Hi Wildo
If I disable the epel repo then the error changes:
[root@ninja ~]# yum install --disablerepo=epel ceph
Loaded plugins: langpacks, priorities, product-id, subscription-manager
10 packages excluded due to repository priority protections
Resolving Dependencies
.
-- Finished Dependency
Hi Wildo
Has something broken with this move? The following has worked for me repeatedly
over the last 2 months:
This a.m. I tried to install ceph using the following repo file:
[root@citrus ~]# cat /etc/yum.repos.d/ceph.repo
[ceph]
name=Ceph packages for $basearch
On 03/09/2015 12:54 PM, HEWLETT, Paul (Paul)** CTR ** wrote:
Hi Wildo
Has something broken with this move? The following has worked for me
repeatedly over the last 2 months:
It shouldn't have broken anything, but you never know.
The machine rsyncs the data from ceph.com directly. The
Hi,
I'm trying to fix an issue within 0.93 on our internal cloud related
to incomplete pg's (yes, I realise the folly of having the dev release
- it's a not-so-test env now, so I need to recover this really). I'll
detail the current outage info;
72 initial (now 65) OSDs
6 nodes
* Update to 0.92
Hi Karan,
We faced same issue and resolved after increasing the open file limit and
maximum no of threads
Config reference
/etc/security/limit.conf
root hard nofile 65535
sysctl -w kernel.pid_max=4194303
http://tracker.ceph.com/issues/10554#change-47024
Cheers
Mohamed Pakkeer
On Mon, Mar
On 03/09/2015 02:47 PM, HEWLETT, Paul (Paul)** CTR ** wrote:
Hi Wildo
If I disable the epel repo then the error changes:
[root@ninja ~]# yum install --disablerepo=epel ceph
Loaded plugins: langpacks, priorities, product-id, subscription-manager
10 packages excluded due to repository
Hi,
I'm trying to fix an issue within 0.93 on our internal cloud related
to incomplete pg's (yes, I realise the folly of having the dev release
- it's a not-so-test env now, so I need to recover this really). I'll
detail the current outage info;
72 initial (now 65) OSDs
6 nodes
* Update to 0.92
When did you make the change?
It worked on Friday albeit with these extra lines in ceph.repo:
[Ceph-el7]
name=Ceph-el7
baseurl=http://eu.ceph.com/rpms/rhel7/noarch/
enabled=1
gpgcheck=0
which I removed when I discovered this no longer existed.
Regards
Paul Hewlett
Senior Systems Engineer
On 03/09/2015 02:27 PM, HEWLETT, Paul (Paul)** CTR ** wrote:
When did you make the change?
Yesterday
It worked on Friday albeit with these extra lines in ceph.repo:
[Ceph-el7]
name=Ceph-el7
baseurl=http://eu.ceph.com/rpms/rhel7/noarch/
enabled=1
gpgcheck=0
which I removed when I
Hi,
Since the recent reports of rsync failing on eu.ceph.com I moved
eu.ceph.com to a new machine.
It went from physical to a KVM VM backed by RBD, so it's now running on
Ceph.
URLs or rsync paths haven't changed, it's still eu.ceph.com and
available over IPv4 and IPv6.
This Virtual Machine is
Hi Karan,
as you are actually writing in your own book, the problem is the sysctl
setting kernel.pid_max. I've seen in your bug report that you were
setting it to 65536, which is still to low for high density hardware.
In our cluster, one OSD server has in an idle situation about 66.000
Threads
Hi All,
I just created this little bash script to retrieve the /dev/disk/by-id
string for each OSD on a host. Our disks are internally mounted so have no
concept of drive bays, this should make it easier to work out what disk has
failed.
#!/bin/bash
DISKS=`ceph-disk list | grep ceph data`
Hi Wildo
It seems that your move coincided with yet another change in the EPEL repo.
For anyone who is interested, I fixed this by:
1. ensuring that check_obsoletes=1 is in
/etc/yum/pluginconf.d/priorities.conf
2. Install libunwind explicitly:
yum install libunwind
Thanks Guys kernel.pid_max=4194303 did the trick.
