的会话是否已经建立(测试和状态查看用)。</summary>
public bool IsConnected(string agentId)
{
lock (_connections)
{
return _connections.TryGetValue(agentId, out Running? running) && running.Connection.IsConnected;
}
}
public AgentConnection? Connection(string agentId)
{
lock (_connections)
{
return _connections.TryGetValue(agentId, out Running? running) ? running.Connection : null;
}
}
public async Task RunAsync(CancellationToken ct)
{
if (_env.Directory.Agents.Count == 0)
{
_log.Warn("no agents known yet: probe.json has an empty bootstrap list");
}
try
{
while (!ct.IsCancellationRequested)
{
await ReconcileAsync(_env.Directory.Agents, ct);
await _changed.WaitAsync(ct);
}
}
catch (OperationCanceledException)
{
// 停止
}
finally
{
await ReconcileAsync([], CancellationToken.None);
}
}
public void Dispose()
{
_env.Identity.Dispose();
_changed.Dispose();
}
private async Task ReconcileAsync(IReadOnlyList<BootstrapAgent> agents, CancellationToken ct)
{
var wanted = agents.ToDictionary(a => a.AgentId);
var stopped = new List<Running>();
lock (_connections)
{
foreach ((string id, Running running) in _connections.ToList())
{
if (!wanted.TryGetValue(id, out BootstrapAgent? endpoint) || !AgentDirectory.Equal(endpoint, running.Connection.Endpoint))
{
stopped.Add(running);
_connections.Remove(id);
}
}
}
foreach (Running running in stopped)
{
await running.StopAsync();
}
if (ct.IsCancellationRequested)
{
return;
}
lock (_connections)
{
foreach ((string id, BootstrapAgent endpoint) in wanted)
{
if (!_connections.ContainsKey(id))
{
_connections[id] = Running.Start(new AgentConnection(endpoint, _env));
}
}
}
}
private sealed class Running : IDisposable
{
private readonly CancellationTokenSource _cts = new();
private Task _run = Task.CompletedTask;
private Running(AgentConnection connection)
{
Connection = connection;
}
public AgentConnection Connection { get; }
public static Running Start(AgentConnection connection)
{
var running = new Running(connection);
running._run = Task.Run(() => connection.RunAsync(running._cts.Token), CancellationToken.None);
return running;
}
public async Task StopAsync()
{
await _cts.CancelAsync();
try
{
await _run.WaitAsync(TimeSpan.FromSeconds(10));
}
catch (Exception ex) when (ex is OperationCanceledException or TimeoutException)
{
// 没有及时退出:不再等
}
Dispose();
}
public void Dispose() => _cts.Dispose();
}
}
@@ -5,7 +5,7 @@
</PropertyGroup>
<ItemGroup>
<PackageReference Include="xunit.v3" />
<ProjectReference Include="..\..\src\Monitor.Probe\Monitor.Probe.csproj" />
<ProjectReference Include="..\..\src\MoShiro.Probe\MoShiro.Probe.csproj" />
</ItemGroup>
<ItemGroup>
<Using Include="Xunit" />
@@ -1,11 +1,36 @@
using Fleet.Core;
using Fleet.Hosting;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Probe.Tests;
namespace MoShiro.Probe.Tests;
public sealed class ProbeTests
{
[Fact]
public void AgentDirectory_DualStack_PersistsBothPaths_AndRemovingIpv6KeepsIpv4()
{
string directory = NewDirectory();
try
{
var agents = new AgentDirectory(new ConfigStore(directory), directory, NullNodeLog.Instance);
ProbeAgentList list = List(1, "agt_dual");
list.Agents[0].Ipv6Host = "2001:db8::1";
agents.Apply(list);
Assert.Equal(new[] { "ipv4", "ipv6" }, agents.Agents.Select(a => a.IpVersion));
var reloaded = new AgentDirectory(new ConfigStore(directory), directory, NullNodeLog.Instance);
Assert.Equal(2, reloaded.Agents.Count);
Assert.Equal("2001:db8::1", reloaded.Agents[1].Host);
list.Version = 2;
list.Agents[0].Ipv6Host = null;
