Hello,
No new information. Every two night server OSD 1 freeze with a load > 500.
It's every 2 days. Sometime during scrub, sometime during fstrim,
sometime during nothing...
But this night, this OSD server came not a life after some minutes as
before... 8 hours without this server and all its OSD (12/36).
This morning, I restart it and now after some hours :
HEALTH_WARN 1 pgs degraded; 1 pgs recovering; 1 pgs stuck unclean;
recovery 304/46002595 objects degraded (0.001%); recovery 11288/46002595
objects misplaced (0.025%); recovery 3/9779473 unfound (0.000%)
pg 50.2dd is stuck unclean for 23531.224308, current state
active+recovering+degraded+remapped, last acting [7,28]
pg 50.2dd is active+recovering+degraded+remapped, acting [7,28], 3 unfound
recovery 304/46002595 objects degraded (0.001%)
recovery 11288/46002595 objects misplaced (0.025%)
recovery 3/9779473 unfound (0.000%)
Pool 50.2dd is a RBD filesystem, XFS with replicat 2x.
So what is the best solution ?
#ceph pg 50.2dd mark_unfound_lost delete
or
#ceph pg 50.2dd mark_unfound_lost revert
?
What can have more impact to RBD/XFS filesystem ? a xfs_repair required
after ?
So, I will probably try ceph version 10.2.6 this evening because I
really found nothing to fix...
Why this freeze ? why only this server OSD freeze and not others ? why
every 2 days ? It's crazy.
I already checked all : disk, network, soft, all servers are equals.
(all issues started the day after upgrade to 10.2.5 from 10.2.3).
Thanks for your help.
Regards,
Le 02/03/2017 à 15:34, [email protected] a écrit :
Hello,
So, I need maybe some advices : 1 week ago (last 19 feb), I upgraded
my stable Ceph Jewel from 10.2.3 to 10.2.5 (YES, It was maybe a bad idea).
I never had problem with Ceph 10.2.3 since last upgrade, last 23
September.
So since my upgrade (10.2.5), every 2 days, the first OSD server
totaly Freeze. Load go > 500 and come back after somes minutes… I lost
all OSD from this server (12/36) during issue.
[...]
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
