It's just because the PG hadn't been scrubbed since the error occurred;
then you upgraded, it scrubbed, and the error was found. You can deep-scrub
all your PGs to check them if you like, but as I've said elsewhere this
issue -- while scary! -- shouldn't actually damage any of your user data,
so ju
Greg,
This error occurred AFTER the upgrade. I upgraded to 0.80.4 last night and
this error cropped up this afternoon. I ran `ceph pg repair 3.7f` (after I
copied the pgs) which returned the cluster to health. However, I'm
concerned that this showed up again so soon after I upgraded to 0.80.4.
Is
Keep in mind that this has thrown out all the auth info in your
cluster, so if you ever do enable cephx you'll need to re-assign all
the keys. And you might be in line for some other strangeness as well
that I haven't foreseen down the line.
In the meanwhile, I've forwarded things on to our full-t
Thanks Greg. I appreciate the advice, and very quick replies too :)
On 18 July 2014 23:35, Gregory Farnum wrote:
> On Fri, Jul 18, 2014 at 3:29 PM, James Eckersall
> wrote:
> > Thanks Greg.
> >
> > Can I suggest that the documentation makes this much clearer? It might
> just be me, but I cou
Thanks Greg. Just for posterity, "ceph-kvstore-tool /var/lib/ceph/mon/store.db
set auth last_committed ver 0" did the trick and we're back to HEALTH_OK.
Cheers,
Lincoln Bryant
On Jul 18, 2014, at 4:15 PM, Gregory Farnum wrote:
> Hmm, this log is just leaving me with more questions. Could you ta
On Fri, Jul 18, 2014 at 3:29 PM, James Eckersall
wrote:
> Thanks Greg.
>
> Can I suggest that the documentation makes this much clearer? It might just
> be me, but I couldn't glean this from the docs, so I expect I'm not the only
> one.
>
> Also, can I clarify how many pg's you would suggest is
Thanks Greg.
Can I suggest that the documentation makes this much clearer? It might just be
me, but I couldn't glean this from the docs, so I expect I'm not the only one.
Also, can I clarify how many pg's you would suggest is a decent number for my
setup?
80 OSD's across 4 nodes. 5 pools.
I'
Yes, that's expected behavior. Since the cluster can't move data
around on its own, and lots of things will behave *very badly* if some
of their writes go through but others don't, the cluster goes
read-only once any OSD is full. That's why nearfull is a warn
condition; you really want to even out
Hi,
I have a ceph cluster running on 0.80.1 with 80 OSD's.
I've had fairly uneven distribution of the data and have been keeping it
ticking along with "ceph osd reweight XX 0.x" commands on a few OSD's while
I try and increase the pg count of the pools to hopefully better balance
the data.
Tonig
The config option change in the upgrade will prevent *new* scrub
errors from occurring, but it won't resolve existing ones. You'll need
to run a scrub repair to fix those up.
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com
On Fri, Jul 18, 2014 at 2:59 PM, Randy Smith wrote:
>
Greetings,
I upgraded to 0.80.4 last night to resolve the inconsistent pg scrub errors
I was seeing. Unfortunately, they are continuing.
$ ceph health detail
HEALTH_ERR 1 pgs inconsistent; 1 scrub errors
pg 3.7f is active+clean+inconsistent, acting [0,4]
And here's the relevant log entries.
201
Hmm, this log is just leaving me with more questions. Could you tar up
the "/var/lib/ceph/mon/store.db" (substitute actual mon store path as
necessary) and upload it for me? (you can use ceph-post-file to put it
on our servers if you prefer.) Just from the log I don't have a great
idea of what's go
You can change the pools configuration using for example one of this two
options:
1. define a different crush_ruleset for the pools to allow replicas to be
placed on the same host (so 3 replicas can be accommodated on two hosts).
2. set the number of replicas to two
Option 2 is very easy:
$ c
There's nothing built in to the system but I think some people have
had success with scripts that set nobackfill during the day, and then
trigger them regularly at night. Try searching the list archives. :)
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com
On Fri, Jul 18, 2014 at
On 07/17/2014 03:07 PM, Dmitry Borodaenko wrote:
> The meeting is in 2 hours, so you still have a chance to particilate
> or at least lurk :)
Note that this spec has 4 members of nova-core sponsoring it for an
exception on the etherpad tracking potential exceptions. We'll review
further in the me
Hi Sage,
A little favour -- It will be helpful if you could CC release announcements to
ceph-maintain...@lists.ceph.com
please. It is much easier to notice there comparing to higher volume
{devel,users} lists.
Thank you very much.
--
All the best,
Dmitry Smirnov
GPG key : 4096R/53968D1
Agree with Iban,, adding one more OSD may help.
I met the same problem when I created cluster following quick start guide. 2
OSD nodes gave me health_warn and then adding one more OSD made everything
okay.
From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of Iban
Cabrill
Hi Pratik,
I am not an expert, but I think you need one more OSD server, the default
pools (rbd, metadata, data) have 3 replicas by default.
Regards, I
El 18/07/2014 14:19, "Pratik Rupala"
escribió:
> Hi,
>
> I am deploying firefly version on CentOs 6.4. I am following quick
> installation ins
On Fri, 18 Jul 2014, Anand Bhat wrote:
> I have question on intention of Ceph setmaxosd command. From source code, it
> appears as if this is present as a way to limit the number of OSDs in the
> Ceph cluster.
Yeah. It's basically sizing the array of OSDs in the OSDMap. It's a bit
obsolete sin
Hi,
I am deploying firefly version on CentOs 6.4. I am following quick
installation instructions available at ceph.com.
Kernel version in CentOs 6.4 is 2.6.32-358.
I am using virtual machines for all the nodes. As per the setup, there
are one admin-node, one monitor node and two OSD nodes.
I
Is there any known workarounds to schedule deep scrubs to run nightly?
Latency does go up a little bit when it runs so I’d rather that it didn’t
affect our daily activities.
Kind Regards,
David
___
ceph-users mailing list
ceph-users@lists.ceph.com
http
21 matches
Mail list logo
