Versions (relevant - OpenSearch/Dashboard/Server OS/Browser):
Opensearch 2.19.x
Describe the issue:
We have an ISM that deletes indices when they’re 10 days old. We’ve added “snapshot” step right before the “delete” step. This works just fine, but now we want to manage (i.e. delete old snapshots) as well, there is no option in ISM for that.
I know about SM (snapshotmanagement), where we can state “create” and “delete” policies, but we cannot link it in any way to the ISM, nor can we address the SMpolicy from the ISMpolicy. The ISM created snapshots are also not managed (i.e. not seen/detectable/deleteable ??) by the SMpolicy delete-section …
So, now the question is, how can we have a coherent snapshot and index management this way ? The SMpolicy cron schedule and the ISMpolicy “1d old”-condition are not working nicely together. We want to prevent missing data in the snapshots, but short of having a every_minute SMpolicy cron schedule I see no concrete way of doing this.
I also noticed that AWS snapshot/index management has no real alternative for this according to their documentation : Unleash the power of Snapshot Management to take automated snapshots using Amazon OpenSearch Service | AWS Big Data Blog
How do you people manage this ?
Grtz,
Gert
Configuration:
Relevant Logs or Screenshots: