Scrubbing

All of lore.kernel.org
 help / color / mirror / Atom feed

* Scrubbing
@ 2011-04-21 15:56 Zenon Panoussis
       [not found] ` <4DB05926.6020406@hq.newdream.net>
  0 siblings, 1 reply; 5+ messages in thread
From: Zenon Panoussis @ 2011-04-21 15:56 UTC (permalink / raw)
  To: ceph-devel

Hi

Two OSDs, 264 pgs on each. This is killing the OSD:

2011-04-21 17:49:52.673431    pg v102789: 528 pgs: 299 active+clean, 229 active+clean+scrubbing; 5675 MB data, 26696 MB used, 395 GB / 446 GB avail

Is there some way to tell it to favour serial scrubbing over
parallel?

Z



^ permalink raw reply	[flat|nested] 5+ messages in thread

[parent not found: <4DB05926.6020406@hq.newdream.net>]

[parent not found: <4DB06BFC.9010207@provocation.net>]

* Re: Scrubbing
       [not found]   ` <4DB06BFC.9010207@provocation.net>
@ 2011-04-21 23:30     ` Samuel Just
  0 siblings, 0 replies; 5+ messages in thread
From: Samuel Just @ 2011-04-21 23:30 UTC (permalink / raw)
  To: ceph-devel

  Hmm, it's supposed to throttle background scrubbing.  Could we get 
access to the osd logs?

Thanks,
-Sam

On 04/21/2011 10:40 AM, Zenon Panoussis wrote:
> On 04/21/2011 06:19 PM, Samuel Just wrote:
>
> Hi
>
>> Well, that is definitely not supposed to happen, sorry about
>> that!  Did you request the scrubs or did it start them automatically?
> I had my one monitor mounted by a client. I said "ceph osd scrub 0"
> and it behaved quite nicely, scrubbing two or three pgs at a time
> (I have default osd threads). Then I did "umount mon:/" on the
> client and as soon as umount returned (it took a few seconds) the
> scrubbing exploded.
>
> It's 0.26 tarball release BTW.
>
> Z
>


^ permalink raw reply	[flat|nested] 5+ messages in thread

* scrubbing
@ 2012-01-02 11:10 Jens Rehpöhler
  2012-01-02 11:13 ` scrubbing Wido den Hollander
  0 siblings, 1 reply; 5+ messages in thread
From: Jens Rehpöhler @ 2012-01-02 11:10 UTC (permalink / raw)
  To: ceph-devel

[-- Attachment #1: Type: text/plain, Size: 1563 bytes --]

Happy new year  !!

another year another question:

I got the following from ceph -w :

2012-01-02 12:05:17.294350    pg v830384: 1254 pgs: 18 creating, 1235
active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
GB / 14887 GB avail
2012-01-02 12:05:23.763982    pg v830385: 1254 pgs: 18 creating, 1235
active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
GB / 14887 GB avail
2012-01-02 12:05:27.295568    pg v830386: 1254 pgs: 18 creating, 1235
active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
GB / 14887 GB avail
2012-01-02 12:05:28.765457    pg v830387: 1254 pgs: 18 creating, 1235
active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
GB / 14887 GB avail

ceph pg dump | grep scrub:
pg_stat    objects    mip    degr    unf    kb    bytes    log   
disklog    state    v    reported    up    acting    last_scrub
34.2    1583    0    0    0    6125554    6272545912    3319    3319   
active+clean+scrubbing    933'385350    926'63653    [3,4]    [3,4]   
927'337222    2011-12-27 13:20:42.436092

This happens once a week. If i restart ceph i takes a moment an the
"scrubbing" disappears.

-- 
mit freundlichen Grüssen

Jens Rehpöhler

----------------------------------------------------------------------
Filoo GmbH
Moltkestr. 25a
33330 Gütersloh
HRB4355 AG Gütersloh

Geschäftsführer: S.Grewing | J.Rehpöhler | C.Kunz
Telefon: +49 5241 8673012 | Mobil: +49 151 54645798
Hotline: 07000-3378658 (14 Ct/min) Fax: +49 5241 8673020



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 262 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: scrubbing
  2012-01-02 11:10 scrubbing Jens Rehpöhler
@ 2012-01-02 11:13 ` Wido den Hollander
       [not found]   ` <4F019647.50808@filoo.de>
  0 siblings, 1 reply; 5+ messages in thread
From: Wido den Hollander @ 2012-01-02 11:13 UTC (permalink / raw)
  To: Jens Rehpöhler; +Cc: ceph-devel

Hi,

On 01/02/2012 12:10 PM, Jens Rehpöhler wrote:
> Happy new year  !!