reloaded.Apply(list);
Assert.Equal("ipv4", Assert.Single(reloaded.Agents).IpVersion);
}
finally
{
Directory.Delete(directory, recursive: true);
}
}
[Fact]
public void ClockEstimator_UsesTheSampleWithTheShortestRoundTrip()
{
@@ -30,23 +55,22 @@ public sealed class ProbeTests
return path;
}
private static ProbeAgentsResult List(long version, params string[] ids) => new()
private static ProbeAgentList List(long version, params string[] ids) => new()
{
Version = version,
Agents = [.. ids.Select(id => new AgentEndpoint { AgentId = new NodeId(id), Host = id + ".example", TcpPort = 7443, Spki = "c3BraQ==" })],
};
[Fact]
public void AgentDirectory_StartsFromBootstrap_IgnoresStaleLists_PersistsChanges_AndReloadsThem()
public void AgentDirectory_StartsEmpty_IgnoresStaleLists_PersistsChanges_AndReloadsThem()
{
string directory = NewDirectory();
try
{
var file = new ProbeFile { ProbeId = "prb_x", Bootstrap = [new BootstrapAgent { AgentId = "agt_boot", Host = "boot.example", TcpPort = 7443, Spki = "c3BraQ==" }] };
var directoryA = new AgentDirectory(new ConfigStore(directory), directory, file, NullNodeLog.Instance);
var directoryA = new AgentDirectory(new ConfigStore(directory), directory, NullNodeLog.Instance);
int changes = 0;
directoryA.Changed += () => changes++;
Assert.Equal(["agt_boot"], directoryA.Agents.Select(a => a.AgentId));
Assert.Empty(directoryA.Agents);
directoryA.Apply(List(5, "agt_1", "agt_2"));
Assert.Equal(1, changes);
@@ -57,8 +81,8 @@ public sealed class ProbeTests
Assert.Equal(1, changes);
Assert.Equal(5, directoryA.Version);
// The next start uses the saved list, not the bootstrap one
var directoryB = new AgentDirectory(new ConfigStore(directory), directory, file, NullNodeLog.Instance);
// The next start uses the saved list while the panel is not reachable yet
var directoryB = new AgentDirectory(new ConfigStore(directory), directory, NullNodeLog.Instance);
Assert.Equal(["agt_1", "agt_2"], directoryB.Agents.Select(a => a.AgentId));
Assert.Equal(5, directoryB.Version);
}
@@ -69,15 +93,30 @@ public sealed class ProbeTests
}
[Fact]
public void AgentDirectory_FallsBackToBootstrap_WhenTheSavedFileIsCorrupt()
public void AgentDirectory_StartsEmpty_WhenTheSavedFileIsCorrupt()
{
string directory = NewDirectory();
try
{
new ConfigStore(directory).Write(AgentDirectory.FileName, "{ not json"u8.ToArray());
var file = new ProbeFile { Bootstrap = [new BootstrapAgent { AgentId = "agt_boot", Host = "h", TcpPort = 1, Spki = "AA==" }] };
var loaded = new AgentDirectory(new ConfigStore(directory), directory, file, NullNodeLog.Instance);
Assert.Equal(["agt_boot"], loaded.Agents.Select(a => a.AgentId));
var loaded = new AgentDirectory(new ConfigStore(directory), directory, NullNodeLog.Instance);
Assert.Empty(loaded.Agents);
}
finally
{
Directory.Delete(directory, recursive: true);
}
}
[Fact]
public void ConfigStore_CreatesSubdirectories_ForFreshVolumes()
{
// 新建的命名卷里没有 update/:面板转来的更新清单要能写进去
string directory = NewDirectory();
try
{
new ConfigStore(directory).Write("update/offer.manifest", "{}"u8.ToArray());
Assert.True(File.Exists(Path.Combine(directory, "update", "offer.manifest")));
}
finally
{
@@ -1,7 +1,7 @@
using System.Net;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>积木的公共部分:环境、参数、把行切成列块交出去。</summary>
public abstract class BlockBase : IBlock
@@ -1,6 +1,6 @@
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>
/// 一种采集方法(Kind)的两个角色:拨号方(Probe)和监听方(Agent),以及它能不能作为目标任务(Agent 自己探测一个地址,没有对端)运行。
