Re: [ceph-users] xattr on cephfs with dosattrib

2013-05-16 Thread Tim Mohlmann
oup:users:--- > group:300:rwx > mask::rwx > other::--- it is probably the mask wich is offending. There is not a single linux filesystem supporting filemask. I don't know is mask is defined / supported in the cp preserve options. try `cp -a --no-preserve=mask`. Tim O

Re: [ceph-users] xattr on cephfs with dosattrib

2013-05-16 Thread Tim Mohlmann
Hi, `cp -a` implies to preserve attributes. It is obvious that there are some attributes set, not supported by any linux filesystem. I have the same experience running `rsync -a` between a fat32 or ntfs filesytem to ext4 or btrfs, it can not preserve attributes. Try to copy the same files to a

Re: [ceph-users] OSD state flipping when cluster-network in high utilization

2013-05-15 Thread Tim Mohlmann
Just for my information / learning process: This ping checking, is het a process itself, or is it part of the OSD process? (Maybe as a sub-process). In that case one could play with nice settings? Eg: marking the process as a realtime, or just a very low nice value. Then, at least, it would not

Re: [ceph-users] Hardware recommendation / calculation for large cluster

2013-05-13 Thread Tim Mohlmann
May 2013 00:25:19 Dmitri Maziuk wrote: > On 2013-05-12 08:34, Tim Mohlmann wrote: > > As for choking the backplane: That would just slow things down a bit, am I > > right? > > A bit, a lot, or not at all -- I think IRL you'll have to test it under > your workload and see

Re: [ceph-users] Hardware recommendation / calculation for large cluster

2013-05-12 Thread Tim Mohlmann
Hi, On Sunday 12 May 2013 18:05:16 Leen Besselink wrote: > > I did see you mentioned you wanted to have, many disks in the same machine. > > Not just machines with let's say 12 disks for example. > > Did you know you need the CPU-power of a 1Ghz Xeon core per OSD for the > times when recovery

Re: [ceph-users] Hardware recommendation / calculation for large cluster

2013-05-12 Thread Tim Mohlmann
Hi, On Saturday 11 May 2013 16:22:15 Dimitri Maziuk wrote: > SuperMicro has a new 4U chassis w/ 72x3.5" drives (2/canister). You can > double the number of drives. (With faster drives you may be getting > close to chocking the expander backplane, though.) Just checked their site and those are awes

Re: [ceph-users] Hardware recommendation / calculation for large cluster

2013-05-12 Thread Tim Mohlmann
-daemon. > Gosh, I feel embarresed. This ectually was my main concern / bottleneck. Thanks for pointing this out. Seems Ceph really rocks in deploying affordable data clusters. Regards, Tim > On Sat, May 11, 2013 at 03:42:59PM +0200, Tim Mohlmann wrote: > > Hi, > > >

[ceph-users] Hardware recommendation / calculation for large cluster

2013-05-11 Thread Tim Mohlmann
ted brainstorming on some business opportunities. If this story would be feasible, I would use this information to make a business and investment plan and look for investors. Thanks and best regards, Tim Mohlmann ___ ceph-users mailing list ceph-users@l