All of lore.kernel.org
 help / color / mirror / Atom feed
* clean shutdown and failover of osd
@ 2013-04-21  5:15 James Harper
       [not found] ` <6035A0D088A63A46850C3988ED045A4B4F360D32-mzsoxcrO4/2UD0RQwgcqbDSf8X3wrgjD@public.gmane.org>
  0 siblings, 1 reply; 5+ messages in thread
From: James Harper @ 2013-04-21  5:15 UTC (permalink / raw)
  To: ceph-devel

I'm doing some testing with ceph trying to figure out why my performance is so bad, and have noticed that there doesn't seem to be a way to cleanly stop an osd, or at least under debian /etc/init.d/ceph stop seems to just kill the OSD resulting in the client also stopping io until it figures out why the node isn't responding.

Is this how it is supposed to work? I would have thought it would be possible to shut down an osd cleanly such that failover is immediate. I understand that an abrupt failure like a node crashing would result in a freeze for a bit.

Thanks

James


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

* Re: clean shutdown and failover of osd
       [not found] ` <6035A0D088A63A46850C3988ED045A4B4F360D32-mzsoxcrO4/2UD0RQwgcqbDSf8X3wrgjD@public.gmane.org>
@ 2013-04-21  5:44   ` Gregory Farnum
       [not found]     ` <CAPYLRzjU-BsA16rv+zSgR8ngcOm5L-tZ54kMNq=DrZWZr+VPFQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
  0 siblings, 1 reply; 5+ messages in thread
From: Gregory Farnum @ 2013-04-21  5:44 UTC (permalink / raw)
  To: James Harper; +Cc: ceph-devel-u79uwXL29TY76Z2rM5mHXA, ceph-users

[ This is a good query for ceph-users. ]

On Sat, Apr 20, 2013 at 10:15 PM, James Harper
<james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
> I'm doing some testing with ceph trying to figure out why my performance is so bad, and have noticed that there doesn't seem to be a way to cleanly stop an osd, or at least under debian /etc/init.d/ceph stop seems to just kill the OSD resulting in the client also stopping io until it figures out why the node isn't responding.
>
> Is this how it is supposed to work? I would have thought it would be possible to shut down an osd cleanly such that failover is immediate. I understand that an abrupt failure like a node crashing would result in a freeze for a bit.

In bobtail that's about all there is, yes. If you're in control you
can shut it down and issue a "ceph osd down" command to tell the
system without having to wait for heartbeats to notice it.

The current next branch (will be Cuttlefish) includes code so that
OSDs attempt to notify the system before they get turned off.
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com

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

* Re: clean shutdown and failover of osd
       [not found]     ` <CAPYLRzjU-BsA16rv+zSgR8ngcOm5L-tZ54kMNq=DrZWZr+VPFQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
@ 2013-04-21  5:51       ` James Harper
       [not found]         ` <6035A0D088A63A46850C3988ED045A4B4F360DEF-mzsoxcrO4/2UD0RQwgcqbDSf8X3wrgjD@public.gmane.org>
  0 siblings, 1 reply; 5+ messages in thread
From: James Harper @ 2013-04-21  5:51 UTC (permalink / raw)
  To: Gregory Farnum; +Cc: ceph-devel-u79uwXL29TY76Z2rM5mHXA, ceph-users

> 
> [ This is a good query for ceph-users. ]
> 

Well... this is embarrassing. In reading the docs at http://ceph.com/docs/master/start/get-involved/ there was no mention of a users list so I just assumed there wasn't one. Looking again I see that if I go to the link from the main page http://ceph.com/resources/mailing-list-irc/ it's obviously there and I just never looked there.

Sorry for the noise. Maybe the web/docs admin could add a link to the get-involved docs page?