- Karan -
On 09 Mar 2015, at 14:48, Christian Eichelmann
christian.eichelm...@1und1.de wrote:
Hi Karan,
as you are actually writing in your own book, the problem is the sysctl
setting kernel.pid_max. I've seen in your bug report that
You'll probably have to recreate osds with the same ids (empty ones),
let them boot, stop them, and mark them lost. There is a feature in the
tracker to improve this behavior: http://tracker.ceph.com/issues/10976
-Sam
On Mon, 2015-03-09 at 12:24 +, joel.merr...@gmail.com wrote:
Hi,
I'm
- Original Message -
From: Steffen Winther ceph.u...@siimnet.dk
To: ceph-users@lists.ceph.com
Sent: Monday, March 9, 2015 12:43:58 AM
Subject: Re: [ceph-users] S3 RadosGW - Create bucket OP
Steffen W Sørensen stefws@... writes:
Response:
HTTP/1.1 200 OK
Date: Fri, 06 Mar
Umm.. Too many Threads are created in SimpleMessenger, every pipe
should create two working threads for sending and receiving messages.
Thus, AsyncMessenger would be promissing but still in development.
Regards
Ning Yao
2015-03-09 20:48 GMT+08:00 Christian Eichelmann
Hi Mike,
I was using bs_aio with the krbd and still saw a small caching effect. I'm
not sure if it was on the ESXi or tgt/krbd page cache side, but I was
definitely seeing the IO's being coalesced into larger ones on the krbd
device in iostat. Either way, it would make me potentially nervous to
Thank you Nick for explaining the problem with 4k writes.Queue depth used
in this setup is 256 the maximum supported.
Can you clarify that adding more nodes will not increase iops.In general
how will we increase iops of a ceph cluster.
Thanks for your help
On Sat, Mar 7, 2015 at 5:57 PM, Nick
Hi All,
This is a new release of ceph-deploy that changes a couple of behaviors.
On RPM-based distros, ceph-deploy will now automatically enable
check_obsoletes in the Yum priorities plugin. This resolves an issue
many community members hit where package dependency resolution was
breaking due to
- Original Message -
From: Steffen Winther ceph.u...@siimnet.dk
To: ceph-users@lists.ceph.com
Sent: Monday, March 9, 2015 1:25:43 PM
Subject: Re: [ceph-users] S3 RadosGW - Create bucket OP
Yehuda Sadeh-Weinraub yehuda@... writes:
If you're using apache, then it filters out
I know I'm not even close to this type of a problem yet with my small
cluster (both test and production clusters) - but it would be great if
something like that could appear in the cluster HEALTHWARN, if Ceph could
determine the amount of used processes and compare them against the current
limit
Either option #1 or #2 depending on if your data has hot spots or you need
to use EC pools. I'm finding that the cache tier can actually slow stuff
down depending on how much data is in the cache tier vs on the slower tier.
Writes will be about the same speed for both solutions, reads will be a
Hi David also for the Calamari or gui monitoring interface is there any way to
get user account and passwd of inktank since the repo to install Calamari seems
to be only for people inside of inktank
Jesus Chavez
SYSTEMS ENGINEER-C.SALES
jesch...@cisco.commailto:jesch...@cisco.com
Phone: +52
Nick Fisk nick@... writes:
My Ceph cluster comprises of 4 Nodes each with the following:-
10x 3TB WD Red Pro disks - EC pool k=3 m=3 (7200rpm)
2x S3700 100GB SSD's (20k Write IOPs) for HDD Journals
1x S3700 400GB SSD (35k Write IOPs) for cache tier - 3x replica
If I have following 4x node
Hi Tony,
sounds like an good idea!
Udo
On 09.03.2015 21:55, Tony Harris wrote:
I know I'm not even close to this type of a problem yet with my small
cluster (both test and production clusters) - but it would be great if
something like that could appear in the cluster HEALTHWARN, if Ceph
could
Can you run the Fio test again but with a queue depth of 32. This will probably
show what your cluster is capable of. Adding more nodes with SSD's will
probably help scale, but only at higher io depths. At low queue depths you are
probably already at the limit as per my earlier email.
From:
On 03/05/2015 07:19 PM, Josh Durgin wrote:
client.libvirt
key:
caps: [mon] allow r
caps: [osd] allow class-read object_prefix rbd_children, allow rw
class-read pool=rbd
This includes everything except class-write on the pool you're using.
You'll need that so that a copy_up
I've found commit 9b9a682fe035c985e416ee1c112fa58f9045a27c and I see
that when 'osd heartbeat use min delay socket = true' it will mark the
packet with DSCP CS6. Based on the setting of the socket in
msg/simple/Pipe.cc is it possible that this can apply to both OSD and
monitor? I don't understand
On Mon, 9 Mar 2015, Karan Singh wrote:
Thanks Guys kernel.pid_max=4194303 did the trick.
Great to hear! Sorry we missed that you only had it at 65536.
This is a really common problem that people hit when their clusters start
to grow. Is there somewhere in the docs we can put this to catch
Yehuda Sadeh-Weinraub yehuda@... writes:
If you're using apache, then it filters out zero Content-Length.
