Hi all,
I just submitted an OpenStack Forum proposal for a Ceph BoF session at
OpenStack Sydney. If you're interested in seeing this happen then
please hit up http://forumtopics.openstack.org/cfp/details/46 with
your comments / +1's.
--
Cheers,
~Blairo
___
Hello,
We are working on a POC with containers (kubernetes) and cephfs (for permanent
storage).
The main idea is to give to a user access to a subdirectory of the cephfs but
be sure he won't be able to access to the rest of the storage. As
k8s works, the user will have access to the yml file wh
Maybe this will get you started with the permissions for only this fs
path /smb
sudo ceph auth get-or-create client.cephfs.smb mon 'allow r' mds 'allow
r, allow rw path=/smb' osd 'allow rwx pool=fs_meta,allow rwx
pool=fs_data'
-Original Message-
From: Yoann Moulin [mailto:yoann.m
>> We are working on a POC with containers (kubernetes) and cephfs (for
>> permanent storage).
>>
>> The main idea is to give to a user access to a subdirectory of the
>> cephfs but be sure he won't be able to access to the rest of the
>> storage. As k8s works, the user will have access to the
Hi all,
I use puppet to deploy and manage my clusters.
Recently, as I have been doing a removal of old hardware and adding of
new I've noticed that sometimes the "ceph osd create" is returning
repeated IDs. Usually it's on the same server, but yesterday I saw it
in different servers.
I was expec
I think that is because of the older kernel client, like mentioned here?
https://www.mail-archive.com/ceph-users@lists.ceph.com/msg39734.html
-Original Message-
From: Yoann Moulin [mailto:yoann.mou...@epfl.ch]
Sent: vrijdag 29 september 2017 10:00
To: ceph-users
Subject: Re: [ceph-u
Do you mean that after you delete and remove the crush and auth entries for the
OSD, when you go to create another OSD later it will re-use the previous OSD ID
that you have destroyed in the past?
Because I have seen that behaviour as well - but only for previously allocated
OSD IDs that have
We are working on a POC with containers (kubernetes) and cephfs (for
permanent storage).
The main idea is to give to a user access to a subdirectory of the
cephfs but be sure he won't be able to access to the rest of the
storage. As k8s works, the user will have acc
Quoting Yoann Moulin (yoann.mou...@epfl.ch):
>
> Kernels on client is 4.4.0-93 and on ceph node are 4.4.0-96
>
> What is exactly an older kernel client ? 4.4 is old ?
See
http://docs.ceph.com/docs/master/cephfs/best-practices/#which-kernel-version
If you're on Ubuntu Xenial I would advise to us
Hi,
I'm trying to use CEPH-12.2.0 as storage for with Bareos-16.2.4 backup with
libradosstriper1 support.
Libradosstriber was suggested on this list to solve the problem, that
current CEPH-12 discourages users from using object with very big size
(>128MB). Bareos treat Rados Object as Volume and
Quoting Christian Balzer (ch...@gol.com):
>
> On Thu, 28 Sep 2017 22:36:22 + Gregory Farnum wrote:
>
> > Also, realize the deep scrub interval is a per-PG thing and (unfortunately)
> > the OSD doesn't use a global view of its PG deep scrub ages to try and
> > schedule them intelligently acros
Hi,
I discovered that my cluster starts to make slow requests and all disk
activity get blocked.
This happens once a day. And the ceph OSD get 100% CPU. In the ceph
health I get something like:
2017-09-29 10:49:01.227257 [INF] pgmap v67494428: 764 pgs: 1
active+recovery_wait+degraded+inc
On Fri, Sep 29, 2017 at 9:44 AM, Adrian Saul
wrote:
>
> Do you mean that after you delete and remove the crush and auth entries for
> the OSD, when you go to create another OSD later it will re-use the previous
> OSD ID that you have destroyed in the past?
>
The issue is that has been giving th
On 2017-09-29 10:44, Adrian Saul wrote:
> Do you mean that after you delete and remove the crush and auth entries for
> the OSD, when you go to create another OSD later it will re-use the previous
> OSD ID that you have destroyed in the past?
>
> Because I have seen that behaviour as well - bu
>> Kernels on client is 4.4.0-93 and on ceph node are 4.4.0-96
>>
>> What is exactly an older kernel client ? 4.4 is old ?
>
> See
> http://docs.ceph.com/docs/master/cephfs/best-practices/#which-kernel-version
>
> If you're on Ubuntu Xenial I would advise to use
> "linux-generic-hwe-16.04". Curr
Quoting Yoann Moulin (yoann.mou...@epfl.ch):
>
> >> Kernels on client is 4.4.0-93 and on ceph node are 4.4.0-96
> >>
> >> What is exactly an older kernel client ? 4.4 is old ?
