Hi,
We’re going to take down one OSD node for maintenance (add cpu + ram) which
might take 10-20 minutes.
What’s the best practice here in a production cluster running dumpling
0.67.7-1~bpo70+1?
Kind Regards,
David Majchrzak
___
ceph-users mailing li
On 06/13/2014 10:56 AM, David wrote:
Hi,
We’re going to take down one OSD node for maintenance (add cpu + ram) which
might take 10-20 minutes.
What’s the best practice here in a production cluster running dumpling
0.67.7-1~bpo70+1?
I suggest:
$ ceph osd set noout
This way NO OSD will be m
Thanks Wido,
So during no out data will be degraded but not resynced, which won’t interrupt
operations ( running default 3 replicas and a normal map, so each osd node only
has 1 replica of the data)
Do we need to do anything after bringing the node up again or will it resynch
automatically?
Ki
On 06/13/2014 11:18 AM, David wrote:
Thanks Wido,
So during no out data will be degraded but not resynced, which won’t interrupt
operations ( running default 3 replicas and a normal map, so each osd node only
has 1 replica of the data)
Do we need to do anything after bringing the node up again
Alright, thanks! :)
Kind Regards,
David Majchrzak
13 jun 2014 kl. 11:21 skrev Wido den Hollander :
> On 06/13/2014 11:18 AM, David wrote:
>> Thanks Wido,
>>
>> So during no out data will be degraded but not resynced, which won’t
>> interrupt operations ( running default 3 replicas and a normal
The thread creation depends on the OSD number per host, as well as the
cluster size. You have really a lot (40!!) OSDs on a single node, but the
good part is that you¹ve got a small cluster (only 4 nodes).
If you already run into the problem then the only way is to increase
pid_max. Remember to re
Thanks, John.
That seems will take care of monitors, how about osd? Any idea how to
change IP addresses without triggering a resync?
Fred
Sent from my Samsung Galaxy S3
On Jun 12, 2014 1:21 PM, "John Wilkins" wrote:
> Fred,
>
> I'm not sure it will completely answer your question, but I would
On 06/13/2014 01:41 PM, Fred Yang wrote:
Thanks, John.
That seems will take care of monitors, how about osd? Any idea how to
change IP addresses without triggering a resync?
IPs of OSDs are dynamic. Their IP is no part of the data distribution.
Simply renumber them and restart the daemon.
Wido,
So the cluster reference osd based on the hostname, or the GUID(hopefully)?
Note that I mentioned in original email the hostname associated to the IP
will also be changed as well, it will be as simple as changing IP and
restart osd? I remembered I tested in Dumpling a while ago and it didn't
I recently changed IP and hostname of an osd node running dumpling and had
no problems.
You do need to have your ceph.conf file built correctly or your osds won't
start. Make sure the new IPs and new hostname are in there before you
change the IP.
The crushmap showed a new bucket (host name) cont
On Fri, Jun 13, 2014 at 7:09 AM, Ke-fei Lin wrote:
> Hi list,
>
> I deployed a Windows 7 VM with qemu-rbd disk, and got an unexpected booting
> phase performance.
>
> I discovered that when booting the Windows VM up, there are consecutive ~2
> minutes that `ceph -w` gives me an interesting log lik
2014-06-13 21:23 GMT+08:00 Andrey Korolyov :
> On Fri, Jun 13, 2014 at 7:09 AM, Ke-fei Lin wrote:
> > Hi list,
> >
> > I deployed a Windows 7 VM with qemu-rbd disk, and got an unexpected
> booting
> > phase performance.
> >
> > I discovered that when booting the Windows VM up, there are consecuti
On Fri, Jun 13, 2014 at 5:50 PM, Ke-fei Lin wrote:
> 2014-06-13 21:23 GMT+08:00 Andrey Korolyov :
>
>> On Fri, Jun 13, 2014 at 7:09 AM, Ke-fei Lin wrote:
>> > Hi list,
>> >
>> > I deployed a Windows 7 VM with qemu-rbd disk, and got an unexpected
>> > booting
>> > phase performance.
>> >
>> > I di
2014-06-13 22:04 GMT+08:00 Andrey Korolyov :
>
> On Fri, Jun 13, 2014 at 5:50 PM, Ke-fei Lin wrote:
> > Thanks, Andrey.
> >
> > I tried `virsh domblkstat hda` (only this VM in whole cluster) and
got
> > these values:
> >
> > hda rd_req 70682
> > hda rd_bytes 229894656
> > hda wr_req 1067
> > hda
Hi,
while testint firefly i cam into the sitation where i had a client where
the latest dumpling packages where installed (0.67.9).
As my pool has hashppool false and the tunables are set to default it
can talk to my firefly ceph sotrage.
For random 4k writes using fio with librbd and 32 jo
Right now, no.