@@ -1,7 +1,7 @@
using System.Net.Sockets;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>批量传输的收发循环(两端共用)。</summary>
internal static class BulkIo
@@ -1,8 +1,8 @@
using System.Net;
using System.Net.Sockets;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>
/// 回显积木的行表(拨号方,行的主导方):发出时记 size、t1,回复到达时按序号回填 t4,到点把已经稳定的行切成列块。
@@ -1,9 +1,10 @@
using System.Net.NetworkInformation;
using System.Net.Sockets;
using System.Security.Authentication;
using Monitor.Protocol;
using Fleet.Core;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>什么也不做的监听方:connect 的另一端就是 TCP 监听器本身(接受连接、认证失败就关闭),不需要积木,但需要占一个流编号。</summary>
public sealed class IdleListener : BlockBase
@@ -306,7 +307,8 @@ public sealed class IcmpEchoSolo : BlockBase
TimeSpan interval = TimeSpan.FromSeconds(1.0 / Spec.Int("rate_hz"));
int size = (int)Spec.Int("size");
byte[] payload = new byte[Math.Max(0, size - 28)];
using var ping = new Ping();
using LinuxIcmpEcho? linuxPing = OperatingSystem.IsLinux() ? LinuxIcmpEcho.TryCreate(Env.PeerAddress!, payload.Length) : null;
using Ping? ping = linuxPing is null ? new Ping() : null;
try
{
while (!ct.IsCancellationRequested)
@@ -315,8 +317,10 @@ public sealed class IcmpEchoSolo : BlockBase
long? t4 = null;
try
{
PingReply reply = await ping.SendPingAsync(Env.PeerAddress!, TimeSpan.FromSeconds(2), payload, null, ct);
if (reply.Status == IPStatus.Success)
bool receivedReply = linuxPing is not null
? await linuxPing.SendAsync(TimeSpan.FromSeconds(2), ct)
: (await ping!.SendPingAsync(Env.PeerAddress!, TimeSpan.FromSeconds(2), payload, null, ct)).Status == IPStatus.Success;
if (receivedReply)
{
t4 = Env.Clock.NowUs();
}
@@ -325,6 +329,11 @@ public sealed class IcmpEchoSolo : BlockBase
{
// 发送失败(例如本机没有路由):记为丢失
}
catch (PlatformNotSupportedException) when (OperatingSystem.IsLinux())
{
Env.Log.Warn("ICMP permission unavailable", ("hint", "allow the mapped process GID in net.ipv4.ping_group_range, or grant effective NET_RAW"));
throw;
}
long key = _log.Store.Append();
_log.Store.Set(key, _seq, key);
@@ -1,8 +1,8 @@
using System.Diagnostics;
using System.Net;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>
/// 单向数据流(UDP train),监听方(Agent,行的主导方):按声明的码率均匀发包,发出的每个包 (seq, size, t_send) 记在按序号的记录里,
@@ -1,9 +1,9 @@
using System.Net;
using System.Text.Json;
using Fleet.Core;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>
/// 数据面会话:<c>session_id</c> 和密钥(由点对点握手里的 ECDH 派生)。两端的数据面包都用它认证。
@@ -1,4 +1,4 @@
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>节点的单调时钟(微秒):t1…t4、t_send、t_recv 都用它;Probe 和 Agent 各自一个,块头里的时钟锚点把它对应到 UTC。</summary>
public sealed class MonoClock
@@ -1,6 +1,6 @@
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>
/// 一侧记录的行:只含这一侧拥有的列(见 <see cref="StreamSpec.ColumnsOf"/>),按连接键(通常是 <c>seq</c>)索引的环形缓冲,
@@ -1,4 +1,4 @@
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>一个运行中的积木:自己的取消令牌和后台任务。Probe 和 Agent 共用。</summary>
public sealed class RunningBlock : IDisposable
@@ -1,8 +1,8 @@
using System.Net;
using System.Text.Json;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>节点的本地硬上限(写在本机配置里,Backend 无法修改)。</summary>
public sealed class NodeLimits
@@ -1,8 +1,8 @@
using System.Text.Json;
using Fleet.Core;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>Agent 单调时钟(微秒)与 UTC(毫秒)的对应关系:块头里的时钟锚点。</summary>
public readonly record struct ClockAnchor(long AgentMonoUs, long UtcMs);
@@ -69,6 +69,7 @@ public static class ChunkFinalizer
Phase = source.Phase,
AgentId = agent,
