Ceph

01/31/2024

10:04 AM rgw Bug #37668: AbortMultipartUpload causes data loss(NoSuchKey) when CompleteMultipartUpload request timeout

We also encountered a similar problem on v16.2.10.
The s3 index object exists, but it return 404 error when getting ... xiaobao wen

12/27/2023

07:48 AM mgr Bug #59580: memory leak (RESTful module, maybe others?)

Hello.
We suspect we have encountered similar problems.
About 370GB of memory was consumed in five minutes. Oom ... xiaobao wen

11/27/2023

06:04 AM rgw Bug #63642: rgw: rados objects wrongly deleted

rados object list. shadow_xxxxx.2_3, 2_4 lost... xiaobao wen

06:00 AM rgw Bug #63642: rgw: rados objects wrongly deleted

We checked the s3 user's logs. The multipart upload retries were automatically attempted by the s3-transfer SDK.
W... xiaobao wen

03:49 AM rgw Bug #63642: rgw: rados objects wrongly deleted

We encountered data loss when using multipart upload. We found that some rados objects were lost.
h2. Logs on prod... xiaobao wen

03:23 AM rgw Bug #63642 (Pending Backport): rgw: rados objects wrongly deleted

xiaobao wen

10/23/2023

03:50 AM Ceph Documentation #63285 (New): update document for `DEBUGGING SLOW REQUESTS`

When we tried to solve the problem of slow requests in the new cluster, we found that the official website documentat... xiaobao wen

08/18/2023

06:08 AM rgw Bug #61763: RGW issue sync loop restart master zone

last ~100 lines of log xiaobao wen

08/17/2023

03:25 AM rgw Bug #61763: RGW issue sync loop restart master zone

Casey Bodley wrote:
> it's hard to tell what the problem is when all of the rgw logging is disabled by @0/ 0 rgw@. c... xiaobao wen

08/10/2023

09:20 AM mgr Bug #62385: mgr/status: wrong kb_used and kb_avail units in command `ceph osd status` results

https://github.com/ceph/ceph/pull/52856 xiaobao wen