05060ca337
fix: 降低 Todo Note Git 快照内存峰值
2026-07-12 12:10:38 +02:00
d35c14fe4a
fix: 保留 Todo 写后健康计数
2026-07-10 07:33:16 +02:00
fbd9dbbae1
fix: use owner-level unidesk state
2026-07-10 05:09:13 +02:00
0184bd7334
feat: migrate todo note to nc01 github storage
2026-07-10 03:47:30 +02:00
4bb930bb36
fix: target decision center data repo
2026-07-10 03:22:37 +02:00
6fc459e1fb
feat: add github-backed decision storage
2026-07-10 03:22:37 +02:00
50a166a2ce
fix: proxy internal-sidecar microservices directly
2026-07-10 03:22:37 +02:00
0f14bc16a5
feat: deploy decision center on nc01 k8s
2026-07-10 03:22:37 +02:00
8983657fe3
fix: handle expired frontend sessions globally
Pipelines as Code CI / hwlab-web-probe-sentinel-nc01- Success
2026-07-10 01:02:59 +02:00
bd4722c9a8
fix: allow provider data channels without token gate
2026-07-08 08:43:32 +02:00
bc7406ecbd
fix: allow provider registration without token gate
2026-07-08 08:30:47 +02:00
deb2044037
fix: update provider gateway default master host
2026-07-08 05:49:17 +00:00
5bd0f3921e
fix: complete provider upgrade after final heartbeat
2026-07-05 18:21:18 +00:00
42f605dceb
fix: add provider rootfs diagnostic
2026-07-05 18:08:17 +00:00
41414b8211
fix: guard provider gateway runtime dependencies
2026-07-05 16:55:38 +00:00
ef68531e17
fix: quiet late ssh data frames after provider sessions
2026-06-29 16:18:41 +00:00
d88778d7da
fix: buffer early ssh data frames in provider gateway
2026-06-29 15:42:15 +00:00
076c4b643d
feat: add JD01 YAML-first deployment support
2026-06-29 08:13:34 +00:00
3cea6d96b7
fix: fail explicit host ssh cwd
2026-06-29 06:59:34 +00:00
4a636abb73
fix: isolate provider egress tunnel diagnostics
2026-06-28 16:15:41 +00:00
0b923fc5ee
fix: bump provider gateway for egress diagnostics ( #970 )
...
Co-authored-by: Codex <codex@noreply.local >
2026-06-26 13:00:52 +08:00
d3d542fbd3
fix: improve egress and job diagnostics ( #969 )
...
Co-authored-by: Codex <codex@noreply.local >
2026-06-26 12:51:58 +08:00
754789c43c
fix: route D601 PK01 postgres through master relay ( #967 )
...
Co-authored-by: Codex <codex@noreply.local >
2026-06-26 12:22:30 +08:00
2980e675c4
fix(hwlab): route d601 v03 postgres through gateway ( #957 )
...
Co-authored-by: Codex <codex@noreply.local >
2026-06-26 11:53:59 +08:00
84a4b70a26
fix(workbench): scope trace loading indicators ( #797 )
...
Co-authored-by: Codex <codex@noreply.local >
2026-06-24 07:24:18 +08:00
9349f801a6
fix: drive frontend timezone from yaml config
2026-06-20 05:00:35 +00:00
5d484d1ac4
fix: improve wechat collector deploy visibility
2026-06-13 14:54:34 +00:00
19930f56ce
feat: add personal wechat wcf collector deploy
2026-06-13 14:33:56 +00:00
37f48d4655
feat: enforce read-only wechat archive ingestion
2026-06-13 06:22:04 +00:00
c818d0cd44
feat: wire wechat archive through langbot and n8n
2026-06-13 04:32:52 +00:00
b41847853a
fix: restore node resource status sync
2026-06-12 14:18:55 +00:00
f012b4ab81
Merge pull request #267 from pikasTech/fix/issue191-decision-diary-summary
...
fix: preserve decision diary summary
2026-06-11 18:01:43 +08:00
f4d2b72405
fix: harden CI cargo downloads
2026-06-11 09:54:46 +00:00
fc3a095858
fix: preserve decision diary summary
...
Keep diary summary as an independent Decision Center field when body content is supplied from --body-file, and cover the CLI/backend contract.
2026-06-11 17:35:35 +08:00
08a46a22c7
fix: clarify todo-note route misses
2026-06-11 16:50:57 +08:00
0bef6e577a
feat: add marker-only sub2api sentinel reporting
2026-06-11 08:40:43 +00:00
20dda46e03
fix: manage platform-infra network policy
2026-06-10 16:12:34 +00:00
9710a1a6f6
docs: prefer stdin AgentRun queue handoff
2026-06-10 05:32:59 +00:00
5e34f2c2f6
fix: roll sub2api on config changes
2026-06-09 09:50:02 +00:00
24efae9f52
fix: allow sub2api http upstream URLs
2026-06-09 09:45:48 +00:00
80ef698498
feat: add platform infra sub2api deployment
2026-06-09 02:24:17 +00:00
987f41d258
feat: freeze legacy code queue entry
2026-06-09 01:14:23 +00:00
16b0c5659c
fix: drop stale ssh data channels
2026-06-07 10:52:08 +00:00
9e304db0fc
fix: wait for ssh opened before broker stdin
2026-06-07 05:50:32 +00:00
9812c694b1
fix: cache backend core builds and flush ssh output
2026-06-07 04:31:58 +00:00
d56673d62d
fix: exit ssh broker on session termination
2026-06-07 03:55:47 +00:00
b6efa89c9e
fix: initialize provider gateway config constants before read
2026-06-07 03:22:31 +00:00
fe1b91dcbd
feat: add provider ssh tcp data pool
2026-06-07 02:02:38 +00:00
0c28961f9a
feat: add scoped ssh client passthrough
2026-06-02 07:39:21 +00:00
0f62fa6d06
fix(ci): route npm registry direct in docker build NO_PROXY
...
The D601 provider-gateway egress proxy (127.0.0.1:18789) is a
WebSocket-tunneled CONNECT proxy to backend-core. Long npm downloads
(playwright tarball ~3MB, plus tarball streaming) trigger ECONNRESET
mid-install, which then compounds into ENOTEMPTY on the next npm
retry because the partial global install is not cleaned up. The
in-dockerfile retry loop (5 attempts) exhausts and the image build
fails at Step 12/37.
The build's --build-arg NO_PROXY only listed localhost / 127.0.0.1 /
::1 / host.docker.internal, so every npm fetch went through the
proxy. Direct curl tests from D601 to registry.npmjs.org succeed at
HTTP 200 with ~280 KB/s, well above what the WS tunnel can sustain
for multi-MB downloads.
Persist the fix in both D601 and G14 pipeline manifests by adding
registry.npmjs.org and .registry.npmjs.org to NO_PROXY so docker
build's npm install goes direct. The proxy is still used for apt-get
and any other traffic that doesn't match the expanded NO_PROXY list,
so unrelated egress continues through the provider-gateway.
Refs: #189
2026-06-01 08:21:32 +00:00
Codex