ProbeId = source.ProbeId,
IpVersion = source.IpVersion,
Target = null,
Params = new Dictionary<string, JsonElement>(resolvedParams),
Port = port,
@@ -1,8 +1,8 @@
using System.Text.Json;
using Fleet.Core;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>Agent 一侧为合并保留的按键记录(由正在运行的积木提供)。</summary>
public interface IRecordSource
@@ -29,7 +29,7 @@ public static class FragmentMerger
StreamSpec? streamSpec = spec.FindStream(stream.Name);
if (streamSpec is null)
{
return Result<IReadOnlyList<RawStream>>.Fail(MonitorErrors.ValidationFailed, $"stream {stream.Name} is not declared by {spec.Id}");
return Result<IReadOnlyList<RawStream>>.Fail(MoShiroErrors.ValidationFailed, $"stream {stream.Name} is not declared by {spec.Id}");
}
Result<RawStream> merged = Join(streamSpec, stream, fragment.Header, records.Records(streamId, stream.Name));
@@ -71,7 +71,7 @@ public static class FragmentMerger
RawColumn? key = probe.Find(spec.Key);
if (key is null || probeColumns.Any(c => probe.Find(c.Name) is null))
{
return Result<RawStream>.Fail(MonitorErrors.ValidationFailed, $"a fragment of stream {spec.Name} needs the columns {string.Join(", ", new[] { spec.Key }.Concat(probeColumns.Select(c => c.Name)))}");
return Result<RawStream>.Fail(MoShiroErrors.ValidationFailed, $"a fragment of stream {spec.Name} needs the columns {string.Join(", ", new[] { spec.Key }.Concat(probeColumns.Select(c => c.Name)))}");
}
return spec.Driver == Side.Probe ? JoinOnProbeRows(spec, probe, key, agent) : JoinOnAgentRows(spec, probe, key, header, agent);
@@ -118,7 +118,7 @@ public static class FragmentMerger
{
if (!TryGetRange(header, out uint from, out uint to))
{
return Result<RawStream>.Fail(MonitorErrors.ValidationFailed, $"a fragment of stream {spec.Name} must declare its sequence range");
return Result<RawStream>.Fail(MoShiroErrors.ValidationFailed, $"a fragment of stream {spec.Name} must declare its sequence range");
}
// 行取自 Agent 的记录:范围内发出过的每一个包都有一行,Probe 没有报告到达的就是丢失
@@ -3,8 +3,8 @@
<IsAotCompatible>true</IsAotCompatible>
</PropertyGroup>
<ItemGroup>
<InternalsVisibleTo Include="Monitor.Measurement.Tests" />
<InternalsVisibleTo Include="MoShiro.Measurement.Tests" />
<AdditionalFiles Include="$(MSBuildThisFileDirectory)../../BannedSymbols.Node.txt" />
<ProjectReference Include="..\Monitor.Protocol\Monitor.Protocol.csproj" />
<ProjectReference Include="..\MoShiro.Protocol\MoShiro.Protocol.csproj" />
</ItemGroup>
</Project>
@@ -0,0 +1,84 @@
using System.Buffers.Binary;
using System.Net;
using System.Net.Sockets;
namespace MoShiro.Measurement;
/// <summary>Linux ping sockets permit ICMP echo without CAP_NET_RAW when ping_group_range allows the process group.</summary>
internal sealed class LinuxIcmpEcho : IDisposable
{
private readonly Socket _socket;
private readonly byte[] _request;
private readonly byte[] _response;
private readonly byte _replyType;
private ushort _sequence;
private LinuxIcmpEcho(IPAddress address, int payloadSize)
{
bool ipv6 = address.AddressFamily == AddressFamily.InterNetworkV6;
_socket = new Socket(address.AddressFamily, SocketType.Dgram, ipv6 ? ProtocolType.IcmpV6 : ProtocolType.Icmp);
try
{
_socket.Connect(new IPEndPoint(address, 0));
}
catch
{
_socket.Dispose();
throw;
}
_request = new byte[8 + payloadSize];
_request[0] = ipv6 ? (byte)128 : (byte)8;
_replyType = ipv6 ? (byte)129 : (byte)0;
_response = new byte[_request.Length + 64];
// The Linux ping socket supplies the identifier and checksum.
