A repop is a sub-operation between primaries and replicas mostly.
That op only shows a duration of 1.3 seconds and the delay you
mentioned previously was under a second. Do you see larger delays? Are
they always between "sub_op_committed" and "commit_sent"?
What is your workload and how heavily
clients also need to access the OSDs and MDS servers
Paul
--
Paul Emmerich
Looking for help with your Ceph cluster? Contact us at https://croit.io
croit GmbH
Freseniusstr. 31h
81247 München
www.croit.io
Tel: +49 89 1896585 90
On Thu, Mar 21, 2019 at 1:02 PM Andres Rojas Guerrero wrote:
>
>
On 03/21/2019 11:27 AM, Maged Mokhtar wrote:
>
> Though i do not recommend changing it, if there is a need to lower
> fast_io_fail_tmo, then osd_heartbeat_interval + osd_heartbeat_grace sum
> need to be lowered as well, their default sum is 25 sec, which i would
> assume why fast_io_fail_tmo is
This behaviour is still an issue in mimic 13.2.5 and nautilus 14.2.0.
I've logged https://tracker.ceph.com/issues/38841 for this. Apologies if
this has already been done.
Simon
On 06/03/2019 20:17, Simon Ironside wrote:
Yes, as I said that bug is marked resolved. It's also marked as only
Though i do not recommend changing it, if there is a need to lower
fast_io_fail_tmo, then osd_heartbeat_interval + osd_heartbeat_grace sum
need to be lowered as well, their default sum is 25 sec, which i would
assume why fast_io_fail_tmo is set to this. you would want to have your
higher
On Thu, Mar 21, 2019 at 12:14 PM Eugen Block wrote:
>
> Hi Dan,
>
> I don't know about keeping the osd-id but I just partially recreated
> your scenario. I wiped one OSD and recreated it. You are trying to
> re-use the existing block.db-LV with the device path (--block.db
> /dev/vg-name/lv-name)
On 21/03/19 16:15 +0100, Dan van der Ster wrote:
On Thu, Mar 21, 2019 at 1:50 PM Tom Barron wrote:
On 20/03/19 16:33 +0100, Dan van der Ster wrote:
>Hi all,
>
>We're currently upgrading our cephfs (managed by OpenStack Manila)
>clusters to Mimic, and want to start enabling snapshots of the
On Thu, Mar 21, 2019 at 1:50 PM Tom Barron wrote:
>
> On 20/03/19 16:33 +0100, Dan van der Ster wrote:
> >Hi all,
> >
> >We're currently upgrading our cephfs (managed by OpenStack Manila)
> >clusters to Mimic, and want to start enabling snapshots of the file
> >shares.
> >There are different ways
It's just the design of the iSCSI protocol. Sure, you can lower the
timeouts (see "fast_io_fail_tmo" [1]) but you will just end up w/ more
false-positive failovers.
[1] http://docs.ceph.com/docs/master/rbd/iscsi-initiator-linux/
On Thu, Mar 21, 2019 at 10:46 AM li jerry wrote:
>
> Hi Maged
>
>
Hi Maged
thank you for your reply.
To exclude the osd_heartbeat_interval and osd_heartbeat_grace factors, I
cleared the current lio configuration, redeployed two CENTOS7 (not in any ceph
role), and deployed rbd-target-api, rbd-target-gw, trum-runner on it. ;
And do the following test
1.
They released Luminous 12.2.11 and now my large objects are starting to
count down ( after running the rm command suggested in the release notes ).
Seems dynamic sharding will clean up after itself now to! So case closed!
-Brent
-Original Message-
From: ceph-users On Behalf Of Brent
On 20/03/19 16:33 +0100, Dan van der Ster wrote:
Hi all,
We're currently upgrading our cephfs (managed by OpenStack Manila)
clusters to Mimic, and want to start enabling snapshots of the file
shares.
There are different ways to approach this, and I hope someone can
share their experiences with:
Hi all, we have deployed a Ceph cluster configured with two public networks:
[global]
cluster network = 10.100.188.0/23
fsid = 88f62260-b8de-499f-b6fe-5eb66a967083
mon host = 10.100.190.9,10.100.190.10,10.100.190.11
mon initial members = mon1,mon2,mon3
osd_pool_default_pg_num = 4096
public
Hi Dan,
I don't know about keeping the osd-id but I just partially recreated
your scenario. I wiped one OSD and recreated it. You are trying to
re-use the existing block.db-LV with the device path (--block.db
/dev/vg-name/lv-name) instead the lv notation (--block.db
vg-name/lv-name):
Martin, my thanks to Croit for making this repository available.
I have been building Ceph from source on Ubuntu Cosmic for the last few
days.
It is much more convenient to use a repo.
On Thu, 21 Mar 2019 at 09:32, Martin Verges wrote:
> Hello,
>
> we strongly believe it would be good for Ceph
Hello,
we strongly believe it would be good for Ceph to have the packaged
directly on the official Debian mirrors, but for everyone out there
having trouble with Ceph on Debian we are glad to help.
If Ceph is not available on Debian, it might affect a lot of other
Software, for example Proxmox.
Hello Ceph,
What is the best way to find out how the RocksDB is currently performing? I
need to build a business case for NVME devices for RocksDB.
Kind regards,
Glen Baars
This e-mail is intended solely for the benefit of the addressee(s) and any
other named recipient. It is confidential and
On Thu, Mar 21, 2019 at 8:51 AM Gregory Farnum wrote:
>
> On Wed, Mar 20, 2019 at 6:06 PM Dan van der Ster wrote:
>>
>> On Tue, Mar 19, 2019 at 9:43 AM Erwin Bogaard
>> wrote:
>> >
>> > Hi,
>> >
>> >
>> >
>> > For a number of application we use, there is a lot of file duplication.
>> > This
Beware of start using this influx. I have 80GB db, and I regret using
it. I have to move now to storing data in graphite. Collectd has also a
plugin for that.
Influx cannot downsample properly when having tags I think (still wait
for a response to this [0])
What I have understood is that
On Wed, Mar 20, 2019 at 6:06 PM Dan van der Ster wrote:
> On Tue, Mar 19, 2019 at 9:43 AM Erwin Bogaard
> wrote:
> >
> > Hi,
> >
> >
> >
> > For a number of application we use, there is a lot of file duplication.
> This wastes precious storage space, which I would like to avoid.
> >
> > When
Hello Brad,
It doesn't seem to be a set of OSDs, the cluster has 160ish OSDs over 9 hosts.
I seem to get a lot of these ops also that don't show a client.
"description": "osd_repop(client.14349712.0:4866968 15.36
e30675/22264 15:6dd17247:::rbd_data.2359ef6b8b4567.0042766
21 matches
Mail list logo