Saturday, 2 March 2019, 04.20 +0100 from satha...@gmail.com
:
>56 OSD, 6-node 12.2.5 cluster on Proxmox
>
>We had multiple drives fail(about 30%) within a few days of each other, likely
>faster than the cluster could recover.
Hov did so many drives break?
Jesper
56 OSD, 6-node 12.2.5 cluster on Proxmox
We had multiple drives fail(about 30%) within a few days of each other,
likely faster than the cluster could recover.
After the dust settled, we have 2 out of 896 pgs stuck inactive. The failed
drives are completely inaccessible, so I can't mount them and
True, but not before you unmap it from the previous server. It's like
physically connecting a harddrive to two servers at the same time. Neither
knows what the other is doing to it and can corrupt your data. You should
always make sure to unmap an rbd before mapping it to another server.
On Fri, M
It has to be mounted from somewhere, if that server goes offline, you need
to mount it from somewhere else right?
On Thu, Feb 28, 2019 at 11:15 PM David Turner wrote:
> Why are you making the same rbd to multiple servers?
>
> On Wed, Feb 27, 2019, 9:50 AM Ilya Dryomov wrote:
>
>> On Wed, Feb 2
Hi
I already used the balancer in my ceph luminous cluster a while ago when
all the OSDs were using filestore.
Now, after having added some bluestore OSDs, if I try to create a plan:
[root@ceph-mon-01 ~]# ceph balancer status
{
"active": false,
"plans": [],
"mode": "crush-compat"
}
On Fri, Mar 1, 2019 at 1:09 PM Anthony D'Atri wrote:
>
> Thanks! I'll do some experimentation. Our orchestration service might need
> an overhaul to manage parents / children together.
>
> Related question, in this thread:
>
> https://www.spinics.net/lists/ceph-users/msg35257.html
>
> there was
Hello,
My question is how crush distributes chunks throughout the cluster with
erasure coded pools. Currently, we have 4 OSD nodes with 36 drives(OSD
daemons) per node. If we use ceph_failire_domaon=host, then we are
necessarily limited to k=3,m=1, or k=2,m=2. We would like to explore k>3,
m>2 mod
Hi All
Exporting cephfs with the CEPH_FSAL
I set the following on a dir:
setfattr -n ceph.quota.max_bytes -v 1 /dir
setfattr -n ceph.quota.max_files -v 10 /dir
>From an NFSv4 client, the quota.max_bytes appears to be completely ignored,
I can go GBs over the quota in the dir. The *quota
I believe the question was in regards to which formula to use. There are two
different formulas here [1] and here [2].
The difference being the additional steps used to calculate the appropriate PG
counts for a pool. In Nautilus though, this mostly moot as the mgr service now
has a module to au
+1, I also think's it's strange that deleting OSD by "osd out -> osd
purge" causes two rebalances instead of one.
--
With best regards,
Vitaliy Filippov
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-us
More on the subject can be found here:
https://ceph.com/geen-categorie/difference-between-ceph-osd-reweight-and-ceph-osd-crush-reweight/
On Fri, Mar 1, 2019 at 2:22 PM Darius Kasparavičius wrote:
>
> Hi,
>
> Setting crush weight to 0 removes the osds weight from crushmap, by
> modifying hosts tot
Hi,
Setting crush weight to 0 removes the osds weight from crushmap, by
modifying hosts total weight. Which forces rebalancing of data across
all the cluster. Setting and OSD to out only modifies "REWEIGHT"
status, which balances data inside the same host.
On Fri, Mar 1, 2019 at 12:25 PM Paul Emm
Hi - we are using ceph 12.2.4 and bug#24445 hitting, which caused 10
min IO pause on ceph cluster..
Is this bug fixed?
bug: https://tracker.ceph.com/issues/24445/
Thanks
Swami
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/l
Hi!
Hmm.
"ceph osd crush reweight" and "ceph osd reweight" - do the same or not?
I use "ceph osd crush reweight"
- Original Message -
From: "Paul Emmerich"
To: "Fyodor Ustinov"
Cc: "David Turner" , "ceph-users"
Sent: Friday, 1 March, 2019 12:24:37
Subject: Re: [ceph-users] Right way
resending, not sure the prev email reached the mailing list...
Hi Chen,
thanks for the update. Will prepare patch to periodically reset
StupidAllocator today.
And just to let you know below is an e-mail from AdamK from RH which
might explain the issue with the allocator.
Also please note
On Fri, Mar 1, 2019 at 11:17 AM Fyodor Ustinov wrote:
> May be. But "ceph out + ceph osd purge" causes double relocation, and "ceph
> reweight 0 + ceph osd purge" - causes only one.
No, the commands "ceph osd out X" and "ceph osd reweight X 0" do the
exact same thing: both set reweight to 0.
Pa
Hi Chen,
thanks for the update. Will prepare patch to periodically reset
StupidAllocator today.
And just to let you know below is an e-mail from AdamK from RH which
might explain the issue with the allocator.
Also please note that StupidAllocator might not perform full
defragmentation in r
Hi!
May be. But "ceph out + ceph osd purge" causes double relocation, and "ceph
reweight 0 + ceph osd purge" - causes only one.
- Original Message -
From: "Paul Emmerich"
To: "Fyodor Ustinov"
Cc: "David Turner" , "ceph-users"
Sent: Friday, 1 March, 2019 11:54:20
Subject: Re: [ceph-us
"out" is internally implemented as "reweight 0"
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 Fri, Mar 1, 2019 at 10:48 AM Fyodor Ustinov wrote:
>
> Hi!
>
> As fa
Hi!
As far as I understand, reweight also does not lead to the situation "a period
where one copy / shard
is missing".
- Original Message -
From: "Paul Emmerich"
To: "Fyodor Ustinov"
Cc: "David Turner" , "ceph-users"
Sent: Friday, 1 March, 2019 11:32:54
Subject: Re: [ceph-users] Righ
> Backports should be available in v12.2.11.
s/v12.2.11/ v12.2.12/
Sorry for the typo.
原始邮件
发件人：谢型果10072465
收件人：d...@vanderster.com ;
抄送人：ceph-users@lists.ceph.com ;
日 期 ：2019年03月01日 17:09
主 题 ：Re: [ceph-users] ceph osd pg-upmap-items not working
On Fri, Mar 1, 2019 at 8:55 AM Fyodor Ustinov wrote:
>
> Hi!
>
> Yes. But I am a little surprised by what is written in the documentation:
the point of this is that you don't have a period where one copy/shard
is missing if you wait for it to take it out.
Yeah, there'll be an unnecessary small da
See https://github.com/ceph/ceph/pull/26179
Backports should be available in v12.2.11.
Or you can manually do it by simply adopting
https://github.com/ceph/ceph/pull/26127 if you are eager to get out of the
trap right now.
原始邮件
发件人：DanvanderSter
收件人：Kári Bertil
Hi,
some news, it seem that it's finally stable for me since 1week. (around 0,7ms
of commit latency average)
http://odisoweb1.odiso.net/osdstable.png
The biggest change is the 18/02, where I have finished to rebuild all my osd,
with 2 osd of 3TB for 1NVME 6TB.
(previously I only have done it
24 matches
Mail list logo
