Bug #18258
rgw: radosgw-admin orphan find goes into infinite loop
0%
Description
The job is running 6 days already:
radosgw-admin orphans find --pool=.rgw.buckets --job-id=dev1
I think job is stuck in a loop, because i can see the same number of linked objects in same shards, for example:
storing 12 entries at orphan.scan.dev1.linked.63
...
storing 8 entries at orphan.scan.dev1.linked.63
...
storing 15 entries at orphan.scan.dev1.linked.63
..
storing 19 entries at orphan.scan.dev1.linked.63
...
storing 12 entries at orphan.scan.dev1.linked.63
...
On another cluster the job is running for 5 days already. Same ceph version 10.2.2.
Related issues
History
#1 Updated by Yehuda Sadeh over 7 years ago
- Subject changed from Loop in radosgw-admin orphan find to rgw: radosgw-admin orphan find goes into infinite loop
- Assignee set to Matt Benjamin
- Priority changed from Normal to High
#2 Updated by Marius Vaitiekunas over 7 years ago
Some more details.
We've upgraded rgw's to 10.2.5, but the job gets into loop anyway. At the moment we have to rgw's running
Our cluster has many leaked objects - could be 20 000 000, it's about 50TB of data.
#3 Updated by Wido den Hollander about 7 years ago
- Affected Versions v10.2.5 added
- Affected Versions deleted (
v10.2.2)
#4 Updated by Nathan Cutler about 7 years ago
- Backport set to jewel
#5 Updated by Yehuda Sadeh about 7 years ago
- Status changed from New to Pending Backport
- Backport changed from jewel to jewel, kraken
#6 Updated by Loïc Dachary about 7 years ago
- Copied to Backport #18912: jewel: rgw: radosgw-admin orphan find goes into infinite loop added
#7 Updated by Loïc Dachary about 7 years ago
- Copied to Backport #18913: kraken: rgw: radosgw-admin orphan find goes into infinite loop added
#8 Updated by Nathan Cutler about 7 years ago
- Duplicates Bug #18331: RGW leaking data added
#9 Updated by Nathan Cutler about 7 years ago
- Status changed from Pending Backport to Duplicate