No one had this problem? I found a forum/mail list post from 2013 with
the same issue but no responses either.
Any pointers appreciated.
Daniel
On 2014-11-14 20:20, Daniel Hoffman wrote:
Hi All.
Running a Ceph Cluster (firefly) ceph version 0.80.5
We use ceph mainly for backups via the r
Dear all
i have one Ceph Firefily test cluster on Debian Wheezy too, i want to
upgrade ceph from Firefly to Giant, could you tell me how to do upgrade ?
i saw the release notes like below , bu ti do not know how to upgrade,
could you give me some guide ?
*Upgrade Sequencin
Ok. I will do that. Thanks
--Jiten
> On Nov 14, 2014, at 4:57 PM, Gregory Farnum wrote:
>
> It's still creating and storing keys in case you enable it later.
> That's exactly what the error is telling you and that's why it's not
> working.
>
>> On Fri, Nov 14, 2014 at 4:45 PM, JIten Shah w
Will do. We haven't really had any stability issues once the containers are
running - they are the same daemons that would run on your host OS, they
just happen to be in a container. We have seen a few cases during startup
where there are placement groups stuck inactive/peering, which prevents the
It's still creating and storing keys in case you enable it later.
That's exactly what the error is telling you and that's why it's not
working.
On Fri, Nov 14, 2014 at 4:45 PM, JIten Shah wrote:
> But I am not using “cephx” for authentication. I have already disabled that.
>
> —Jiten
>
> On Nov 1
You didn't remove them from the auth monitor's keyring. If you're
removing OSDs you need to follow the steps in the documentation.
-Greg
On Fri, Nov 14, 2014 at 4:42 PM, JIten Shah wrote:
> Hi Guys,
>
> I had to rekick some of the hosts where OSD’s were running and after
> re-kick, when I try to
But I am not using “cephx” for authentication. I have already disabled that.
—Jiten
On Nov 14, 2014, at 4:44 PM, Gregory Farnum wrote:
> You didn't remove them from the auth monitor's keyring. If you're
> removing OSDs you need to follow the steps in the documentation.
> -Greg
>
> On Fri, Nov
Hi Guys,
I had to rekick some of the hosts where OSD’s were running and after re-kick,
when I try to run puppet and install OSD’s again, it gives me a key mismatch
error (as below). After the hosts were shutdown for rekick, I removed the OSD’s
from the osd tree and the crush map too. Why is it
Ceph in Docker is very intriguing to me, but I understood that there were
still a number of stability and implementation issues. What is your
experience? Please post a link to your blog when you are done, I'd be
interested in reading it.
On Fri, Nov 14, 2014 at 4:15 PM, Christopher Armstrong
wrot
Hi Loic,
Thanks for the reply! I started implementing these first using ceph-deploy,
but ran into lots of issues because there is no upstart daemon running
within the containers. So while ceph-deploy would successfully log into the
containers remotely (I had to run sshd in the containers), it woul
Any other suggestions why several osds are going down on Giant and causing IO
to stall? This was not happening on Firefly.
Thank s
- Original Message -
On 11/14/2014 01:50 PM, Andrei Mikhailovsky wrote:
> Wido, I've not done any changes from the default settings. There are no
> fi
Will there be RPMs built for this release?
Thanks,
On Tue, Nov 11, 2014 at 5:24 PM, Sage Weil wrote:
> This is the first development release after Giant. The two main
> features merged this round are the new AsyncMessenger (an alternative
> implementation of the network layer) from Haomai Wang
I have identical regionmaps in both clusters.
I only created the zone's pools in that cluster. I didn't delete the
default .rgw.* pools, so those exist in both zones.
Both users need to be system on both ends, and have identical access and
secrets. If they're not, this is likely your problem.
Hi,
Take a look at the built in perf counters -
http://ceph.com/docs/master/dev/perf_counters/. Through this you can get
individual daemon performance as well as some cluster level statistics.
