Hi Craig, Thanks for replying.
When i started that osd, Ceph Log from "ceph -w" warns pgs 7.9d8 23.596,
23.9c6, 23.63 can't recovery as pasted log.
Those pgs are "active+degraded" state.
#ceph pg map 7.9d8
osdmap e102808 pg 7.9d8 (7.9d8) -> up [93,49] acting [93,49] (When start
osd.21 then pg
Hello,
I was having problems with a node in my cluster (Ceph v0.80.7/Debian
Wheezy/Kernel 3.12), so I rebooted it and the disks were relabled when
it came back up. Now all the symlinks to the journals are broken. The
SSDs are now sda, sdb, and sdc but the journals were sdc, sdd, and sde:
root@cep
I am actually curious about one more thing.
In the image -> rbd case, is rbd_secret_uuid config option really used? I
am running nova-compute as a non-root user, so virsh secret shouldn't be
accessible unless we get it via rootwrap. I had to make ceph keyring file
readable to the nova-compute user
Thanks. I found the commit on git and confirms 0.80.7 fixes the issue.
On Friday, October 24, 2014, Josh Durgin wrote:
> On 10/24/2014 08:21 AM, Xu (Simon) Chen wrote:
>
>> Hey folks,
>>
>> I am trying to enable OpenStack to use RBD as image backend:
>> https://bugs.launchpad.net/nova/+bug/12263
Yeah, looks like it. If I disable the rbd ccahe:
$ tail /etc/ceph/ceph.conf
...
[client]
rbd cache = false
then the 2-4M reads work fine (no invalid reads in valgrind either).
I'll let the fio guys know.
Cheers
Mark
On 25/10/14 06:56, Gregory Farnum wrote:
There's an issue in master branch
It looks like you're running into http://tracker.ceph.com/issues/5699
You're running 0.80.7, which has a fix for that bug. From my reading of
the code, I believe the fix only prevents the issue from occurring. It
doesn't work around or repair bad snapshots created on older versions of
Ceph.
Wer
You can deploy multiple RadosGW in a single cluster. You'll need to setup
zones (see http://ceph.com/docs/master/radosgw/federated-config/). Most
people seem to be using zones for geo-replication, but local replication
works even better. Multiple zones don't have to be replicated either. For
ex
No, MDS and RadosGW store their data in different pools. There's no way
for them to access the other's data.
All of the data is stored in RADOS, and can be accessed via the rados CLI.
It's not easy, and you'd probably have to spend a lot of time reading the
source code to do it.
On Fri, Oct 24,
I have a number of PGs which are marked as incomplete. I'm at a loss for how to
go about recovering these PGs and believe they're suffering from the "lost
time" symptom. How do I recover these PGs? I'd settle for sacrificing the "lost
time" and just going with what I've got. I've lost the abilit
Hi,
Given HDFS is far from ideal for small files, I am examining the
possibility of using Hadoop on top Ceph. I found mainly one online resource
about it https://ceph.com/docs/v0.79/cephfs/hadoop/. I am wondering whether
there is any reference implementation or blog post you are aware of, about
ha
On Fri, Oct 24, 2014 at 8:17 AM, Dane Elwell wrote:
> Hi list,
>
> We're using the object storage in production and billing people based
> on their usage, much like S3. We're also trying to produce things like
> hourly bandwidth graphs for our clients.
>
> We're having some issues with the API not
On 10/24/2014 08:21 AM, Xu (Simon) Chen wrote:
Hey folks,
I am trying to enable OpenStack to use RBD as image backend:
https://bugs.launchpad.net/nova/+bug/1226351
For some reason, nova-compute segfaults due to librados crash:
./log/SubsystemMap.h: In function 'bool
ceph::log::SubsystemMap::sh
FWIW the specific fio read problem appears to have started after 0.86
and before commit 42bcabf.
Mark
On 10/24/2014 12:56 PM, Gregory Farnum wrote:
There's an issue in master branch temporarily that makes rbd reads
greater than the cache size hang (if the cache was on). This might be
that. (Ja
There's an issue in master branch temporarily that makes rbd reads
greater than the cache size hang (if the cache was on). This might be
that. (Jason is working on it: http://tracker.ceph.com/issues/9854)
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com
On Thu, Oct 23, 2014 at 5
Thanks! I'll continue with Apache 2.2 until the next release.
