Fix Missing IgnoredUnavailable Flag in 7.x SLM Retention Task (#56616)
Without the flag we run into the situation where a broken repository (broken by some old 6.x version of ES that is missing some snap-${uuid}.dat blobs fails to run the SLM retention task since it always errors out).
This commit is contained in:
parent
4240b97d0e
commit
c104c9a11b
|
@ -270,6 +270,7 @@ public class SnapshotRetentionTask implements SchedulerEngine.Listener {
|
|||
for (String repository : repositories) {
|
||||
client.admin().cluster()
|
||||
.prepareGetSnapshots(repository)
|
||||
.setIgnoreUnavailable(true)
|
||||
.execute(ActionListener.wrap(resp -> {
|
||||
final Set<SnapshotState> retainableStates =
|
||||
new HashSet<>(Arrays.asList(SnapshotState.SUCCESS, SnapshotState.FAILED, SnapshotState.PARTIAL));
|
||||
|
|
Loading…
Reference in New Issue