We could add a minimum read size to librbd when caching is enabled...
that would not be particularly difficult.
sage
On Fri, 13 Jun 2014, Ke-fei Lin wrote:
> 2014-06-13 22:04 GMT+08:00 Andrey Korolyov :
> >
> > On Fri, Jun 13, 2014 at 5:50 PM, Ke-fei Lin wrote:
> > > Thanks,
In my belief, lot of sequential small reads will be aggregated after
all when targeting filestore contents (of course if the moment of issuing
next one is not dependent on status of previous read, otherwise
they`ll be separated in time in such way that the rotating media
scheduler will not be able
2014-06-14 0:11 GMT+08:00 Sage Weil :
> Right now, no.
>
> We could add a minimum read size to librbd when caching is enabled...
> that would not be particularly difficult.
>
> sage
Thanks, so is it possible to set some options like *readahead* in librbd
or QEMU? Seems no docs mentioned this...
B
2014-06-14 0:25 GMT+08:00 Andrey Korolyov :
> In my belief, lot of sequential small reads will be aggregated after
> all when targeting filestore contents (of course if the moment of issuing
> next one is not dependent on status of previous read, otherwise
> they`ll be separated in time in such way
On Sat, 14 Jun 2014, Ke-fei Lin wrote:
> 2014-06-14 0:11 GMT+08:00 Sage Weil :
> > Right now, no.
> >
> > We could add a minimum read size to librbd when caching is enabled...
> > that would not be particularly difficult.
> >
> > sage
>
> Thanks, so is it possible to set some options like *readahe
Hey,
I did try this, it didn't work though, so I think I still have to patch
the kernel though, as the user_xattr is not allowed on tmpfs.
Thanks for the description though.
I think the next step in this is to do it all virtual, maybe on the same
hardware to avoid network.
Any problems with
Hey,
That sounds awful. Have you had any luck in increasing the performance?
Cheers,
Josef
Christian Balzer skrev 2014-05-23 17:57:
For what it's worth (very little in my case)...
Since the cluster wasn't in production yet and Firefly (0.80.1) did hit
Debian Jessie today I upgraded it.
Big m
Shayan Saeed writes:
>
>
>
> Hi,
> I am following the standard deployment guide for ceph firefly. When I try
to do the step 5 for collecting the key, it gives me warnings saying that
keyrings not found for bootstrap-mds, bootstrap-osd and admin due to which
the next step for deploying osds fai
Hello, there,
I am trying to run ceph from source code. configure, make and make install
worked fine. But after done these steps, I can't see the binary files in
/etc/init.d/. My current OS is Centos6.5. I also tried Ubuntu 12.04, the same
issue occurred which said "unknown job ceph..." when I
I don't know anybody who makes much use of "make install", so it's
probably not putting the init system scripts into place. So make sure
they aren't there, copy them from the source tree, and try again?
Patches to fix are welcome! :)
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.co
Hey,
Just examing what happened to an OSD, that was just turned off. Data has
been moved away from it, so hesitating to turned it back on.
Got the below in the logs, any clues to what the assert talks about?
Cheers,
Josef
-1 os/FileStore.cc: In function 'virtual int FileStore::read(coll_t,
The OSD did a read off of the local filesystem and it got back the EIO
error code. That means the store got corrupted or something, so it
killed itself to avoid spreading bad data to the rest of the cluster.
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com
On Fri, Jun 13, 2014 a
Hi Greg,
Thanks for the clarification. I believe the OSD was in the middle of a
deep scrub (sorry for not mentioning this straight away), so then it
could've been a silent error that got wind during scrub?
What's best practice when the store is corrupted like this?
Cheers,
Josef
Gregory Far
On Fri, Jun 13, 2014 at 5:25 PM, Josef Johansson wrote:
> Hi Greg,
>
> Thanks for the clarification. I believe the OSD was in the middle of a deep
> scrub (sorry for not mentioning this straight away), so then it could've
> been a silent error that got wind during scrub?
Yeah.
>
> What's best pr
Thanks for the quick response.
Cheers,
Josef
Gregory Farnum skrev 2014-06-14 02:36:
On Fri, Jun 13, 2014 at 5:25 PM, Josef Johansson wrote:
Hi Greg,
Thanks for the clarification. I believe the OSD was in the middle of a deep
scrub (sorry for not mentioning this straight away), so then it cou
I compile and run from the src build quite often. Here is my recipe:
$ ./autogen.sh
$ ./configure --prefix=/usr --sysconfdir=/etc --localstatedir=/var
--with-radosgw
$ time make
$ sudo make install
$ sudo cp src/init-ceph /etc/init.d/ceph
$ sudo cp src/init-radosgw /etc/init.d/radosgw
$ sudo ch
31 matches
Mail list logo
