Loki on ecs crashes when cleaning up chunks

[JanOgr]

Describe the bug I have a case of a loki task shutting down when running on ECS, this occurs when trying to remove a chunk. My questions are how do I configure loki to not crash? And how can I reload the config on a task that is already running (deployment procedures at my company are extremely retarded, i.e I need to try it out before deploying and the issue only occurs in one of the environments).

level=error ts=2024-03-26T12:56:20.496619944Z caller=controller.go:139 component=bloom-compactor org_id=fake table=index_19801 ownership=0000000000000000-ffffffffffffffff msg="failed to delete block" err="error deleting block file: remove /loki/chunks/bloom/index_19801/fake/blocks/1062290f1be392d6-faef15eea9fe36f6/1710859179552-1710888334904-78fa9ef7.tar.gz: no such file or directory" block=bloom/index_19801/fake/blocks/1062290f1be392d6-faef15eea9fe36f6/1710859179552-1710888334904-78fa9ef7.tar.gz

Expected behavior Not to crash **Environment: aws ECS fargate

• Infrastructure: [docker image]
• Deployment tool: [cloudformation]

[JStickler]

Questions have a better chance of being answered if you ask them on the community forums.