yifan-c commented on code in PR #346:
URL: https://github.com/apache/cassandra-sidecar/pull/346#discussion_r3271685995
##########
server/src/main/java/org/apache/cassandra/sidecar/restore/RestoreJobDiscoverer.java:
##########
@@ -628,6 +628,22 @@ Map<Integer, Map<UUID, RestoreJobStatus>> jobsByDay()
}
}
+ /**
+ * Immediately processes a restore job without waiting for the next
discovery loop iteration.
+ * Called by UpdateRestoreJobHandler after a phase signal (STAGE_READY) is
written to DB.
+ * This is safe to call concurrently with the discovery loop — the DB
write is the durable
+ * source of truth, and duplicate processing is deduplicated by existing
idempotency checks.
+ *
+ * @param restoreJob the restore job to process immediately
+ */
+ public void processJobNow(RestoreJob restoreJob)
+ {
+ initLocalDatacenterMaybe();
+ RestoreJobManagerGroup restoreJobManagers =
restoreJobManagerGroupSingleton.get();
+ restoreJobManagers.updateRestoreJob(restoreJob);
+ processSidecarManagedJobMaybe(restoreJob);
Review Comment:
Good catch. A simple/clean way to address it is to wrap the lines in
`processJobNow` with the same `isExecuting / finally` pattern used in
`tryExecuteDiscovery`. The CAS operation on `isExecuting` provides a
happens-before edge.
--
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]