James

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

* Re: clean shutdown and failover of osd
       [not found]         ` <6035A0D088A63A46850C3988ED045A4B4F360DEF-mzsoxcrO4/2UD0RQwgcqbDSf8X3wrgjD@public.gmane.org>
@ 2013-04-21  5:53           ` Gregory Farnum
       [not found]             ` <CAPYLRzhCrG3_iwC1TeFm=K3yowx38SbVfk8jTzGn+qq_zbYcGw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
  0 siblings, 1 reply; 5+ messages in thread
From: Gregory Farnum @ 2013-04-21  5:53 UTC (permalink / raw)
  To: James Harper, John Wilkins; +Cc: ceph-devel-u79uwXL29TY76Z2rM5mHXA, ceph-users

On Sat, Apr 20, 2013 at 10:51 PM, James Harper
<james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
>>
>> [ This is a good query for ceph-users. ]
>>
>
> Well... this is embarrassing. In reading the docs at http://ceph.com/docs/master/start/get-involved/ there was no mention of a users list so I just assumed there wasn't one. Looking again I see that if I go to the link from the main page http://ceph.com/resources/mailing-list-irc/ it's obviously there and I just never looked there.
>
> Sorry for the noise. Maybe the web/docs admin could add a link to the get-involved docs page?

Haha, yep. Thanks for pointing that out — John? :)
-Greg

Software Engineer #42 @ http://inktank.com | http://ceph.com

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

* Re: clean shutdown and failover of osd
       [not found]             ` <CAPYLRzhCrG3_iwC1TeFm=K3yowx38SbVfk8jTzGn+qq_zbYcGw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
@ 2013-04-22 20:41               ` John Wilkins
  0 siblings, 0 replies; 5+ messages in thread
From: John Wilkins @ 2013-04-22 20:41 UTC (permalink / raw)
  To: Gregory Farnum; +Cc: ceph-devel-u79uwXL29TY76Z2rM5mHXA, ceph-users


[-- Attachment #1.1: Type: text/plain, Size: 1046 bytes --]

Will do.


On Sat, Apr 20, 2013 at 10:53 PM, Gregory Farnum <greg-4GqslpFJ+cxBDgjK7y7TUQ@public.gmane.org> wrote:

> On Sat, Apr 20, 2013 at 10:51 PM, James Harper
> <james.harper-NMzNsA1hOHcW+bLBXbPJGg@public.gmane.org> wrote:
> >>
> >> [ This is a good query for ceph-users. ]
> >>
> >
> > Well... this is embarrassing. In reading the docs at
> http://ceph.com/docs/master/start/get-involved/ there was no mention of a
> users list so I just assumed there wasn't one. Looking again I see that if
> I go to the link from the main page
> http://ceph.com/resources/mailing-list-irc/ it's obviously there and I
> just never looked there.
> >
> > Sorry for the noise. Maybe the web/docs admin could add a link to the
> get-involved docs page?
>
> Haha, yep. Thanks for pointing that out — John? :)
> -Greg
>
> Software Engineer #42 @ http://inktank.com | http://ceph.com
>



-- 
John Wilkins
Senior Technical Writer
Intank
john.wilkins-4GqslpFJ+cxBDgjK7y7TUQ@public.gmane.org
(415) 425-9599
http://inktank.com

[-- Attachment #1.2: Type: text/html, Size: 1938 bytes --]

[-- Attachment #2: Type: text/plain, Size: 178 bytes --]

_______________________________________________
ceph-users mailing list
ceph-users-idqoXFIVOFJgJs9I8MT0rw@public.gmane.org
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

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

end of thread, other threads:[~2013-04-22 20:41 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-04-21  5:15 clean shutdown and failover of osd James Harper
     [not found] ` <6035A0D088A63A46850C3988ED045A4B4F360D32-mzsoxcrO4/2UD0RQwgcqbDSf8X3wrgjD@public.gmane.org>
2013-04-21  5:44   ` Gregory Farnum
     [not found]     ` <CAPYLRzjU-BsA16rv+zSgR8ngcOm5L-tZ54kMNq=DrZWZr+VPFQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-04-21  5:51       ` James Harper
     [not found]         ` <6035A0D088A63A46850C3988ED045A4B4F360DEF-mzsoxcrO4/2UD0RQwgcqbDSf8X3wrgjD@public.gmane.org>
2013-04-21  5:53           ` Gregory Farnum
     [not found]             ` <CAPYLRzhCrG3_iwC1TeFm=K3yowx38SbVfk8jTzGn+qq_zbYcGw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-04-22 20:41               ` John Wilkins

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.