I do not have cache pool in it
___
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-le...@ceph.io
Hi Community,
currently i’m installing an nvme only storage cluster with cephadm from scratch
(v.17.2.5). Everything works fine. Each of my nodes (6) has 3 enterprise nvme’s
with 7TB capacity.
At the beginning I only installed one OSD per nvme, now I want to use four
instead of one but I’m str
thank you Robert!
this sounds like split brain again... but we have a quorum system by using 3
monitor nodes. so only the room with the majority of the ceph-mons is available
for I/O.
if the room with the majority of the ceph-mons is the one that is cut-off, I
suppose we'd need to do this:
go
Hello Joffrey,
You could be hitting the slow backfill/recovery issue with mclock_scheduler.
To confirm the above could you please provide the output of the following
commands?
1. ceph versions
2. ceph config show osd. | grep osd_max_backfills
3. ceph config show osd. | grep osd_recovery_max_activ
great, thank you Anthony! :)
so size 4 / min_size 2 would be a lot better (of course)
we have to stay at 3/2 for now though, because our OSDs are filled 60% in sum
maybe someone can answer additional questions:
- what is the best practice to avoid a full OSD scenario, where ceph tries to
recre
ceph 16.2.11,
is safe to enable scrub and deep scrub during backfilling ?
I have log recovery-backfilling due to a new crushmap , backfilling is going
slow and deep scrub interval as expired so I have many pgs not deep-scrubbed
in time.
Best regards
Alessandro
__
I initially ran the upgrade fine but it failed @ around 40/100 on an osd, so
after waiting for along time i thought I'd try restarting it and then
restarting the upgrade.
I am stuck with the below debug error, I have tested docker pull from other
servers and they dont fail for the ceph images b
The conditional policy for the List operations does not work as expected
for the bucket with tenant. With buckets and users without a tenant,
everything is fine.
The owner of t1\bucket1 is t1$user1.
Bucket Policy:
{
"Version":"2012-10-17",
"Statement":[
{
"Sid":"PolicyForUser2Pref
Update:
I tested same hard drives on different servers, and the sata hard drives do
show up with command:
ceph orch device ls
The working systems are Intel pci v3 using chipset SATA controller.
The systems not working are AMD pci v5 using chipset SATA controller.
Both systems are running
Hello,
I’m planning a Ceph multisite deployment with two clusters. One being a primary
one the clients will be intereacting with and a second one with zone tier type
set as an archive one. There is also a one way sync from the primary zone set -
I would like to have this cluster as a "backup zo
On Wed, Mar 01, 2023 at 08:39:56AM -0500, Daniel Gryniewicz wrote:
We're actually writing this for RGW right now. It'll be a bit before
it's productized, but it's in the works.
Just curious, what is the use cases for this feature?
S3 against CephFS?
--
Kai Stian Olstad
___
Thanks.
The question is indeed mainly for rbd. In particolar the old clients I was
referring to are a proxmox cluster and a Openstack installation
Regards, Massimo
Il ven 3 mar 2023, 15:53 Daniel Gryniewicz ha scritto:
> I can't speak for RBD, but for RGW, as long as you upgrade all the RGWs
>
Hello Joffrey,
I am not sure why my previous reply did not go through. I am replying on
this
thread again.
The slow backfills could also be due to the mclock_scheduler throttling the
recovery operations aggressively and this issue is currently being fixed. To
confirm if mclock_scheduler is causing
Ok, Thanks. You mean that the autoscale feature is... stupid ?
I'm going to change pgp_num and use the legacy formula OSDs * 100 / pool
size.
Le jeu. 2 mars 2023 à 17:04, Curt a écrit :
> I see autoscale_mode on all pools and I'm guessing this is your largest
> pool bkp365-ncy.rgw.buckets.data,
This is not speculation: I have personally experienced this with an inherited
2R cluster.
> On Mar 3, 2023, at 04:07, Janne Johansson wrote:
>
>
> Do not assume the last PG needs to die in a horrible fire, killing
> several DC operators with it, it only takes a REALLY small outage, a
> fluke
I can't speak for RBD, but for RGW, as long as you upgrade all the RGWs
themselves, clients will be fine, since they speak S3 to the RGWs, not
RADOS.
Daniel
On 3/3/23 04:29, Massimo Sgaravatto wrote:
Dear all
I am going to update a ceph cluster (where I am using only rbd and rgw,
i.e. I didn'
Hello,
Before we start I'm fully aware that this kind of setup is not
recommended by any means and I'm familiar with it's implications. I'm
just trying to practice extreme situations, just in case...
I have a test cluster with:
3 nodes with Proxmox 7.3 + Ceph Quincy 17.2.5
3 monitors + 3 man
On 03.03.23 11:16, Jeremy Hansen wrote:
3/3/23 2:13:53 AM[WRN]unable to calc client keyring client.admin
placement PlacementSpec(label='_admin'): Cannot place : No matching
hosts for label _admin
I keep seeing this warning in the logs. I’m not really sure what action
to take to resolve this
Hi,
what's your ceph version? You probably don't have host labels set
(check 'ceph orch host ls') and cephadm tries to copy a keyring to
admin hosts with the _admin label [1]. Also check out how to put
keyrings under management [2], I haven't dealt with that too intense
yet. But basically
3/3/23 2:13:53 AM[WRN]unable to calc client keyring client.admin placement
PlacementSpec(label='_admin'): Cannot place : No matching hosts for label _admin
I keep seeing this warning in the logs. I’m not really sure what action to take
to resolve this issue.
Thanks
-jeremy
signature.asc
Desc
Dear all
I am going to update a ceph cluster (where I am using only rbd and rgw,
i.e. I didn't deploy cephfs) from Octtopus to Quincy
Before doing that I would like to understand if some old nautilus clients
(that I can't update for several reasons) will still be able to connect
In general: I am
Den fre 3 mars 2023 kl 01:07 skrev :
> it is unclear for us what min_size means besides what it does. i hope someone
> can clear this up :)
> someone pointed out "split brain" but I am unsure about this.
>
> i think what happens in the worst case is this:
> only 1 PG is available, client writes ch
(for archival purposes)
On Thu, Mar 2, 2023 at 6:04 PM Milind Changire wrote:
> The docs for the ceph kernel module will be updated appropriately in the
> kernel documentation.
> Thanks for pointing out your pain point.
>
> --
> Milind
>
>
> On Thu, Mar 2, 2023 at 1:41 PM Shawn Weeks
> wrote:
>
On 02.03.23 09:16, stefan.pin...@bearingpoint.com wrote:
so if one room goes down/offline, around 50% of the PGs would be left with only
1 replica making them read-only.
Most people forget the other half of the cluster in such a scenario.
For us humans it is obvious that one room is down, be
24 matches
Mail list logo
