docker: make home wait for solr via healthcheck in compose.override.yaml

[modi02]

Closes #12321

Summary

Replaces the manual Solr polling loop in docker/ol-home-start.sh with
Docker's native healthcheck and depends_on mechanism.

Changes

• Added healthcheck to the solr service in compose.override.yaml
  using the /solr/openlibrary/admin/ping endpoint (consistent with fix: Add wait_for_solr to prevent race condition on startup (#11013) #11570)
• Added depends_on: solr: condition: service_healthy to the home
  service in compose.override.yaml
• Removed the "Waiting for Solr..." polling loop from docker/ol-home-start.sh
• Skipped the postgres bonus task to avoid conflicts with Add docker health checks to compose.override.yaml for postgres and make home wait on db #12322

Testing

Run docker compose up and verify the home service starts only after
Solr is healthy, without the manual polling loop.

Checklist

• healthcheck uses /solr/openlibrary/admin/ping endpoint
• home service depends_on solr with condition: service_healthy
• Solr wait loop removed from ol-home-start.sh

[mekarpeles]

Thank you for this contribution, @modi02!

@RayBB is assigned to this PR and currently has:

• 11 open PR(s) of equal or higher priority to review first

PR triage checklist (maintainers / Pam)

• PR description — not empty; explains what the change does and how to verify it
• References an issue — PR body contains a #NNN reference
  • Linked issue is triaged — has a Priority: * label (not just Needs: Triage)
  • Linked issue is assigned — has at least one assignee
• Commit history clean — no WIP/fixup/conflict noise; commit messages are meaningful
• CI passing — no failing check-runs
• Test cases present — configuration/infrastructure change; no unit tests needed
• Proof of testing — PR body includes a description of how to verify the change

Note

This comment was automatically generated by Pam, Open Library's Project AI Manager, on behalf of @mekarpeles. Pam is designed to provide status visibility, perform basic project management functions and relevant codebase research, and provide actionable feedback so contributors aren't left waiting.

[Copilot]

Pull request overview

Updates the local Docker dev workflow so the home service waits for Solr readiness via Docker Compose healthchecks, replacing the manual Solr polling loop previously embedded in docker/ol-home-start.sh.

Changes:

• Removed the “Waiting for Solr…” polling loop from docker/ol-home-start.sh.
• Added a Solr HTTP healthcheck in compose.override.yaml and made home depend on Solr being healthy before starting.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
docker/ol-home-start.sh	Removes the in-container Solr wait loop so startup sequencing is handled by Compose.
compose.override.yaml	Introduces Solr healthcheck + home depends_on gating on service_healthy.

[Copilot]

healthcheck is indented at the same level as the service definitions (e.g., solr-updater:), so it will be treated as a separate service named healthcheck rather than the solr service’s healthcheck. This will prevent depends_on: condition: service_healthy from working and may make the compose file invalid. Indent healthcheck: (and its fields) so it is under the solr: service block.

[RayBB]

Please restore the new line at the end of the script file and use the health check that we added in the most recent pull request. You may need to branch off the most recent version of master.

https://github.com/internetarchive/openlibrary/pull/11570/changes

[RayBB]

Also, please include screenshots of what happens when you run Docker Compose up and preferably timings before and after.

[RayBB]

Almost there, just need to remove the submodule changes you committed.

Should be possible to do this with make git or follow our git guide on docs.openlibrary.org for more info.

[RayBB]

This is awesome, good job! It cuts startup time by about 24% in my case (8.501s → 6.473s)

[modi02]

Hi @RayBB, glad the improvement was useful! Apologies for the delayed comment, had some trouble attaching the screenshots earlier, but it looks like you were able to verify the fix directly.