opensource/coder
3
0
Fork 0
mirror of https://github.com/coder/coder.git synced 2026-09-24 15:04:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: recover chatd from poisoned chain anchor on retry (#25097)

Browse Source 
When OpenAI's Responses API returns `Previous response with id ... not
found` for a chained turn, classify it as a `ChainBroken` retry, clear
`previous_response_id`, exit chain mode, reload full history, and let
`chatretry` retry. Self-heals chats whose anchor was poisoned before
#25074 stopped truncated streams from being persisted as a successful
turn with a stored response id.

The new state is exposed via the existing
`coderd_chatd_stream_retries_total` counter as a
`chain_broken="true"|"false"` label. Aggregating queries (`sum`, `rate`
over `provider`/`model`/`kind`) keep working without changes; raw-series
matchers without aggregation will now see two series per `(provider,
model, kind)` where they previously saw one. The metric is internal-only
so the blast radius should be small, but if you have dashboards that
index by exact label matchers without aggregation they will need an
extra `sum` or an explicit `chain_broken` selector.

> 🤖 This PR was created with the help of Coder Agents, and was reviewed by a human 🧑‍💻
This commit is contained in:
Cian Johnston
2026-05-11 17:43:40 +01:00
committed by GitHub
parent c2dfaa406a
commit e8508b2d90
8 changed files with 814 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/admin/integrations/prometheus.md
+1 -1
View File
@@ -207,7 +207,7 @@ deployment. They will always be available from the agent.
| `coderd_chatd_stream_buffer_dropped_total` | counter | Number of chat stream buffer events dropped due to the per-chat buffer cap. | |
| `coderd_chatd_stream_buffer_events` | gauge | Sum of current buffer lengths across all chat streams. | |
| `coderd_chatd_stream_buffer_size_max` | gauge | Maximum current buffer length across all chat streams. | |
| `coderd_chatd_stream_retries_total` | counter | Total LLM stream retries. | `kind` `model` `provider` |
| `coderd_chatd_stream_retries_total` | counter | Total LLM stream retries. | `chain_broken` `kind` `model` `provider` |
| `coderd_chatd_stream_subscribers` | gauge | Current number of chat stream subscribers across all chat streams. | |
| `coderd_chatd_streams_active` | gauge | Current number of chat stream state entries (in-flight plus retained). | |
| `coderd_chatd_tool_errors_total` | counter | Total tool calls that returned an error result. | `model` `provider` `tool_name` |