Nothing much radosgw can do about it.
You can try using the radosgw civetweb frontend, see if it changes anything.
Thanks, only no difference...
Req:
PUT /mssCl/ HTTP/1.1
Host:
Yeah, I was thinking about that and will be the alternative for me too...
Regards.
Italo Santos
http://italosantos.com.br/
On Friday, March 6, 2015 at 18:20, ghislain.cheval...@orange.com wrote:
Message d'origine
De : CHEVALIER Ghislain IMT/OLPS
Hello Georgios,
I thought which had some admin alternative to do that, but I realised don’t
have once the bucket belongs to a specify user. So the alternative is, after
create the user authenticate with created credentials to create the bucket.
Thanks
Regards.
Italo Santos
On Mon, Mar 9, 2015 at 2:28 PM, Samuel Just sj...@redhat.com wrote:
You'll probably have to recreate osds with the same ids (empty ones),
let them boot, stop them, and mark them lost. There is a feature in the
tracker to improve this behavior: http://tracker.ceph.com/issues/10976
-Sam
Thanks
Hello Community need help to fix a long going Ceph problem.
Cluster is unhealthy , Multiple OSDs are DOWN. When i am trying to restart
OSD’s i am getting this error
2015-03-09 12:22:16.312774 7f760dac9700 -1 common/Thread.cc: In function 'void
Thread::create(size_t)' thread 7f760dac9700 time
Hi,
On 09/03/2015 04:06, kenmasida wrote :
I have resolved the problem,thank you very much。 When I use ceph-fuse to
mount the client,it work well.
Good news but can you give the kernel version of your client cephfs OS?
Like you, I had one problem with cephfs in the client side and it come
*Check Max Threadcount:* If you have a node with a lot of OSDs, you may be
hitting the default maximum number of threads (e.g., usually 32k),
especially during recovery. You can increase the number of threads using
sysctl to see if increasing the maximum number of threads to the maximum
possible
Great Karan.
On Mon, Mar 9, 2015 at 9:32 PM, Karan Singh karan.si...@csc.fi wrote:
Thanks Guys kernel.pid_max=4194303 did the trick.
- Karan -
On 09 Mar 2015, at 14:48, Christian Eichelmann
christian.eichelm...@1und1.de wrote:
Hi Karan,
as you are actually writing in your own book,
hello All,
I just setup single node ceph with no replication to
familiarize with ceph.
using 2 intel S3500 SSD 800 Gb and 8Gb RAM and 16 core CPU.
Os is ubuntu 14.04 64 bit ,kbd is loaded (modprobe kbd)
When running bonniee++ against /dev/rbd0
it shows a seekrate of 892.2/s.
How
2015-03-10 3:01 GMT+08:00 Sage Weil s...@newdream.net:
On Mon, 9 Mar 2015, Karan Singh wrote:
Thanks Guys kernel.pid_max=4194303 did the trick.
Great to hear! Sorry we missed that you only had it at 65536.
This is a really common problem that people hit when their clusters start
to grow.
Jian,
Thanks for the clarification. I'll mark traffic destined for the
monitors as well. We are getting ready to put our first cluster into
production. If you are interested we will be testing the heartbeat
priority to see if we can saturate the network (not an easy task for
40 Gb) and keep the
Only OSD calls set_socket_priority().
See https://github.com/ceph/ceph/pull/3353
On Tue, Mar 10, 2015 at 3:36 AM, Robert LeBlanc rob...@leblancnet.us wrote:
I've found commit 9b9a682fe035c985e416ee1c112fa58f9045a27c and I see
that when 'osd heartbeat use min delay socket = true' it will mark
we use `rados export poolA /opt/zs.rgw-buckets` export ceph cluster pool
named poolA into localdir /opt/ .and import the directroy /opt/zs.rgw-buckets
into another ceph cluster pool named hello , and following the error :shell
rados import /opt/zs.rgw-buckets hello --create[ERROR]
Steffen W Sørensen stefws@... writes:
Response:
HTTP/1.1 200 OK
Date: Fri, 06 Mar 2015 10:41:14 GMT
Server: Apache/2.2.22 (Fedora)
Connection: close
Transfer-Encoding: chunked
Content-Type: application/xml
This response makes the App say:
S3.createBucket, class S3, code
Hi,
I have a 4 node ceph cluster and the operating system used on the nodes is
Ubuntu
14.04.
The ceph cluster currently has 12 osds spread across the 4 nodes. Currently
one of the nodes has been restored after an operating system file system
corruption
which basically made the node and the osds
45 matches
Mail list logo