Random.Shared.NextBytes(_request.AsSpan(8));
}
public static LinuxIcmpEcho? TryCreate(IPAddress address, int payloadSize)
{
try
{
return new LinuxIcmpEcho(address, payloadSize);
}
catch (SocketException)
{
// Preserve System.Net.Ping's raw-socket path for hosts that allow NET_RAW instead.
return null;
}
}
public async ValueTask<bool> SendAsync(TimeSpan timeout, CancellationToken ct)
{
ushort sequence = unchecked(++_sequence);
BinaryPrimitives.WriteUInt16BigEndian(_request.AsSpan(6), sequence);
using var deadline = CancellationTokenSource.CreateLinkedTokenSource(ct);
deadline.CancelAfter(timeout);
try
{
await _socket.SendAsync(_request, SocketFlags.None, deadline.Token);
while (true)
{
int count = await _socket.ReceiveAsync(_response, SocketFlags.None, deadline.Token);
if (IsReply(_response.AsSpan(0, count), _request, _replyType, sequence))
{
return true;
}
}
}
catch (OperationCanceledException) when (!ct.IsCancellationRequested)
{
return false;
}
catch (SocketException)
{
return false;
}
}
internal static bool IsReply(ReadOnlySpan<byte> reply, ReadOnlySpan<byte> request, byte replyType, ushort sequence) =>
reply.Length == request.Length && reply.Length >= 8 && reply[0] == replyType && reply[1] == 0 &&
BinaryPrimitives.ReadUInt16BigEndian(reply[6..]) == sequence && reply[8..].SequenceEqual(request[8..]);
public void Dispose() => _socket.Dispose();
}
@@ -5,9 +5,9 @@ using System.Net.Sockets;
using System.Security.Authentication;
using System.Security.Cryptography;
using System.Security.Cryptography.X509Certificates;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>拨号方的一条数据面 TCP 连接(已经完成 TLS 和首帧认证)。</summary>
public sealed class TcpLink : IDisposable
@@ -39,7 +39,7 @@ public sealed class TcpLink : IDisposable
/// </summary>
public static async Task<TcpLink> OpenAsync(BlockEnv env, int port, bool tls, uint streamId, bool sendAuth, CancellationToken ct)
{
var client = new TcpClient { NoDelay = true };
var client = new TcpClient(env.PeerAddress!.AddressFamily) { NoDelay = true };
try
{
long start = env.Clock.NowUs();
@@ -153,6 +153,11 @@ public sealed class TcpAcceptor
foreach (int port in _ports)
{
var listener = new TcpListener(_bind, port);
if (_bind.Equals(IPAddress.IPv6Any))
{
listener.Server.DualMode = true;
}
listener.Start(128);
_listeners.Add(listener);
bound.Add(((IPEndPoint)listener.LocalEndpoint).Port);
@@ -1,8 +1,8 @@
using System.Buffers.Binary;
using System.Net.Sockets;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>内核对一条 TCP 连接的采样(Linux <c>TCP_INFO</c>)。</summary>
public readonly record struct TcpInfoSample(long SrttUs, long RttVarUs, long Cwnd, long TotalRetrans, long SegsOut, long BytesAcked);
@@ -2,9 +2,9 @@ using System.Collections.Concurrent;
using System.Net;
using System.Net.Sockets;
using Fleet.Core;
using Monitor.Protocol;
using MoShiro.Protocol;
namespace Monitor.Measurement;
namespace MoShiro.Measurement;
/// <summary>一个通过校验的 UDP 包,交给按 <c>stream_id</c> 登记的处理函数。处理函数在接收循环里同步执行,不能阻塞。</summary>
public readonly ref struct UdpDelivery
@@ -0,0 +1,154 @@
using System.Globalization;
using System.Net;
using System.Net.NetworkInformation;
using System.Net.Sockets;
using MoShiro.Protocol;
namespace MoShiro.Measurement;
/// <summary>诊断采集进程所在的网络环境。宿主网络容器看到的是主机;桥接容器看到自己的网络环境。</summary>
public static class Ipv6Diagnostics
{
public static Ipv6Diagnostic Sample(bool? listening = null, string? networkScope = null, string? hostProcRoot = null)
{
bool supported = Socket.OSSupportsIPv6;
bool? address = null;
bool? route = null;
if (supported)
{
if (OperatingSystem.IsLinux())
{
address = ParseAddresses(Read("/proc/net/if_inet6"));
route = ParseRoutes(Read("/proc/net/ipv6_route"));
}