Other (cluster-level) disk space utilization and pool utilization/performance
is available through “c
Well I upgraded both clusters to giant this morning just to see if that would
help, and it didn’t. I have a couple questions though. I have the same
regionmap on both clusters, with both zones in it, but then i only have the
buckets and zone info for one zone in each cluster, is this right? Or d
Hi Guys,
I got 3 MONs and 3 OSD’s configured via puppet but when trying to run MDS on
the MON servers, I am running into the error, which I can’t figure out.
BTW, I already manually installed MDS on cephmon001 and trying to install the
second MDS via puppet to see if I can have multiple MDS’s
Hi,
Thanks for your replies :]
Indeed, I did not think about the /sys/class/leds, but unfortunately I have
nothing in there on my systems.
This is kernel related, so I presume it would be the module duty to expose leds
there (in my case, mpt2sas) ... that would indeed be welcome !
/sys/block is
Hi,
I 'm trying to collect performance data for Ceph
I 'm looking to run some commands .. on regular intervals. to collect data.
Apart from "ceph osd perf" . Are there other commands one can use.
Can I also track how much data is being replicated ?
Does Ceph maintain performance counters for
HI list,
I am running ceph-deploy with a non-standard cluster name.
At the activate stage of the OSD the data directory /var/lib/ceph/osd/-0
is not created so the mount fails., If I manually create the directory it works.
I can't quite figure out when the directory is supposed to be created.
- Original Message -
On 11/14/2014 01:50 PM, Andrei Mikhailovsky wrote:
> Wido, I've not done any changes from the default settings. There are no
> firewalls between the ceph cluster members and I do not see a great deal of
> network related errors either. There is a tiny amount of TX e
On 11/14/2014 01:50 PM, Andrei Mikhailovsky wrote:
> Wido, I've not done any changes from the default settings. There are no
> firewalls between the ceph cluster members and I do not see a great deal of
> network related errors either. There is a tiny amount of TX errors on the
> network interfa
Wido, I've not done any changes from the default settings. There are no
firewalls between the ceph cluster members and I do not see a great deal of
network related errors either. There is a tiny amount of TX errors on the
network interface, which accounts to 0.0001% of the total packets.
Regar
On 11/14/2014 05:54 AM, Florent Bautista wrote:
On 11/14/2014 12:52 PM, Alexandre DERUMIER wrote:
Unfortunately I didn't, do you think hdparm could wrong results ?
I really don't known how hdparm is doing his bench (block size ? number of
thread ?).
BTW, do you have also upgraded librbd on
On 14-11-14 13:26, Andrei Mikhailovsky wrote:
> Hello guys,
>
> Since upgrading my cluster to Giant from the previous stable release I
> started having massive problems with client IO. I've done the upgrade 2
> weeks ago and since then the IO on my ceph cluster has been unavailable
> 3 times alrea
Hello guys,
Since upgrading my cluster to Giant from the previous stable release I started
having massive problems with client IO. I've done the upgrade 2 weeks ago and
since then the IO on my ceph cluster has been unavailable 3 times already.
Quick info on my storage cluster - 3 mons, 2 osd
On 13-11-14 19:39, Stephan Seitz wrote:
>
Indeed, there must be something! But I can't figure it out yet. Same
controllers, tried the same OS, direct cables, but the latency is 40%
higher.
>
> Wido,
>
> just an educated guess:
>
> Did you check the offload settings of your
On 11/14/2014 12:52 PM, Alexandre DERUMIER wrote:
>>> Unfortunately I didn't, do you think hdparm could wrong results ?
> I really don't known how hdparm is doing his bench (block size ? number of
> thread ?).
>
>
> BTW, do you have also upgraded librbd on your kvm node ? (and restarted the
> vm
>>Unfortunately I didn't, do you think hdparm could wrong results ?
I really don't known how hdparm is doing his bench (block size ? number of
thread ?).
BTW, do you have also upgraded librbd on your kvm node ? (and restarted the vm)
- Mail original -
De: "Florent Bautista"
À: "Ale
Unfortunately I didn't, do you think hdparm could wrong results ?
