Object is stuck in GC cycle #3629
Description
Expected Behavior
GC drops the expired object, it's gone.
Current Behavior
Oct 09 16:09:05 titan4 neofs-node[717444]: info log/log.go:12 local object storage operation {"shard_id": "FvTvbcGePikfcYZGLPsfJe", "address": "754iyTDY8xUtZJZfheSYLUn7jvCkxr79RcbjMt81QykC/96bCoxn8XorTQmVMDrMfi5afmXFSUKUjSGbFGcDUuRaJ", "op": "metabase DELETE"}
Oct 09 16:10:05 titan4 neofs-node[717444]: info log/log.go:12 local object storage operation {"shard_id": "FvTvbcGePikfcYZGLPsfJe", "address": "754iyTDY8xUtZJZfheSYLUn7jvCkxr79RcbjMt81QykC/96bCoxn8XorTQmVMDrMfi5afmXFSUKUjSGbFGcDUuRaJ", "op": "metabase DELETE"}
Oct 09 16:11:05 titan4 neofs-node[717444]: info log/log.go:12 local object storage operation {"shard_id": "FvTvbcGePikfcYZGLPsfJe", "address": "754iyTDY8xUtZJZfheSYLUn7jvCkxr79RcbjMt81QykC/96bCoxn8XorTQmVMDrMfi5afmXFSUKUjSGbFGcDUuRaJ", "op": "metabase DELETE"}
Oct 09 16:12:05 titan4 neofs-node[717444]: info log/log.go:12 local object storage operation {"shard_id": "FvTvbcGePikfcYZGLPsfJe", "address": "754iyTDY8xUtZJZfheSYLUn7jvCkxr79RcbjMt81QykC/96bCoxn8XorTQmVMDrMfi5afmXFSUKUjSGbFGcDUuRaJ", "op": "metabase DELETE"}
Oct 09 16:13:05 titan4 neofs-node[717444]: info log/log.go:12 local object storage operation {"shard_id": "FvTvbcGePikfcYZGLPsfJe", "address": "754iyTDY8xUtZJZfheSYLUn7jvCkxr79RcbjMt81QykC/96bCoxn8XorTQmVMDrMfi5afmXFSUKUjSGbFGcDUuRaJ", "op": "metabase DELETE"}
Possible Solution
Fix something.
Steps to Reproduce (for bugs)
Put an object with expiration, see how it goes after the epoch change.
Context
Testnet. New epoch ticks do not affect the behavior. A restarted node doesn't have this.
Regression
Looks like, #3522.
Your Environment
- Version used: 0.49.0
- Server setup and configuration:
gc:
remover_batch_size: 100
remover_sleep_interval: 1m
- Operating System and version (
uname -a): Debian oldstable