else if (OperatingSystem.IsWindows() || OperatingSystem.IsMacOS())
{
try
{
address = false;
route = false;
foreach (NetworkInterface adapter in NetworkInterface.GetAllNetworkInterfaces())
{
if (adapter.OperationalStatus != OperationalStatus.Up || adapter.NetworkInterfaceType == NetworkInterfaceType.Loopback)
{
continue;
}
IPInterfaceProperties properties = adapter.GetIPProperties();
foreach (UnicastIPAddressInformation unicast in properties.UnicastAddresses)
{
if (IsGlobal(unicast.Address))
{
address = true;
}
}
foreach (GatewayIPAddressInformation gateway in properties.GatewayAddresses)
{
if (gateway.Address.AddressFamily == AddressFamily.InterNetworkV6 && !gateway.Address.Equals(IPAddress.IPv6Any))
{
route = true;
}
}
}
}
catch (Exception ex) when (ex is NetworkInformationException or PlatformNotSupportedException)
{
address = null;
route = null;
}
}
}
string scope = networkScope is "host" or "container" ? networkScope
: File.Exists("/run/.containerenv") || File.Exists("/.dockerenv") ? "container" : "host";
bool? hostAddress = scope == "host" ? address : null;
bool? hostRoute = scope == "host" ? route : null;
if (scope == "container" && OperatingSystem.IsLinux() && hostProcRoot is not null)
{
// /host/proc/net 通常指向当前进程的网络空间;宿主 PID 1 才是宿主网络。
hostAddress = ParseAddresses(Read(Path.Combine(hostProcRoot, "1/net/if_inet6")));
hostRoute = ParseRoutes(Read(Path.Combine(hostProcRoot, "1/net/ipv6_route")));
}
Ipv6Diagnostic result = Evaluate(supported, listening, address, route);
result.NetworkScope = scope;
result.HostHasGlobalAddress = hostAddress;
result.HostHasDefaultRoute = hostRoute;
if (scope == "container" && address == false && hostAddress == true)
{
result.State = "container_network_not_ready";
}
return result;
}
public static Ipv6Diagnostic Evaluate(bool supported, bool? listening, bool? address, bool? route) => new()
{
SocketSupported = supported,
Listening = listening,
HasGlobalAddress = address,
HasDefaultRoute = route,
State = !supported ? "unsupported" : address == false ? "no_global_address" : listening == false ? "not_listening"
: route == false ? "no_default_route" : address is null || route is null ? "unknown" : "ready",
};
public static bool? ParseAddresses(string? text)
{
if (text is null)
{
return null;
}
foreach (string line in text.Split('\n'))
{
string[] fields = line.Split((char[]?)null, StringSplitOptions.RemoveEmptyEntries);
if (fields.Length != 6 || fields[0].Length != 32 || fields[0][0] is not ('2' or '3') || fields[3] != "00")
{
continue;
}
// tentative、DAD failed 和 deprecated 地址不能作为就绪的公网地址。
if (uint.TryParse(fields[4], NumberStyles.HexNumber, CultureInfo.InvariantCulture, out uint flags) && (flags & 0x68) == 0)
{
return true;
}
}
return false;
}
public static bool? ParseRoutes(string? text)
{
if (text is null)
{
return null;
}
foreach (string line in text.Split('\n'))
{
string[] fields = line.Split((char[]?)null, StringSplitOptions.RemoveEmptyEntries);
if (fields.Length == 10 && fields[0] == "00000000000000000000000000000000" && fields[1] == "00" && fields[9] != "lo"
&& uint.TryParse(fields[8], NumberStyles.HexNumber, CultureInfo.InvariantCulture, out uint flags) && (flags & 1) != 0 && (flags & 0x200) == 0)
{
return true;
}
}
return false;
}
private static bool IsGlobal(IPAddress address) => address.AddressFamily == AddressFamily.InterNetworkV6 && (address.GetAddressBytes()[0] & 0xE0) == 0x20;
private static string? Read(string path)
{
try
{
return File.Exists(path) ? File.ReadAllText(path) : null;
}
catch (Exception ex) when (ex is IOException or UnauthorizedAccessException)
{
return null;
}
}
}
@@ -0,0 +1,81 @@
using System.Runtime.InteropServices;
using MoShiro.Protocol;
namespace MoShiro.Measurement;
