[ceph-users] Ceph fails to recover

2017-09-19 Thread Jonas Jaszkowic
Hi all, I have setup a Ceph cluster consisting of one monitor, 32 OSD hosts (1 OSD of size 320GB per host) and 16 clients which are reading and writing to the cluster. I have one erasure coded pool (shec plugin) with k=8, m=4, c=3 and pg_num=256. Failure domain is host. I am able to reach a HEA

Re: [ceph-users] Ceph fails to recover

2017-09-20 Thread Jonas Jaszkowic
ree`, and > `ceph health detail`? Thank you. > > On Tue, Sep 19, 2017 at 2:59 PM Jonas Jaszkowic > mailto:jonasjaszkowic.w...@gmail.com>> wrote: > Hi all, > > I have setup a Ceph cluster consisting of one monitor, 32 OSD hosts (1 OSD of > size 320GB per host)

Re: [ceph-users] Ceph fails to recover

2017-09-20 Thread Jonas Jaszkowic
your normal usage > and again during recover while testing the settings. For testing, you can > set it as high as you'd like (probably no need to go above 20 as that will > likely saturate your disks' performance) to get the PGs out of the wait > status and into

Re: [ceph-users] Ceph fails to recover

2017-09-20 Thread Jonas Jaszkowic
ulting Practice > Red Hat, Inc. > jelo...@redhat.com <mailto:jelo...@redhat.com> > +1 408-680-6959 > >> On Sep 20, 2017, at 08:26, Jonas Jaszkowic > <mailto:jonasjaszkowic.w...@gmail.com>> wrote: >> >> Thank you, that is very helpful. I didn’t know

Re: [ceph-users] Ceph fails to recover

2017-09-20 Thread Jonas Jaszkowic
sd_max_backfills > > The higher the number the higher the number of PGs being processed at the > same time. > > Regards > Jean-Charles LOPEZ > jeanchlo...@mac.com <mailto:jeanchlo...@mac.com> > > > > JC Lopez > Senior Technical Instructor, Global Storage Consulting

Re: [ceph-users] Ceph fails to recover

2017-09-20 Thread Jonas Jaszkowic
ll being up while it was marked down or that you had some other OSDs > in your cluster be marked down due to not responding or possibly being > restarted due to an OOM killer from the kernel. What is your current `ceph > status`? > > On Wed, Sep 20, 2017 at 11:52 AM Jonas Jaszkowic >

[ceph-users] Different recovery times for OSDs joining and leaving the cluster

2017-09-27 Thread Jonas Jaszkowic
Hello all, I have setup a Ceph cluster consisting of one monitor, 32 OSD hosts (1 OSD of size 320GB per host) and 16 clients which are reading and writing to the cluster. I have one erasure coded pool (shec plugin) with k=8, m=4, c=3 and pg_num=256. Failure domain is host. I am able to reach a

[ceph-users] Requests blocked in degraded erasure coded pool

2017-06-06 Thread Jonas Jaszkowic
I setup a simple Ceph cluster with 5 OSD nodes and 1 monitor node. Each OSD is on a different host. The erasure coded pool has 64 PGs and an initial state of HEALTH_OK. The goal is to deliberately break as many OSDs as possible up to the number of coding chunks m in order to evaluate the read p

Re: [ceph-users] Requests blocked in degraded erasure coded pool

2017-06-07 Thread Jonas Jaszkowic
at the > moment. It's a limit of the architecture. Thank you, this is good to know, particularly because I didn’t find anything about it on the documentation. - Jonas > Am 07.06.2017 um 21:40 schrieb Gregory Farnum : > > > > On Wed, Jun 7, 2017 at 12:30

Re: [ceph-users] Requests blocked in degraded erasure coded pool

2017-06-07 Thread Jonas Jaszkowic
m 07.06.2017 um 22:02 schrieb Gregory Farnum : > > > > On Wed, Jun 7, 2017 at 12:59 PM Jonas Jaszkowic > mailto:jonasjaszko...@googlemail.com>> wrote: >> If you set min_size 2 before taking the OSDs down, that does seem odd. > > I think I don’t get th

[ceph-users] Erasure Coding: Determine location of data and coding chunks

2017-06-19 Thread Jonas Jaszkowic
Hello all, I have a simple question: I have an erasure coded pool with k = 2 data chunks and m = 3 coding chunks, how can I determine the location of the data and coding chunks? Given an object A that is stored on n = k + m different OSDs I want to find out where (i.e. on which OSDs) the data c

Re: [ceph-users] Erasure Coding: Determine location of data and coding chunks

2017-06-20 Thread Jonas Jaszkowic
e the object and chunks > are stored on the cluster. > > Regards, > > Marko Sluga > Independent Trainer > > <1487020143233.jpg> > > W: http://markocloud.com <http://markocloud.com/> > T: +1 (647) 546-4365 > > L + M Consulting Inc. > Ste 212, 2121

[ceph-users] Erasure Coding: Wrong content of data and coding chunks?

2017-06-20 Thread Jonas Jaszkowic
I am currently evaluating erasure coding in Ceph. I wanted to know where my data and coding chunks are located, so I followed the example at http://docs.ceph.com/docs/master/rados/operations/erasure-code/#creating-a-sample-erasure-coded-pool

Re: [ceph-users] Erasure Coding: Wrong content of data and coding chunks?

2017-06-20 Thread Jonas Jaszkowic
> with the exact same values. Thanks for the help! > > On Tue, Jun 20, 2017, 6:48 AM Jonas Jaszkowic <mailto:jonasjaszkowic.w...@gmail.com>> wrote: > I am currently evaluating erasure coding in Ceph. I wanted to know where my > data and coding chunks are located, so I &