As I'm planning a new cluster where to move all my virtual machine
(currently on local storage on each hypervisor) i would like to evaluate
the current IOPS on each server
Knowing the current iops i'll be able to know how many iops i need on ceph
I'm not an expert, do know know how to get this in
Hi,
the most accurate way should be to check on each hostmachine how much
IOPS are flowing through.
You can also visualize this with for example munin.
This way you can also see the peaks.
--
Mit freundlichen Gruessen / Best regards
Oliver Dzombic
IP-Interactive
mailto:i...@ip-interactive.de
Hello,
On Fri, 17 Jun 2016 09:10:10 +0200 Gandalf Corvotempesta wrote:
> As I'm planning a new cluster where to move all my virtual machine
> (currently on local storage on each hypervisor) i would like to evaluate
> the current IOPS on each server
>
> Knowing the current iops i'll be able to k
Thanks for the reply, the service is still showing as failed. How to bring
the osds service up. Ceph osd tree shows all osds as UP.
[root@Admin ceph]# systemctl restart ceph-osd@osd.2.service
[root@Admin ceph]# systemctl status ceph-osd@osd.2.service
● ceph-osd@osd.2.service - Ceph object storage
Hi Mark,
What overhead do you mean? Can it be negligible if I use 4KB (extremly,
same with I/O size) stripe/chunk size for making sure that all random I/O
will spreaded through all OSDs?
Anyway, I love coffee too :)
Best regards,
> Date: Thu, 16 Jun 2016 04:01:37 -0500
> From: Mark Nelson
> T
> Op 17 juni 2016 om 12:12 schreef Lazuardi Nasution :
>
>
> Hi Mark,
>
> What overhead do you mean? Can it be negligible if I use 4KB (extremly,
> same with I/O size) stripe/chunk size for making sure that all random I/O
> will spreaded through all OSDs?
>
Keep in mind that this involves ope
Hello,
what is the diff between below reweight command and which one is
preferable to use?
ceph osd rewight
ceph osd crush reweight
Thanks
Swami
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph
Hi,
Will someone please assist, I am new to cepph and I am trying to map image
and this happens:
cluster-admin@nodeB:~/.ssh/ceph-cluster$ rbd map data_01 --pool data
rbd: sysfs write failed
In some cases useful info is found in syslog - try "dmesg | tail" or so.
rbd: map failed: (13) Permission d
On Fri, Jun 17, 2016 at 2:01 PM, Ishmael Tsoaela wrote:
> Hi,
>
> Will someone please assist, I am new to cepph and I am trying to map image
> and this happens:
>
> cluster-admin@nodeB:~/.ssh/ceph-cluster$ rbd map data_01 --pool data
> rbd: sysfs write failed
> In some cases useful info is found i
Hi Ishmael
You have to apply correct client.keyring permission to a pool on which you
trying to create rbd image.
>From Admin/Management Node: - create a client keyring.
1. Admin Node:-- ceph auth get-or-create client.rbd mon 'allow r' osd 'allow
rwx pool=PoolA'[client.rbd1]key =
AQB
Hello.
I'm trying to debug why OSD does not getting up.
It stops at:
2016-06-17 12:28:55.174468 7f0e60fd78c0 -1 osd.6 6366 log_to_monitors
{default=true}
2016-06-17 12:28:55.185917 7f0e60fd78c0 0 osd.6 6366 done with init,
starting boot process
If I enable debug (debug osd = 20 debug ms =
Hi,
Thank you for the response but with sudo all it does is freeze:
rbd map data_01 --pool data
cluster-admin@nodeB:~/.ssh/ceph-cluster$ date && sudo rbd map data_01
--pool data && date
Fri Jun 17 14:36:41 SAST 2016
On Fri, Jun 17, 2016 at 2:01 PM, Ishmael Tsoaela
wrote:
> Hi,
>
> Will s
On Fri, Jun 17, 2016 at 2:37 PM, Ishmael Tsoaela wrote:
> Hi,
>
> Thank you for the response but with sudo all it does is freeze:
>
> rbd map data_01 --pool data
>
>
> cluster-admin@nodeB:~/.ssh/ceph-cluster$ date && sudo rbd map data_01 --pool
> data && date
> Fri Jun 17 14:36:41 SAST 2016
What'
2016-06-17 10:03 GMT+02:00 Christian Balzer :
> I'm unfamilar with Xen and Xenserver (the later doesn't support RBD, btw),
> but if you can see all the combined activity of your VMs on your HW in the
> dom0 like with KVM/qemu, a simple "iostat" or "iostat -x" will give you the
> average IOPS of a d
What did
sudo rbd -p data ls
show?
If it freezes too, issue is with pool itself (ceph health) or network
connectivity.
On 06/17/2016 03:37 PM, Ishmael Tsoaela wrote:
Hi,
Thank you for the response but with sudo all it does is freeze:
rbd map data_01 --pool data
cluster-admin@nodeB:~/.ss
Thank you very much - we will be testing this soon.