Let this be a great year for Ceph :-)

>
> another year another question:
>
> I got the following from ceph -w :
>
> 2012-01-02 12:05:17.294350    pg v830384: 1254 pgs: 18 creating, 1235
> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
> GB / 14887 GB avail
> 2012-01-02 12:05:23.763982    pg v830385: 1254 pgs: 18 creating, 1235
> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
> GB / 14887 GB avail
> 2012-01-02 12:05:27.295568    pg v830386: 1254 pgs: 18 creating, 1235
> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
> GB / 14887 GB avail
> 2012-01-02 12:05:28.765457    pg v830387: 1254 pgs: 18 creating, 1235
> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
> GB / 14887 GB avail

Which version are you running?

This was a known bug but should be fixed about two weeks ago in the 
master branch.

I also encountered this one, but it's resolved for me right now.

Wido

>
> ceph pg dump | grep scrub:
> pg_stat    objects    mip    degr    unf    kb    bytes    log
> disklog    state    v    reported    up    acting    last_scrub
> 34.2    1583    0    0    0    6125554    6272545912    3319    3319
> active+clean+scrubbing    933'385350    926'63653    [3,4]    [3,4]
> 927'337222    2011-12-27 13:20:42.436092
>
> This happens once a week. If i restart ceph i takes a moment an the
> "scrubbing" disappears.
>

--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 5+ messages in thread

[parent not found: <4F019647.50808@filoo.de>]

* Re: scrubbing
       [not found]   ` <4F019647.50808@filoo.de>
@ 2012-01-03  9:51     ` Wido den Hollander
  0 siblings, 0 replies; 5+ messages in thread
From: Wido den Hollander @ 2012-01-03  9:51 UTC (permalink / raw)
  To: Jens Rehpöhler; +Cc: ceph-devel

Hi,

On 01/02/2012 12:34 PM, Jens Rehpöhler wrote:
> Hi Wido,
>
> Am 02.01.2012 12:13, schrieb Wido den Hollander:
>> Hi,
>>
>> On 01/02/2012 12:10 PM, Jens Rehpöhler wrote:
>>> Happy new year  !!
>>
>> Let this be a great year for Ceph :-)
>>
> i think it will be ;)
>>>
>>> another year another question:
>>>
>>> I got the following from ceph -w :
>>>
>>> 2012-01-02 12:05:17.294350    pg v830384: 1254 pgs: 18 creating, 1235
>>> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
>>> GB / 14887 GB avail
>>> 2012-01-02 12:05:23.763982    pg v830385: 1254 pgs: 18 creating, 1235
>>> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
>>> GB / 14887 GB avail
>>> 2012-01-02 12:05:27.295568    pg v830386: 1254 pgs: 18 creating, 1235
>>> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
>>> GB / 14887 GB avail
>>> 2012-01-02 12:05:28.765457    pg v830387: 1254 pgs: 18 creating, 1235
>>> active+clean, 1 active+clean+scrubbing; 586 GB data, 1378 GB used, 12763
>>> GB / 14887 GB avail
>>
>> Which version are you running?
>>
>> This was a known bug but should be fixed about two weeks ago in the
>> master branch.
>>
>> I also encountered this one, but it's resolved for me right now.
>>
> I use:
>
> ceph version 0.39-15-ge93961c
> (commit:e93961c11119942eae3a4cd14a79f779a5a4d277)
>
> from the "stable" branch.
>
> OS is Debian 6.0.3
>
> So i will wait until it is patched into the stable branch too.

I think it won't be to long until v0.40 comes out. But in the meantime 
you can always try to grab the latest code from the master branch.

Wido

>
> Thank you for the very quick answer.
>

--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2012-01-03  9:51 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-04-21 15:56 Scrubbing Zenon Panoussis
     [not found] ` <4DB05926.6020406@hq.newdream.net>
     [not found]   ` <4DB06BFC.9010207@provocation.net>
2011-04-21 23:30     ` Scrubbing Samuel Just
  -- strict thread matches above, loose matches on Subject: below --
2012-01-02 11:10 scrubbing Jens Rehpöhler
2012-01-02 11:13 ` scrubbing Wido den Hollander
     [not found]   ` <4F019647.50808@filoo.de>
2012-01-03  9:51     ` scrubbing Wido den Hollander

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.