Repository navigation
Route automation queries to the dedicated writer pool - #110
Alex R (alexr17) wants to merge 2 commits into
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
5704af8 to
ca7f345
Compare
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 5704af8ebb
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "Codex (@codex) review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "Codex (@codex) address that feedback".
| BRAINSTORE_WRITER_URL: "http://{{ .Values.brainstore.writer.service.name | default .Values.brainstore.writer.name }}.{{ .Values.global.namespace }}:{{ .Values.brainstore.writer.service.port }}" | ||
| {{- if gt (int (include "braintrust.automationWriter.replicas" .)) 0 }} | ||
| {{- $aw := include "braintrust.automationWriter.config" . | fromYaml }} | ||
| {{- if $aw.routeQueries }} |
There was a problem hiding this comment.
The new raw truthiness check for routeQueries has coverage for true, missing, null, and false, but neither the input matrix nor the tests define what happens for an empty string or 0. Because these shapes directly decide whether the API receives the automation-writer URL, add explicit helm-unittest cases before shipping; the repository treats any unchecked input shape as a release blocker.
AGENTS.md reference: AGENTS.md:L7-L17
Useful? React with 👍 / 👎.
| @@ -0,0 +1,25 @@ | |||
| {{- if gt (int (include "braintrust.automationWriter.replicas" .)) 0 }} | |||
There was a problem hiding this comment.
With positive replicas this template now creates a Service, but braintrust/examples/google-autopilot/values.yaml:154 and braintrust/examples/google-autopilot-cel/values.yaml:174 still explicitly state that no Service is created. Update both supported example comments so operators are not given incorrect resource and networking expectations when enabling this pool.
AGENTS.md reference: AGENTS.md:L9-L12
Useful? React with 👍 / 👎.
|
Codex (@codex) review |
|
Codex Review: Didn't find any major issues. Can't wait for the next one! Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. Codex can also answer questions or update the PR. Try commenting "Codex (@codex) address that feedback". |
1 participant
Automation processing can still compete with regular writers even when its dedicated pool is enabled.
Add a Service and API routing so automation queries use that pool, and wait for it to be ready before starting the API.
routeQueries: falselets operators stage the pool or finish an API rollback before scaling it down.Query routing requires a compatible API release.
--reuse-values; partial replicas-only pool; missing/null Service and annotation mapsv, full/short/all-digit commit SHAs, pre-release and custom tags