Jason
On 6/16/16, 11:11 PM, "Yan, Zheng" wrote:
>On Fri, Jun 17, 2016 at 5:03 AM, Jason Gress wrote:
>> This is the latest default kernel with CentOS7. We also tried a newer
>> kernel (from elrepo), a 4.4 that has the same problem, so I don'
Hi All,
please assist to fix the error:
1 X admin
2 X admin(hosting admin as well)
4 osd each node
cluster a04e9846-6c54-48ee-b26f-d6949d8bacb4
health HEALTH_ERR
819 pgs are stuck inactive for more than 300 seconds
883 pgs degraded
64 pgs stale
I don't think it's the symlink that's the problem, but the path
permissions being something other than open. That may be why you didn't
see this. I am hoping symlinks still work, as I know we will need them
for our application.
Jason
On 6/17/16, 1:50 AM, "ceph-users on behalf of Oliver Dzombic"
Hey cephers,
As a part of my budget this year I am starting the process of raising
awareness around Ceph in the LATAM region. One of the things I am
starting with is having the Ceph Community sponsor FISL in Porto
Alegre, Brazil (13-16 July).
Thankfully our sponsorship negotiations we have been b
Hi,
I found the following on testing performance -
http://tracker.ceph.com/projects/ceph/wiki/Benchmark_Ceph_Cluster_Performance
and have a few questions:
- By testing the block device Do the performance tests take the
overall cluster performance (how long it takes the data to repli
(please reply to the maillist)
Next step is check status for ceph (ceph status). If there is issues
with OSD/placement, IO may hang indefinitely.
If HEALTH_OK, try reduce pool min_size (ceph osd pool set data min_size
1), and retry.
On 06/17/2016 04:01 PM, Ishmael Tsoaela wrote:
cluster-ad
Sounds like you've got deleted objects in the cache tier getting flushed
(i.e., deleted) in the base tier.
-Greg
On Thursday, June 16, 2016, Christian Balzer wrote:
>
> Hello devs and other sage(sic) people,
>
> Ceph 0.94.5, cache tier in writeback mode.
>
> As mentioned before, I'm running a cr
Hey cephers,
For those who have been asking for an official Ceph Slack channel, we
now have a bridge between #ceph/#ceph-devel and corresponding Slack
channels.
https://ceph-storage.slack.com/signup
Right now the email domains that will auto-accept registration are
relatively limited, but any sl
Hello Greg,
The opposite, space is consumed:
http://i.imgur.com/ALBR5dj.png
I can assure you, in that cluster objects don't get deleted.
Christian
On Fri, 17 Jun 2016 08:57:31 -0700 Gregory Farnum wrote:
> Sounds like you've got deleted objects in the cache tier getting flushed
> (i.e., del
Oh, space available drops, not space consumed. Not sure then; caching
has changed a bunch since I worked with it.
On Fri, Jun 17, 2016 at 9:49 AM, Christian Balzer wrote:
>
>
> Hello Greg,
>
> The opposite, space is consumed:
>
> http://i.imgur.com/ALBR5dj.png
>
> I can assure you, in that cluste
On 17 Jun 2016 3:33 p.m., "Carlos M. Perez" wrote:
>
> Hi,
>
>
>
> I found the following on testing performance -
http://tracker.ceph.com/projects/ceph/wiki/Benchmark_Ceph_Cluster_Performance
and have a few questions:
>
>
>
> - By testing the block device Do the performance tests take t
On Fri, May 20, 2016 at 7:32 PM, Daniel Wilhelm wrote:
> Hi
>
>
>
> I relieved to have found a solution to this problem.
>
>
>
> The ansible script for generating the key did not pass the key to the
> following command line and sent therefore an empty string to this script
> (see monitor_secret).
Hi,
I am looking for a way to monitor the utilization of OSD journals - by
observing the utilization pattern over time, I can determine if I have over
provisioned them or not. Is there a way to do this?
When I googled on this topic, I saw one similar request about 4 years back. I
am wondering
Hi Wido,
Do you mean TCP connections overhead is on OSD nodes or on Ceph clients? If
it is the TCP connections on Ceph clients, I think the maximum number will
be not more than the number of OSDs and no matter of the chunk/stripe size
the number of connection will still same if the image is spread
On Fri, 17 Jun 2016 16:29:31 +0530 M Ranga Swami Reddy wrote:
> Hello,
> what is the diff between below reweight command and which one is
> preferable to use?
>
> ceph osd rewight
>
> ceph osd crush reweight
>
Never mind the quite recent and frequent discussions about this,
googling for
"ceph
Hello,
On Fri, 17 Jun 2016 14:51:08 +0200 Gandalf Corvotempesta wrote:
> 2016-06-17 10:03 GMT+02:00 Christian Balzer :
> > I'm unfamilar with Xen and Xenserver (the later doesn't support RBD,
> > btw), but if you can see all the combined activity of your VMs on your
> > HW in the dom0 like with
31 matches
Mail list logo