/// <summary>
/// 读出本机的系统名和架构,只用于在面板上展示。故意只取粗粒度的信息:
/// 「Debian 12」「Windows 11」「x64」,不读主机名、内核版本、CPU 型号。
/// </summary>
public static class PlatformDetector
{
public static PlatformInfo Detect()
{
PlatformInfo info = Detect(File.Exists("/etc/os-release") ? File.ReadAllLines("/etc/os-release") : []);
info.Version = Empty(BuildInfo.Version);
info.Tag = Empty(BuildInfo.Tag);
info.Commit = Empty(BuildInfo.Commit);
info.BuiltAt = Empty(BuildInfo.BuiltAt);
return info;
}
private static string? Empty(string value) => value.Length == 0 ? null : value;
/// <summary>`osRelease` 是 /etc/os-release 的各行(Linux 之外不用)。拆出来是为了好测。</summary>
internal static PlatformInfo Detect(IEnumerable<string> osRelease)
{
string arch = RuntimeInformation.OSArchitecture switch
{
Architecture.X64 => "x64",
Architecture.Arm64 => "arm64",
Architecture.Arm => "arm",
Architecture.X86 => "x86",
var other => other.ToString().ToLowerInvariant(),
};
if (OperatingSystem.IsWindows())
{
// Windows 11 的版本号还是 10.0,靠内部版本号(22000 起)区分
string name = Environment.OSVersion.Version.Build >= 22000 ? "Windows 11" : "Windows 10";
return new PlatformInfo { Os = "windows", Name = name, Arch = arch };
}
if (OperatingSystem.IsMacOS())
{
return new PlatformInfo { Os = "macos", Name = "macOS", Arch = arch };
}
if (OperatingSystem.IsLinux())
{
return new PlatformInfo { Os = "linux", Name = LinuxName(osRelease), Arch = arch };
}
return new PlatformInfo { Os = "other", Name = null, Arch = arch };
}
/// <summary>从 os-release 里取发行版名字和主版本:「Debian GNU/Linux 12 (bookworm)」→「Debian 12」,「OpenWrt 24.10.0」→「OpenWrt 24.10」。</summary>
internal static string LinuxName(IEnumerable<string> osRelease)
{
var values = new Dictionary<string, string>(StringComparer.Ordinal);
foreach (string line in osRelease)
{
int eq = line.IndexOf('=', StringComparison.Ordinal);
if (eq > 0)
{
values[line[..eq].Trim()] = line[(eq + 1)..].Trim().Trim('"', '\'');
}
}
string name = values.GetValueOrDefault("NAME") ?? values.GetValueOrDefault("ID") ?? "Linux";
name = name.Replace(" GNU/Linux", string.Empty, StringComparison.Ordinal).Trim();
string? version = values.GetValueOrDefault("VERSION_ID");
if (string.IsNullOrEmpty(version))
{
return name;
}
// 只留 主.次:Alpine 的 3.22.1、OpenWrt 的 24.10.0 都截成两段;Debian 的 12 本来就一段
string[] parts = version.Split('.');
return $"{name} {string.Join('.', parts.Take(2))}";
}
}
@@ -0,0 +1,100 @@
using System.Text;
using System.Text.Json;
using Fleet.Core;
using MoShiro.Protocol;
namespace MoShiro.Measurement;
/// <summary>
/// 节点进程和宿主机更新器之间的文件交换(容器里 <c>/etc/moshiro</c>,也就是数据目录):
/// 节点把面板发来的清单和签名写进 <c>update/</c>,把「更新」请求写成 <c>update/apply.request</c>;更新器验签并更新,状态由宿主更新器写在只读挂载的 <c>update-status/status.json</c>。
/// 节点进程不执行任何命令,也不信任清单的内容(验签在宿主机上)。没有安装更新器时目录写不了或没人读,调用只是返回 false。
/// </summary>
public sealed class UpdateBridge
{
private const int MaxManifestBytes = 4096;
private const int MaxSignatureBytes = 512;
private const int MaxStatusBytes = 4096;
private readonly string _updateDirectory;
private readonly string _statusFile;
private readonly Action<string, byte[]> _write;
private readonly INodeLog _log;
/// <param name="write">写数据目录里的文件(相对路径,原子替换):由节点传入 <c>ConfigStore.Write</c>——常驻进程里只有它可以写文件。</param>
public UpdateBridge(string dataDirectory, Action<string, byte[]> write, INodeLog log)
{
_updateDirectory = Path.Combine(dataDirectory, "update");
_statusFile = Path.Combine(dataDirectory, "update-status", "status.json");
_write = write;
_log = log;
}
/// <summary>放下面板发来的清单和签名(和已有的一样就不动,免得每次重连都惊动更新器)。先写清单、后写签名:更新器监视的是签名文件。</summary>
public bool WriteOffer(UpdateOfferParams offer)
{
byte[] manifest = Encoding.UTF8.GetBytes(offer.Manifest);
byte[] signature = Encoding.UTF8.GetBytes(offer.Signature);
