On Wed, Nov 15, 2017 at 6:50 AM David Turner wrote:
> 2 weeks later and things are still deleting, but getting really close to
> being done. I tried to use ceph-objectstore-tool to remove one of the
> PGs. I only tested on 1 PG on 1 OSD, but it's doing something really
>
One thing I have just noticed.
The Abort is always on the next thread along for example the last PG 6.84s12
was thread 7f78721ad700, however the assert was listed for 7f78721ae700
Does this mean the PG 6.84s12 crashed causing the next thread to exit, or was
the thread 7f78721ae700 what caused
Will add to ticket.
But no the cluster does not see the OSD go up, just the OSD fails on the same
assert.
,Ashley
From: David Turner [mailto:drakonst...@gmail.com]
Sent: 18 November 2017 23:19
To: Ashley Merrick
Cc: Eric Nelson ;
The osd shouldn't be about to peer while it's down. I think this is good
information to update your ticket with as it is possible a different code
path than anticipated.
Did your cluster see the osd as up?
On Sat, Nov 18, 2017, 9:32 AM Ashley Merrick wrote:
> Hello,
>
>
Hello,
Added an empty one, and was fine as I guess had no peering to do as had no PG’s.
I did also disable no backfill for a while and some PG’s have moved across
fine, I did try and export a PG on a OSD that fails to boot and import a PG
onto the OSD but then caused the OSD to do the same
Hi,
Is it possible to add new empty osds to your cluster? Or do these also
crash out?
Thanks
On 18 Nov 2017 14:32, "Ashley Merrick" wrote:
> Hello,
>
>
>
> So seems noup does not help.
>
>
>
> Still have the same error :
>
>
>
> 2017-11-18 14:26:40.982827 7fb4446cd700
Hello,
So seems noup does not help.
Still have the same error :
2017-11-18 14:26:40.982827 7fb4446cd700 -1 *** Caught signal (Aborted) **in
thread 7fb4446cd700 thread_name:tp_peering
ceph version 12.2.1 (3e7492b9ada8bdc9a5cd0feafd42fbca27f9c38e) luminous (stable)
1: (()+0xa0c554)
Hello,
Will try with the noup now and see if makes any difference.
Is effecting both BS & FS OSD’s and effecting different host’s and different
PG’s seems to be no form of pattern.
,Ashley
From: David Turner [mailto:drakonst...@gmail.com]
Sent: 18 November 2017 22:19
To: Ashley Merrick
Does letting the cluster run with noup for a while until all down disks are
idle, and then letting them come in help at all? I don't know your
specific issue and haven't touched bluestore yet, but that is generally
sound advice when is won't start.
Also is there any pattern to the osds that are
Can you capture a blktrace while perform fstrim to record the discard
operations? A 1TB trim extent would cause a huge impact since it would
translate to approximately 262K IO requests to the OSDs (assuming 4MB
backing files).
On Fri, Nov 17, 2017 at 6:19 PM, Brendan Moloney
Is there a way to rebuild the contents of .rgw.buckets.index pool removed by
accident ?
Thanks in advance.
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Hello,
Any further suggestions or work around's from anyone?
Cluster is hard down now with around 2% PG's offline, on the occasion able to
get an OSD to start for a bit but then will seem to do some peering and again
crash with "*** Caught signal (Aborted) **in thread 7f3471c55700
Just a couple of points.
There is no way you can be writing over 7000 iops to 27x7200rpm disks at a
replica level of 3. As Mark has suggested, with a 1GB test file, you are only
touching a tiny area on each physical disk and so you are probably getting a
combination of short stroking from the
On Sat, Nov 18, 2017 at 4:49 PM, Fred Gansevles wrote:
> Hi,
>
> Currently our company has +/- 50 apps where every app has its own
> data-area on NFS.
> We need to switch S3, using Ceph, as our new data layer with
> every app using its own s3-bucket, equivalent to the NFS
Hi,
Currently our company has +/- 50 apps where every app has its own
data-area on NFS.
We need to switch S3, using Ceph, as our new data layer with
every app using its own s3-bucket, equivalent to the NFS data-area.
The sizes of the data-areas, depending on the app, varies from 1.3 GB to
358
15 matches
Mail list logo