Re: [ceph-users] Help with the Hammer to Jewel upgrade procedure without loosing write access to the buckets

2017-01-26 Thread George Mihaiescu
Hi Mohammed, Thanks for the hint, I think I remember seeing this when Jewel came out but I assumed it must be a mistake, or a mere recommendation but not a mandatory requirement because I always upgraded the OSDs last ones. Today I upgraded my OSD nodes in the test environment to Jewel and

Re: [ceph-users] osd_snap_trim_sleep keeps locks PG during sleep?

2017-01-26 Thread Samuel Just
Just an update. I think the real goal with the sleep configs in general was to reduce the number of concurrent snap trims happening. To that end, I've put together a branch which adds an AsyncReserver (as with backfill) for snap trims to each OSD. Before actually starting to do trim work, the

[ceph-users] Ceph Tech Talk in ~2 hrs

2017-01-26 Thread Patrick McGarry
Hey cephers, Just a reminder that the 'Getting Started with Ceph Development' Ceph Tech Talk [0] is start in about 2 hours. Sage is going to walk through the process from start to finish, so if you have coworkers, friends, or anyone that might be interested in getting started with Ceph, please

Re: [ceph-users] Ceph on Proxmox VE

2017-01-26 Thread David Turner
Is there a guide on how to add Proxmox to an existing ceph deployment? I haven't quite gotten it so that Proxmox can manage ceph, just look at it and access it. On Thu, Jan 26, 2017, 6:08 AM Martin Maurer wrote: > Hello Everyone! > > We just created a new tutorial for

Re: [ceph-users] MDS flapping: how to increase MDS timeouts?

2017-01-26 Thread John Spray
On Thu, Jan 26, 2017 at 8:18 AM, Burkhard Linke wrote: > HI, > > > we are running two MDS servers in active/standby-replay setup. Recently we > had to disconnect active MDS server, and failover to standby works as > expected. > > > The filesystem

Re: [ceph-users] Issue with upgrade from 0.94.9 to 10.2.5

2017-01-26 Thread Piotr Dałek
On 01/24/2017 03:57 AM, Mike Lovell wrote: i was just testing an upgrade of some monitors in a test cluster from hammer (0.94.7) to jewel (10.2.5). after upgrade each of the first two monitors, i stopped and restarted a single osd to cause changes in the maps. the same error messages showed up

[ceph-users] Ceph on Proxmox VE

2017-01-26 Thread Martin Maurer
Hello Everyone! We just created a new tutorial for installing Ceph Jewel on Proxmox VE. The Ceph Server integration in Proxmox VE is already available since three years and is a widely used component for smaller deployments to get a real open source hyper-converged virtualization and storage

Re: [ceph-users] Suddenly having slow writes

2017-01-26 Thread Mark Nelson
On 01/26/2017 06:16 AM, Florent B wrote: On 01/24/2017 07:26 PM, Mark Nelson wrote: My first thought is that PGs are splitting. You only appear to have 168PGs for 9 OSDs, that's not nearly enough. Beyond the poor data distribution and associated performance imbalance, your PGs will split

Re: [ceph-users] 1 pgs inconsistent 2 scrub errors

2017-01-26 Thread Eugen Block
Glad I could help! :-) Zitat von Mio Vlahović : From: Eugen Block [mailto:ebl...@nde.ag] From what I understand, with a rep size of 2 the cluster can't decide which object is intact if one is broken, so the repair fails. If you had a size of 3, the cluster would see 2

Re: [ceph-users] 1 pgs inconsistent 2 scrub errors

2017-01-26 Thread Mio Vlahović
> From: Eugen Block [mailto:ebl...@nde.ag] > > From what I understand, with a rep size of 2 the cluster can't decide > which object is intact if one is broken, so the repair fails. If you > had a size of 3, the cluster would see 2 intact objects an repair the > broken one (I guess). At least we

Re: [ceph-users] 1 pgs inconsistent 2 scrub errors

2017-01-26 Thread Eugen Block
Yes, we have replication size of 2 also From what I understand, with a rep size of 2 the cluster can't decide which object is intact if one is broken, so the repair fails. If you had a size of 3, the cluster would see 2 intact objects an repair the broken one (I guess). At least we didn't

Re: [ceph-users] 1 pgs inconsistent 2 scrub errors

2017-01-26 Thread Mio Vlahović
Hello, > From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On > Behalf Of Eugen Block > I had a similar issue recently, where I had a replication size of 2 (I > changed that to 3 after the recovery). Yes, we have replication size of 2 also... > ceph health detail > HEALTH_ERR 16 pgs

Re: [ceph-users] Inherent insecurity of OSD daemons when using only a "public network"

2017-01-26 Thread Willem Jan Withagen
On 13-1-2017 12:45, Willem Jan Withagen wrote: > On 13-1-2017 09:07, Christian Balzer wrote: >> >> Hello, >> >> Something I came across a while agao, but the recent discussion here >> jolted my memory. >> >> If you have a cluster configured with just a "public network" and that >> network being in

Re: [ceph-users] 1 pgs inconsistent 2 scrub errors

2017-01-26 Thread Eugen Block
I had a similar issue recently, where I had a replication size of 2 (I changed that to 3 after the recovery). ceph health detail HEALTH_ERR 16 pgs inconsistent; 261 scrub errors pg 1.bb1 is active+clean+inconsistent, acting [15,5] zgrep 1.bb1 /var/log/ceph/ceph.log* [...] cluster [INF] 1.bb1

[ceph-users] 1 pgs inconsistent 2 scrub errors

2017-01-26 Thread Mio Vlahović
Hello, We have some problems with 1 pg from this morning, this is what we found so far... # ceph --version ceph version 10.2.0 (3a9fba20ec743699b69bd0181dd6c54dc01c64b9) # ceph -s cluster 2bf80721-fceb-4b63-89ee-1a5faa278493 health HEALTH_ERR 1 pgs inconsistent

Re: [ceph-users] Replacing an mds server

2017-01-26 Thread Peter Maloney
Oh, it says Coming soon somewhere? (Thanks... and I found it now at http://docs.ceph.com/docs/master/rados/deployment/ceph-deploy-mds/ ) I wrote some instructions and tested them (it was very difficult... putting together incomplete docs, old mailing list threads, etc. and tinkering), and

Re: [ceph-users] SIGHUP to ceph processes every morning

2017-01-26 Thread Torsten Casselt
Nice, there it is. Thanks a lot! On 26.01.2017 09:43, Henrik Korkuc wrote: > just to add to what Pawel said: /etc/logrotate.d/ceph.logrotate > > On 17-01-26 09:21, Torsten Casselt wrote: >> Hi, >> >> that makes sense. Thanks for the fast answer! >> >> On 26.01.2017 08:04, Paweł Sadowski wrote:

Re: [ceph-users] SIGHUP to ceph processes every morning

2017-01-26 Thread Henrik Korkuc
just to add to what Pawel said: /etc/logrotate.d/ceph.logrotate On 17-01-26 09:21, Torsten Casselt wrote: Hi, that makes sense. Thanks for the fast answer! On 26.01.2017 08:04, Paweł Sadowski wrote: Hi, 6:25 points to daily cron job, it's probably logrotate trying to force ceph to reopen

[ceph-users] MDS flapping: how to increase MDS timeouts?

2017-01-26 Thread Burkhard Linke
HI, we are running two MDS servers in active/standby-replay setup. Recently we had to disconnect active MDS server, and failover to standby works as expected. The filesystem currently contains over 5 million files, so reading all the metadata information from the data pool took too long,