On 11/14/2014 12:37 PM, Alexandre DERUMIER wrote:
> Do you have tried to bench with something like fio ?
>
>
> - Mail original -
>
> De: "Florent Bautista"
> À: "Alexandre DERUMIER"
> Cc: ceph-us...@ceph.com
> Envoyé:
Do you have tried to bench with something like fio ?
- Mail original -
De: "Florent Bautista"
À: "Alexandre DERUMIER"
Cc: ceph-us...@ceph.com
Envoyé: Vendredi 14 Novembre 2014 12:30:21
Objet: Re: [ceph-users] RBD read performance in Giant ?
Hi,
I don't think this is the proble
Hi,
I don't think this is the problem, because :
- I use kvm RBD disk for my VM, with cache=none option
- after upgrade, I didn't restart my VM to see performance drop (so rbd
cache may not have changed from firefly)
I put this in my config, restarted VM but nothing changed.
On 11/14/2014 12:20
can you try to disable rbd cache ? (which is enable by default in giant)
[client]
rbd_cache = false
- Mail original -
De: "Florent Bautista"
À: ceph-us...@ceph.com
Envoyé: Vendredi 14 Novembre 2014 11:54:19
Objet: [ceph-users] RBD read performance in Giant ?
Hi all,
On a testin
Hi,
On 14/11/2014 12:11, Mallikarjun Biradar wrote:
> Hi,
>
> Default deep scrub interval is once per week, which we can set using
> osd_deep_scrub_interval parameter.
>
> Whether can we reduce it to less than a week or minimum interval is one week?
You can reduce it to a shorter period. It is
Hi,
Default deep scrub interval is once per week, which we can set using osd_
deep_scrub_interval parameter.
Whether can we reduce it to less than a week or minimum interval is one
week?
-Thanks & regards,
Mallikarjun Biradar
___
ceph-users mailing lis
Hi all,
On a testing cluster, I upgraded from Firefly to Giant.
Without changing anything, read performance on a RBD volume (in a VM)
has been divided by 4 !
With Firefly, in the VM (/dev/vda is a Virtio RBD device) :
hdparm -Tt /dev/vda
/dev/vda:
Timing cached reads: 20568 MB in 1.99 sec
Hi Robert,
I've just been testing your ceph check and I have made a small modification to
allow it to adjust itself to suit the autoscaling of the units Ceph outputs.
Here is the relevant section I have modified:-
if line[1] == 'TB':
used = saveint(line[0]) * 1099511627776
Hi,
problem solved it was a very stupid firewall problem.
It was not configured correctly in the monitor part.
The thing that I don't understand is the reason why the OSD didn't
want to start if it was the monitor part with and issue?!
Thank you.
Regards.
Luca
On 11/13/2014 06:40 PM, Luca Ma
Hi All.
Running a Ceph Cluster (firefly) ceph version 0.80.5
We use ceph mainly for backups via the radosGW at the moment.
There had to be an account deleted/bucket removed which had a very large
number of objects and was about 60TB in space.
We have been monitoring it for days now, and the
Hello,
I see following message on calamari GUI.
--
New Calamari Installation
This appears to be the first time you have started Calamari and there are no
clusters currently configured.
3 Ceph servers are connected to Calamari, but no Ceph cluster has been created
What is the role of /var/lib/ceph/radosgw/ceph-radosgw.$ID?
Following the Simple guide and creating the directory it is empty during
runtime. Is this correct behaviour. The S3 gateway seems to be working
though.
Anthony
___
ceph-users mailing list
ceph-
Hi,
Probably the error is causing when salt tries to report the final state summary.
the packages are created. I used it and could install calamari.
now I could see the calamari gui.
Thank you.
On November 14, 2014 at 11:52:26 AM, idzzy (idez...@gmail.com) wrote:
Hi,
Sure, Thanks.
As describe
41 matches
Mail list logo