General

Profile

Vitaliy Filippov

Issues

Activity

08/19/2020

12:53 PM Ceph Bug #46969: Octopus OSDs deadlock with slow ops and make the whole cluster unresponsive
Oops, sorry, there was one more change - I changed shards*threads to 1*16 from default 2*8:
osd_op_num_threads_per...

08/18/2020

11:04 PM bluestore Bug #44731: Space leak in Bluestore
An update here: this was caused by broken compression in Ubuntu builds around 14.2.6-7-8 or so. Old data was compress...
08:34 PM Ceph Bug #46969: Octopus OSDs deadlock with slow ops and make the whole cluster unresponsive
It *seems* the problem has gone away after removing the following non-default variables from the configuration:
#b...

08/14/2020

08:04 AM Ceph Bug #46969 (New): Octopus OSDs deadlock with slow ops and make the whole cluster unresponsive
Hi,
I have another unpleasant bug to report.
Right after I upgraded my cluster to Octopus 15.2.4 I started to e...

07/31/2020

09:26 PM bluestore Bug #46800: Octopus OSD died and fails to start with FAILED ceph_assert(is_valid_io(off, len))
I had bluestore_min_alloc_size_ssd = 4096 before it became the default, but it seems that's all
I've copied RocksD...
03:32 PM bluestore Bug #46800: Octopus OSD died and fails to start with FAILED ceph_assert(is_valid_io(off, len))
Oh, I have one more thing to add. I tried to ran fsck and it gave a slightly different error message
03:16 PM bluestore Bug #46800: Octopus OSD died and fails to start with FAILED ceph_assert(is_valid_io(off, len))
Sorry, too late... :)
01:54 PM bluestore Bug #46800 (New): Octopus OSD died and fails to start with FAILED ceph_assert(is_valid_io(off, len))
Hi
One of my OSDs just died trying to write beyond the end of the device. Now it just fails to start with the same...

03/24/2020

04:50 PM bluestore Bug #44731 (Closed): Space leak in Bluestore
Hi.
I'm experiencing some kind of a space leak in Bluestore. I use EC, compression and snapshots. First I thought ...

06/15/2019

09:36 AM Ceph Revision 47cd1afa (ceph): osd/bluestore: Actually wait until completion in write_sync
This function is only used by RocksDB WAL writing so it must sync data.
This fixes #18338 and thus allows to actuall...

Also available in: Atom