General

Profile

Stefan Priebe

Issues

Activity

04/28/2020

05:11 PM bluestore Bug #44509: using ceph-bluestore-tool bluefs-bdev-new-db results in 128k or 64k leftover on slow ...
Thanks a lot. That solves the problem. Would be good to have a note about it in docs.

04/24/2020

06:44 PM bluestore Bug #44509: using ceph-bluestore-tool bluefs-bdev-new-db results in 128k or 64k leftover on slow ...
Igor did you had a look at those values? Any chance to fix this manually?
06:43 PM bluestore Bug #42928: ceph-bluestore-tool bluefs-bdev-new-db does not update lv tags
OK i was able to solve this by using the lv tools and copy all lv flags from the block device to db and just chaning ...

04/21/2020

05:51 PM bluestore Bug #44509: using ceph-bluestore-tool bluefs-bdev-new-db results in 128k or 64k leftover on slow ...
sure:
```
# ceph-kvstore-tool bluestore-kv /var/lib/ceph/osd/ceph-5 stats
2020-04-21 19:51:16.635 7f439d4b7e00 1 ...
02:33 PM bluestore Bug #42928: ceph-bluestore-tool bluefs-bdev-new-db does not update lv tags
is ceph.db_uuid the LVM uuid?
02:14 PM bluestore Bug #42928: ceph-bluestore-tool bluefs-bdev-new-db does not update lv tags
Those two tags are missing:
ceph.db_device=/dev/BLOCKDB
ceph.db_uuid=k7Oc64-aShi-Z4Y0-Opov-daQM-Vdue-kJpIAN

03/08/2020

06:50 PM bluestore Bug #44509 (Won't Fix): using ceph-bluestore-tool bluefs-bdev-new-db results in 128k or 64k lefto...
Atfer using bluefs-bdev-new-db bluefs-bdev-new-db on 24 osds. All of those OSDs are in the following state even after...

08/21/2019

07:20 AM bluestore Bug #41367 (Duplicate): rocksdb: submit_transaction error: Corruption: block checksum mismatch co...
Since several weeks we have daily 1-8 OSD crashes since ceph 12.2.12.
The OSD log is attached.
```
Trace:
201...

06/19/2019

09:17 AM Ceph Revision 8aea71a1 (ceph): compressor/zstd: improvements
- static linking #define to get the advanced stuff
- hint src size to compressor
- no flush needed before end
- check...

10/30/2018

02:49 PM bluestore Bug #36567: Segmentation fault in BlueStore::Blob::discard_unallocated
Not that good ;-) it always happen, when we trigger a heavy backfill or recovery. But i don't want to pull that many ...

Also available in: Atom