benediktarnold commented on issue #824:
URL: https://github.com/apache/solr-operator/issues/824#issuecomment-4857175120

   Confirming this on the **latest release v0.9.1** (Solr 9.7, GCS repo), and 
adding
   the impact on **recurring** backups, which is severe.
   
   Our operator runs on GKE Autopilot and is restarted routinely (node
   auto-provisioning/upgrades), so the precondition described here — backup 
accepted,
   operator unavailable, async status later returns `notfound` — is easy to hit
   without any manual intervention.
   
   The key thing to add: for a **recurring** `SolrBackup`, a single stuck
   `notfound`/`inProgress` collection wedges the *entire* recurrence, not just 
that
   collection:
   
   - top-level `finished` and `nextScheduled` stay `null`,
   - the schedule never fires again and `maxSaved` pruning never runs,
   - every other collection still reports `successful: true`, and no error 
surfaces.
   
   Ours has been frozen at its original `startTimestamp`, with a few 
per-collection
   statuses stuck in `notfound`/`inProgress`. The schedule was even changed 
during
   that window with no effect, because the recurrence never advanced past the 
wedged
   run.
   
   Suggestion, in addition to making `notfound` terminal: compute
   `finished`/`nextScheduled` from *terminal* per-collection states, so one 
stuck
   collection can't halt the schedule and `maxSaved` retention indefinitely.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to