New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Does any provider support xattrs on openvz nodes?
Antipatiko
Member
Looking forward to deploy a GlusterFS-based node.
http://www.gluster.org 
[root@hera dev]# mount
/dev/simfs on / type reiserfs (rw,usrquota,grpquota)
proc on /proc type proc (rw,relatime)
sysfs on /sys type sysfs (rw,relatime)
none on /dev type tmpfs (rw,relatime)
none on /dev/pts type devpts (rw,relatime)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw,relatime)
Comments
http://serverfault.com/questions/29615/how-to-mount-glusterfs-inside-a-openvz-container means any OpenVZ container should work
Just need fuse enabled. Any provider should do this on request. I believe most do not have it enabled by default.
Just get any KVM VPS, you should not have problems enabling whatever you want.
I for one would not want to enable any OpenVZ features which we have not extensively tested. Stability first.
Fuse is perfectly stable, I've not heard of anyone refusing it on OpenVZ.
+1
In my IT period, I've set up a glusterfs cluster of (then) 5000 nodes, 10.000 RPM, 4x2TB disks in mirror raid, due to the config (mirror and distribute) it had about 6000TB of storage... It was for the MRI, CT, Nuclear and other radiology scans... Performance still is damn good. Now with the recent gluster 3.3 release, they've added auto-self healing. There does not have to be a file operation for nodes to heal themselves after a crash.
The cluster now has trippled in nodes... I whish I could do a df -h or some benchmarks...
I'm sorry if I wasn't clear enough but I'm not trying to mount a volume. Actually, I'm trying to build a replicated volume consisting of openvz VPSes.
This is the error I'm getting:
Glusterfs is not supported on brick: calypso.nullroute.com.ar:/brick.
Setting extended attributes failed, reason: Operation not permitted.
yeah, but where's the fun?
Yes, we do. It's just a matter of mounting the filesystem on the openvz host node with xattrs enabled.
Of course, if you are using a FUSE filesytem you shouldn't need xattrs enabled on the host side anyway (you would set them yourself when you mount your fuse filesystem).
Has anyone successfully been able to pull this off? I've gotten FUSE enabled on a bunch of containers from different providers, but always end up with:
Glusterfs is not supported on brick: nodename:/grid.
Setting extended attributes failed, reason: Operation not permitted.
Also, is there anything to worry about as a provider in giving the sys_admin capability? I wonder if that's the trick.
If you feel uncomfortable giving sys_admin privileges to every VPS mounting Gluster — and most Admins probably do — you can slightly alter GlusterFS source code so that it stores its xattrs in unprivileged namespace: find . -regex '.*/[^/]*\.[ch]' -exec sed -i 's/"trusted\./"user./g' {} + \ && ./configure && make I've tested this with version 1.4.0qa92, which we use in production for over half a year now.From: http://serverfault.com/questions/29615/how-to-mount-glusterfs-inside-a-openvz-container
@Damian, I wonder if that's still valid. They're almost at version 3.4 now, so I bet tons of changes have occurred since version 1.4.0qa92.
Any ideas on what the sys_admin capability is anyway? Info on it is sparse.
Did anyone ever figure out a solution to this? Attempting the build from source with the changes from Damian did not work.