On Fri, Oct 24, 2014 at 8:58 AM, Yehuda Sadeh wrote:
> On Thu, Oct 23, 2014 at 3:51 PM, Craig Lewis
> wrote:
> > I'm having a problem getting RadosGW replication to work after upgrading
> to
> > Apache 2.4 on my primary test cluster
On Fri, Oct 24, 2014 at 8:47 AM, Sergey Nazarov wrote:
> Any update?
>
The short answer is that when the command is executed for second time,
the MDS needs to truncate the file zero length. The speed of truncate
a file is limited by the OSD speed. (creating file and write data to
the file are asy
Bonjour,
Maybe http://ceph.com/docs/giant/rados/troubleshooting/troubleshooting-mon/ can
help ? Joao wrote that a few month ago and it covers a number of scenarios.
Cheers
On 24/10/2014 08:27, HURTEVENT VINCENT wrote:
> Hello,
>
> I was running a multi mon (3) Ceph cluster and in a migration m
Hi,
October 24 2014 5:28 PM, "HURTEVENT VINCENT"
wrote:
> Hello,
>
> I was running a multi mon (3) Ceph cluster and in a migration move, I
> reinstall 2 of the 3 monitors
> nodes without deleting them properly into the cluster.
>
> So, there is only one monitor left which is stuck in probing
On Thu, Oct 23, 2014 at 3:51 PM, Craig Lewis wrote:
> I'm having a problem getting RadosGW replication to work after upgrading to
> Apache 2.4 on my primary test cluster. Upgrading the secondary cluster to
> Apache 2.4 doesn't cause any problems. Both Ceph's apache packages and
> Ubuntu's package
Any update?
On Tue, Oct 21, 2014 at 3:32 PM, Sergey Nazarov wrote:
> Ouch, I think client log is missing.
> Here it goes:
> https://www.dropbox.com/s/650mjim2ldusr66/ceph-client.admin.log.gz?dl=0
>
> On Tue, Oct 21, 2014 at 3:22 PM, Sergey Nazarov wrote:
>> I enabled logging and performed same t
Hello,
I was running a multi mon (3) Ceph cluster and in a migration move, I reinstall
2 of the 3 monitors nodes without deleting them properly into the cluster.
So, there is only one monitor left which is stuck in probing phase and the
cluster is down.
As I can only connect to mon socket, I d
Hey folks,
I am trying to enable OpenStack to use RBD as image backend:
https://bugs.launchpad.net/nova/+bug/1226351
For some reason, nova-compute segfaults due to librados crash:
./log/SubsystemMap.h: In function 'bool
ceph::log::SubsystemMap::should_gather(unsigned
int, int)' thread 7f1b477fe7
Hi list,
We're using the object storage in production and billing people based
on their usage, much like S3. We're also trying to produce things like
hourly bandwidth graphs for our clients.
We're having some issues with the API not returning the correct
statistics. I can see that there is a --sy
Hello Greg and John,
I used the patch on the ceph cluster and tried it again:
/usr/bin/ceph-mds -i th1-mon001 -c /etc/ceph/ceph.conf --cluster ceph
--undump-journal 0 journaldumptgho-mon001
undump journaldumptgho-mon001
start 9483323613 len 134213311
writing header 200.
writing 948332361
It's not stable at Firely for kvstore. But for the master branch, it's
should be no existing/known bug.
On Fri, Oct 24, 2014 at 7:41 PM, Andrey Korolyov wrote:
> Hi,
>
> during recovery testing on a latest firefly with leveldb backend we
> found that the OSDs on a selected host may crash at once,
Hi,
during recovery testing on a latest firefly with leveldb backend we
found that the OSDs on a selected host may crash at once, leaving
attached backtrace. In other ways, recovery goes more or less smoothly
for hours.
Timestamps shows how the issue is correlated between different
processes on s
dear cepher,
Today, I use mds to put/get files from ceph storgate cluster as it is
very easy to use for each side of a company.
But ceph mds is not very stable, So my question:
is it possbile to get the file name and contentes from OSD with radosgw
once MDS crash and how ?
___
27 matches
Mail list logo