> >
> > See
> > http://docs.ceph.com/docs/master/cephfs/best-practices/#which-kernel-version
> >
> > If you're on Ubuntu
Hello,
We see very poor performance when reading/writing rados objects. The speed
is only 3-4MB/sec, compared to 95MB rados benchmarking.
When you look on underline code it uses librados and linradosstripper
libraries (both have poor performance) and the code uses rados_read and
rados_write func
Ok, thanks! So I'll wait a few days for the command to complete and see
what kind to of output it produces then.
Regards,
Andreas
On 29 Sep 2017 12:32 a.m., "Christian Wuerdig"
wrote:
> I'm pretty sure the orphan find command does exactly just that -
> finding orphans. I remember some emails on
Hi,
On 29/09/17 01:00, Brad Hubbard wrote:
> This looks similar to
> https://bugzilla.redhat.com/show_bug.cgi?id=1458007 or one of the
> bugs/trackers attached to that.
Yes, although increasing the timeout still leaves the issue that if the
timeout fires you don't get anything resembling a useful
On 2017-09-29 11:31, Maged Mokhtar wrote:
> On 2017-09-29 10:44, Adrian Saul wrote:
>
> Do you mean that after you delete and remove the crush and auth entries for
> the OSD, when you go to create another OSD later it will re-use the previous
> OSD ID that you have destroyed in the past?
>
>
Kernels on client is 4.4.0-93 and on ceph node are 4.4.0-96
What is exactly an older kernel client ? 4.4 is old ?
>>>
>>> See
>>> http://docs.ceph.com/docs/master/cephfs/best-practices/#which-kernel-version
>>>
>>> If you're on Ubuntu Xenial I would advise to use
>>> "linux-generic-
On Fri, Sep 29, 2017 at 8:58 PM, Matthew Vernon wrote:
> Hi,
>
> On 29/09/17 01:00, Brad Hubbard wrote:
>> This looks similar to
>> https://bugzilla.redhat.com/show_bug.cgi?id=1458007 or one of the
>> bugs/trackers attached to that.
>
> Yes, although increasing the timeout still leaves the issue t
Dear Stefan,
Thanks for your help. You are right. I was missing apt update" after adding
repo.
After doing apt update I am able to install luminous
cadmin@admin:~/my-cluster$ ceph --versionceph version 12.2.1
(3e7492b9ada8bdc9a5cd0feafd42fbca27f9c38e) luminous (stable)
I am not much in practice
"apt-cache policy" shows you the different versions that are possible to
install, and the prioritized order they have.
the highest version will normally be installed unless priorities are
changed.
example:
apt-cache policy ceph
ceph:
Installed: 12.2.1-1~bpo90+1
Candidate: 12.2.1-1~bpo90+1
It sounds like you are doing synchronous reads of small objects here. In
that case you are dominated by the per-op already rather than the
throughout of your cluster. Using aio or multiple threads will let you
parallelism requests.
-Greg
On Fri, Sep 29, 2017 at 3:33 AM Alexander Kushnirenko
wrote:
In cases like this you also want to set RADOS namespaces for each tenant’s
directory in the CephFS layout and give them OSD access to only that
namespace. That will prevent malicious users from tampering with the raw
RADOS objects of other users.
-Greg
On Fri, Sep 29, 2017 at 4:33 AM Yoann Moulin
Hi - I'm trying to get logs of Objecter and librados while doing
operations(read/write) on an rbd image. Here is my ceph.conf:
[global]
debug_objecter = 20
debug_rados = 20
[client]
rbd_cache = false
log file = /self/ceph/ceph-rbd.log
debug rbd = 20
debug objecter
Hi,
> Kernels on client is 4.4.0-93 and on ceph node are 4.4.0-96
>
> What is exactly an older kernel client ? 4.4 is old ?
>
> See
> http://docs.ceph.com/docs/master/cephfs/best-practices/#which-kernel-version
>
> If you're on Ubuntu Xenial I would advise to use
>>
On Fri, Sep 29, 2017 at 7:34 AM Yoann Moulin wrote:
> Hi,
>
> > Kernels on client is 4.4.0-93 and on ceph node are 4.4.0-96
> >
> > What is exactly an older kernel client ? 4.4 is old ?
> >
> > See
> >
> http://docs.ceph.com/docs/master/cephfs/best-practices/#which-kernel-
Hello,
I'm doing some tests on the radosgw on luminous (12.2.1), I have a few
questions.
In the documentation[1], there is a reference to "radosgw-admin region get" but
it seems not to be available anymore.
It should be "radosgw-admin zonegroup get" I guess.
1. http://docs.ceph.com/docs/lumino
>> In cases like this you also want to set RADOS namespaces for each tenant’s
>> directory in the CephFS layout and give them OSD access to only that
>> namespace. That will prevent malicious users from tampering with the raw
>> RADOS objects of other users.