if (manifest.Length is 0 or > MaxManifestBytes || signature.Length is 0 or > MaxSignatureBytes)
{
_log.Warn("ignoring an update offer of an unexpected size");
return false;
}
try
{
string manifestPath = Path.Combine(_updateDirectory, "offer.manifest");
string signaturePath = Path.Combine(_updateDirectory, "offer.sig");
if (Same(manifestPath, manifest) && Same(signaturePath, signature))
{
return true;
}
_write("update/offer.manifest", manifest);
_write("update/offer.sig", signature);
_log.Info("update offer received");
return true;
}
catch (Exception ex) when (ex is IOException or UnauthorizedAccessException)
{
_log.Warn("could not store the update offer (is the updater installed?)", ("err", ex.Message));
return false;
}
}
/// <summary>请求更新:写一个空的 <c>apply.request</c>(更新器验签通过、清单更新才会真的动手)。</summary>
public bool RequestApply()
{
try
{
_write("update/apply.request", []);
_log.Info("update requested");
return true;
}
catch (Exception ex) when (ex is IOException or UnauthorizedAccessException)
{
_log.Warn("could not request the update (is the updater installed?)", ("err", ex.Message));
return false;
}
}
/// <summary>更新器的状态;没有安装更新器、文件不存在或内容不对时返回 null。</summary>
public UpdateStatus? ReadStatus()
{
try
{
if (!File.Exists(_statusFile) || new FileInfo(_statusFile).Length > MaxStatusBytes)
{
return null;
}
return JsonSerializer.Deserialize(File.ReadAllBytes(_statusFile), MoShiroJsonContext.Default.UpdateStatus);
}
catch (Exception ex) when (ex is IOException or UnauthorizedAccessException or JsonException)
{
return null;
}
}
private static bool Same(string path, byte[] content) => File.Exists(path) && File.ReadAllBytes(path).AsSpan().SequenceEqual(content);
}
@@ -1,4 +1,4 @@
namespace Monitor.Protocol;
namespace MoShiro.Protocol;
[Flags]
public enum Scope
@@ -1,4 +1,4 @@
namespace Monitor.Protocol;
namespace MoShiro.Protocol;
/// <summary>
/// Kind 目录。新增一种采集方法 = 在这里加一个静态字段并放进 <see cref="All"/>;
@@ -1,7 +1,7 @@
using System.Globalization;
using System.Text.Json;
namespace Monitor.Protocol;
namespace MoShiro.Protocol;
public enum ParamType
{
@@ -1,7 +1,7 @@
using System.Text.Json.Serialization;
using Fleet.Core;
namespace Monitor.Protocol;
namespace MoShiro.Protocol;
[JsonConverter(typeof(ProbeIdJsonConverter))]
public readonly record struct ProbeId(string Value)
@@ -0,0 +1,13 @@
namespace MoShiro.Protocol;
/// <summary>测量路径的地址族。旧契约缺省为 IPv4;IPv6 使用独立的线路和序列。</summary>
public static class IpVersions
{
public const string V4 = "ipv4";
public const string V6 = "ipv6";
public static bool IsValid(string? value) => value is V4 or V6;
public static string LineSeries(string probeId, string ipVersion) =>
RawChunkHeader.LinePrefix + probeId + (ipVersion == V6 ? ":ipv6" : "");
}
@@ -0,0 +1,31 @@
using System.Text.Json.Serialization;
namespace MoShiro.Protocol;
/// <summary>当前网络环境的 IPv6 诊断。仅上报状态,不传地址、接口名称或路由内容。</summary>
public sealed class Ipv6Diagnostic
{
[JsonPropertyName("socket_supported")]
public bool SocketSupported { get; set; }
[JsonPropertyName("listening")]
public bool? Listening { get; set; }
[JsonPropertyName("network_scope")]
public string NetworkScope { get; set; } = "unknown";
[JsonPropertyName("host_has_global_address")]
public bool? HostHasGlobalAddress { get; set; }
[JsonPropertyName("host_has_default_route")]
public bool? HostHasDefaultRoute { get; set; }
[JsonPropertyName("has_global_address")]
public bool? HasGlobalAddress { get; set; }
[JsonPropertyName("has_default_route")]
public bool? HasDefaultRoute { get; set; }
[JsonPropertyName("state")]
public string State { get; set; } = "unknown";
}
Loaded 100 of 404 files, more files were not shown because too many files have changed in this diff. Show more