Ceph

03/20/2024

02:34 PM RADOS Bug #65008 (New): EC pool - PGs down even if min size is satisfied

Hello I've been evaluating erasure coding ceph setup with following requirements:
- k+m 7+5
- 3 racks
- 5 hosts ... Bartosz Rabiega

03/08/2024

10:19 AM RADOS Bug #62338: osd: choose_async_recovery_ec may select an acting set < min_size

Hello again.
Apparently I got a tiny little bit too excited.
I tested the case described above with 16.2.15 and... Bartosz Rabiega

03/06/2024

07:43 AM RADOS Bug #62338: osd: choose_async_recovery_ec may select an acting set < min_size

Hello. Just FYI, this fixes a very nasty issue in my EC setup.
Here are some details.
The EC setup and crush rule... Bartosz Rabiega

01/31/2017

08:18 AM Ceph Bug #17916: osd/PGLog.cc: 1047: FAILED assert(oi.version == i->first)

Actaully in my case it was enough to delete single faulty object found by strace
Output from strace -e trace=file ... Bartosz Rabiega

01/20/2017

01:18 PM Ceph Bug #17916: osd/PGLog.cc: 1047: FAILED assert(oi.version == i->first)

Managed to find *workaround*:
1. Use strace to find which PG was accessed right before the assertion
2. Move/Remo... Bartosz Rabiega

01/02/2017

12:50 PM Ceph Bug #17916: osd/PGLog.cc: 1047: FAILED assert(oi.version == i->first)

Again this time epochs of objects from divergent priors are quite old (716064 comparing to 1347447) so it looks like ... Bartosz Rabiega

12:17 PM Ceph Bug #17916: osd/PGLog.cc: 1047: FAILED assert(oi.version == i->first)

Hi,
I've been trying to investigate the case from Pawel Sadowski. Using the same OSD data OSD crashes during start... Bartosz Rabiega