[ceph-users] Mons are using a lot of disk space and has a lot of old osd maps

Aleksei Zakharov zakharov.a.g at yandex.ru
Tue Oct 9 00:35:41 PDT 2018


If someone is interested: we've found a workaround in this mailing list: https://www.spinics.net/lists/ceph-users/msg47963.html
It looks like an old bug.
We fixed the issue by restarting all ceph-mon services one by one. Mon's store uses ~500Mb now and osd's removed old osd maps:
~# find /var/lib/ceph/osd/ceph-224/current/meta/ | wc -l
1839

New osd's have only 1.35GiB used space after first start with no weight.


08.10.2018, 22:31, "Aleksei Zakharov" <zakharov.a.g at yandex.ru>:
> As i can see, all pg's are active+clean:
>
> ~# ceph -s
>   cluster:
>     id: d168189f-6105-4223-b244-f59842404076
>     health: HEALTH_WARN
>             noout,nodeep-scrub flag(s) set
>             mons 1,2,3,4,5 are using a lot of disk space
>
>   services:
>     mon: 5 daemons, quorum 1,2,3,4,5
>     mgr: api1(active), standbys: api2
>     osd: 832 osds: 791 up, 790 in
>          flags noout,nodeep-scrub
>
>   data:
>     pools: 10 pools, 52336 pgs
>     objects: 47.78M objects, 238TiB
>     usage: 854TiB used, 1.28PiB / 2.12PiB avail
>     pgs: 52336 active+clean
>
>   io:
>     client: 929MiB/s rd, 1.16GiB/s wr, 31.85kop/s rd, 36.19kop/s wr
>
> 08.10.2018, 22:11, "Wido den Hollander" <wido at 42on.com>:
>>  On 10/08/2018 05:04 PM, Aleksei Zakharov wrote:
>>>   Hi all,
>>>
>>>   We've upgraded our cluster from jewel to luminous and re-created monitors using rocksdb.
>>>   Now we see, that mon's are using a lot of disk space and used space only grows. It is about 17GB for now. It was ~13GB when we used leveldb and jewel release.
>>>
>>>   When we added new osd's we saw that it downloads from monitors a lot of data. It was ~15GiB few days ago and it is ~18GiB today.
>>>   One of the osd's we created uses filestore and it looks like old osd maps are not removed:
>>>
>>>   ~# find /var/lib/ceph/osd/ceph-224/current/meta/ | wc -l
>>>   73590
>>>
>>>   I've tried to run manual compaction (ceph tell mon.NUM compact) but it doesn't help.
>>>
>>>   So, how to stop this growth of data on monitors?
>>
>>  What is the status of Ceph? Can you post the output of:
>>
>>  $ ceph -s
>>
>>  MONs do not trim their database if one or more PGs aren't active+clean.
>>
>>  Wido
>>
>>>   --
>>>   Regards,
>>>   Aleksei Zakharov
>>>
>>>   _______________________________________________
>>>   ceph-users mailing list
>>>   ceph-users at lists.ceph.com
>>>   http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>
> --
> Regards,
> Aleksei Zakharov
>
> _______________________________________________
> ceph-users mailing list
> ceph-users at lists.ceph.com
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

-- 
Regards,
Aleksei Zakharov



More information about the ceph-users mailing list