Hello,
I encountered a problem with index-index-livecycle-management in elasticseach v7.10.2.
I created an ilm policy named zeek_cleanup with rolling indices settings and applied it to some indices via dev-console. Unfortunately, I applied the wrong settings and removed them with the following command.
PUT rita-*/_settings
{
"index": { "lifecycle": { "name": null } }
}
After that, the command GET rita-*/_ilm/explain indicated that no ilm policies were applied to the index. As a result, I wrote a new policy (wit 'enable rollover' = False) named it gc and activated it for the index with the following command.
PUT rita-*/_settings
{
"index": { "lifecycle": { "name": "gc" } }
}
Now the policy name "policy" : "gc", is correct, but the "phase_execution": setting in the output of GET rita-*/_ilm/explain shows the old policy.
Not sure if the problem is related to the previously enabled rollover, but I think rather less.
I think there is a general problem in the ilm functionality. I hope someone can give me a hint for a solution or reproduce the problem. If the problem persists I would create a github issue.
thank you and greetings