Hello,
What's the status with the *-stable-* tags?
https://quay.io/repository/ceph/daemon?tab=tags
No longer build/support?
What should we use until we'll migrate from ceph-ansible to cephadm?
Thanks.
--
Jonas
___
ceph-users mailing list --
You should be able to do `ceph health mute OSD_SLOW_PING_TIME_BACK --sticky` to
mute the health warn/error state as it flaps.
You can also set a TTL for the mute (1d, 1w, 1m) to have it roll off after a
specific time.
Code here is the warning error such as OSD_SLOW_PING_TIME_BACK or
On 1/30/23 15:15, Ana Aviles wrote:
Hi,
Josh already suggested, but I will one more time. We had similar
behaviour upgrading from Nautilus to Pacific. In our case compacting
the OSDs did the trick.
Thanks for chimming in! Unfortunately, in my case neither an online
compaction (ceph tell
oph wait,
i might have been too impatient:
1/30/23 4:43:07 PM[INF]Deploying daemon osd.232 on ceph-a1-06
1/30/23 4:42:26 PM[INF]Found osd claims for drivegroup
dashboard-admin-1661788934732 -> {'ceph-a1-06': ['232']}
1/30/23 4:42:26 PM[INF]Found osd claims -> {'ceph-a1-06': ['232']}
root@ceph-a2-01:/# ceph osd destroy 232 --yes-i-really-mean-it
destroyed osd.232
OSD 232 shows now as destroyed and out in the dashboard.
root@ceph-a1-06:/# ceph-volume lvm zap /dev/sdm
--> Zapping: /dev/sdm
--> --destroy was not specified, but zapping a whole device will remove
the
I think you will have to do the science there, it will be dependent on
thousands of factors like size of OSD, WAL, DB, placement of them,
size of incoming data if it is many small ops or few large ones and so
on. I don't think any one single answer there would work out for many
cases.
Most of us
The 'down' status is why it's not being replaced, vs. destroyed, which would
allow the replacement. I'm not sure why --replace lead to that scenario, but
you will probably need to mark it destroyed for it to be replaced.
OK. How much data will be written to the WAL and elsewhere?
On Mon, Jan 30, 2023 at 3:17 PM Janne Johansson wrote:
> > I'm concerned with the potential increased NVME wear. Assuming writes of
> multiples of the block size, when I write 1GB of data to the CephFS, how
> much data is written to
> I'm concerned with the potential increased NVME wear. Assuming writes of
> multiples of the block size, when I write 1GB of data to the CephFS, how much
> data is written to the disks?
In that case, repl=3 will write 1GB to three PGs. EC 8+3 would write
125M (ie, 1/8 of a GB) to 11 (8+3)
Hi,
Josh already suggested, but I will one more time. We had similar
behaviour upgrading from Nautilus to Pacific. In our case compacting the
OSDs did the trick.
For us there was no performance impact running the compaction (ceph osd
daemon osd.0 compact) although we run them in batches and
Dear Janne,
I'm concerned with the potential increased NVME wear. Assuming writes of
multiples of the block size, when I write 1GB of data to the CephFS, how
much data is written to the disks?
Best wishes,
Manuel
On Mon, Jan 30, 2023 at 2:25 PM Janne Johansson wrote:
> > is there information
> is there information available anywhere about the write amplification for
> CephFS? I found quite some material on write amplification of VMs using
> journaled file system on top of RBD but nothing as it relates to CephFS?
>
> From my understanding I would expect the following:
>
> - for X-rep,
On Sun, Jan 29, 2023 at 1:12 PM Jiatong Shen wrote:
>
> Hello Ilya,
>
>Thank you very much for the clarification! Another question, for some
> historical reasons, there are still some
> Luminous clients existing. Is it dangerous to sparsify an image which is
> still being used by a
Hey all,
I haven't managed to solve this issue yet.
To simplify things, I'm looking to restart one OSD which crashes shortly
after starting.
As mentioned before I've ruled out this being related to hardware.
Not a dev but looking at the log my error occurs at this
Dear all,
is there information available anywhere about the write amplification for
CephFS? I found quite some material on write amplification of VMs using
journaled file system on top of RBD but nothing as it relates to CephFS?
>From my understanding I would expect the following:
- for X-rep,
# ceph orch osd rm status
No OSD remove/replace operations reported
# ceph orch osd rm 232 --replace
Unable to find OSDs: ['232']
It is not finding 232 anymore. It is still shown as down and out in the
Ceph-Dashboard.
pgs: 3236 active+clean
This is the new disk shown as locked
Glad you found it, its quite a long one. We could never confirm that a manual
compaction after an upgrade with quick fix on mount = true would solve the
problem with snap trim. You might want to try off-line fixing using
ceph-bluestore-tool repair --path /var/lib/ceph/osd/OSD_ID
maybe:
I'm reading that thread on spinics. Thanks for pointing that out.
The cluster was upgraded from Nautilus to Octopus first week of
December. No previous version has been used in this cluster. I just
don't remember if we used snapshots after the upgrade.
bluestore_fsck_quick_fix_on_mount is at
Hi Victor,
out of curiosity, did you upgrade the cluster recently to octopus? We and
others observed this behaviour when following one out of two routes to upgrade
OSDs. There was a thread "Octopus OSDs extremely slow during upgrade from
mimic", which seems to have been lost with the recent
Hi,
it's not the best idea to create multiple threads for the same topic.
I don't really have an answer for you, but maybe you could try to add
more rbd-mirror daemons and see if the replication speed increases. I
haven't tried that myself yet, though.
Regards,
Eugen
Zitat von ankit
20 matches
Mail list logo