>
> You mean by doing something
If you're scheduling them appropriately so that no deep scrubs will happen
on their own, then you can just check the cluster status if any PGs are
deep scrubbing at all. If you're only scheduling them for specific pools,
then you can confirm which PGs are being deep scrubbed in a specific pool
wit
Is there a way I could get a performance stats for rbd images? I’m looking for
iops and throughput.
This issue we are dealing with is that there was a sudden jump in throughput
and I want to be able to find out with rbd volume might be causing it. I just
manage the ceph cluster, not the opensta
Hello,
Ich think that the Ceph Users don't recommend on ceph osd on Hardware
RAID. But i haven't found a technical Solution for this.
Can anybody give me so a Solution?
Thanks for your help
Regards
Hauke
--
www.w3-creative.de
www.westchat.de
___
I'm going to assume you're dealing with your scrub errors and have a game
plan for those as you didn't mention them in your question at all.
One thing I'm always leary of when I see blocked requests happening is that
the PGs might be splitting subfolders. It is pretty much a guarantee if
you're a
The reason it is recommended not to raid your disks is to give them all to
Ceph. When a disk fails, Ceph can generally recover faster than the raid
can. The biggest problem with raid is that you need to replace the disk
and rebuild the raid asap. When a disk fails in Ceph, the cluster just
moves
On Thu, 10 Aug 2017, John Spray said:
> On Thu, Aug 10, 2017 at 4:31 PM, Sean Purdy wrote:
> > Luminous 12.1.1 rc
And 12.2.1 stable
> > We added a new disk and did:
> > That worked, created osd.18, OSD has data.
> >
> > However, mgr output at http://localhost:7000/servers showed
> > osd.18 unde
There is no tool on the Ceph side to see which RBDs are doing what.
Generally you need to monitor the mount points for the RBDs to track that
down with iostat or something.
That said, there are some tricky things you could probably do to track down
the RBD that is doing a bunch of stuff (as long a
On 2017-09-29 17:14, Hauke Homburg wrote:
> Hello,
>
> Ich think that the Ceph Users don't recommend on ceph osd on Hardware
> RAID. But i haven't found a technical Solution for this.
>
> Can anybody give me so a Solution?
>
> Thanks for your help
>
> Regards
>
> Hauke
You get better perform
On 2017-09-29 17:13, Matthew Stroud wrote:
> Is there a way I could get a performance stats for rbd images? I'm looking
> for iops and throughput.
>
> This issue we are dealing with is that there was a sudden jump in throughput
> and I want to be able to find out with rbd volume might be causi
His dilemma sounded like he has access to the cluster, but not any of the
clients where the RBDs are used or even the hypervisors in charge of those.
On Fri, Sep 29, 2017 at 12:03 PM Maged Mokhtar wrote:
> On 2017-09-29 17:13, Matthew Stroud wrote:
>
> Is there a way I could get a performance st
Yeah, that is the core problem. I have been working with those teams that
manage those. However, there isn’t a way I can check on my side as it appears.
From: David Turner
Date: Friday, September 29, 2017 at 11:08 AM
To: Maged Mokhtar , Matthew Stroud
Cc: "ceph-users@lists.ceph.com"
Subject:
There is a feature in the backlog for a "rbd top"-like utility which
could provide a probabilistic view of the top X% of RBD image stats
against the cluster. The data collection would be by each OSD
individually which it why it would be probabilistic stats instead of
an absolute. It also would onl
Hi everyone,
thanks for the advice, we consulted it and we're gonna test it out with
cephfs first. Object storage is a possibility if it misbehaves.
Hopefully it will go well :)
On 28/09/17 08:20, Henrik Korkuc wrote:
On 17-09-27 14:57, Josef Zelenka wrote:
Hi,
we are currently working on
Yeah, I don’t have access to the hypervisors, nor the vms on said hypervisors.
Having some sort of ceph-top would be awesome, I wish they would implement that.
Thanks,
Matthew Stroud
On 9/29/17, 11:49 AM, "Jason Dillaman" wrote:
There is a feature in the backlog for a "rbd top"-like utilit
I haven't used the striper, but it appears to make you specify sizes,
stripe units, and stripe counts. I would expect you need to make sure that
the size is an integer multiple of the stripe unit. And it probably
defaults to a 4MB object if you don't specify one?
On Fri, Sep 29, 2017 at 2:09 AM Al
In addition to the points that others made so well:
- When using parity RAID, eg. RAID5 to create OSD devices, one reduces
aggregate write speed — specially if using HDD’s — due to write amplification.
- If using parity or replicated RAID, one might semi-reasonably get away with
reducing Ceph’s
Luis:
As others have mentioned, be sure that when you delete an OSD each step is
completed successfully:
- OSD process is killed
- OSD is marked out/down in the CRUSH map
- ceph osd crush delete osd.xxx
- ceph osd rm osd.xxx
- ceph auth del osd.xxx
- Also be sure to unmount the /var/lib/ceph mou
48 matches
Mail list logo