Search results

1024 should be a good start with 12 OSDs.

Thats right, see https://pve.proxmox.com/wiki/Proxmox_VE_Kernel#Proxmox_VE_4.x

You need datacenter SSDs, here is a good start and a listing of some devices: https://www.sebastien-han.fr/blog/2014/10/10/ceph-how-to-test-if-your-ssd-is-suitable-as-a-journal-device/

Network speed is not the problem, the problem with "hard" is, that the kernel waits forever, if i.e. the NFS host is down.

We had the issue with NFS, as the NFS server had a freeze or network issue. Possible solution is to change NFS mount options from hard to soft like this (in /etc/pve/storage.cfg)

Same here too :(

Ceph ;)

Connect and disconnect solved the issue in most cases for us. But we've kicked DRBD for now as it was horrible unstable :-/

Should work with drbdmanage assign-resource

You need a third node for quorum. This can be a VM or something small. For live migration with 2 nodes you could use DRBD8.

Not the kernel develpment at all, but the RHEL kernels (which PVE uses in a patched version). This ist why newer 3.16 and also 4.0 kernels do not have this problem (and PVE 4.0 also not, depending on its 4.0 kernel). Imho upgrading to PVE 4 or using the backported kernel 3.16 or even the 4.0...

Here: http://pve.proxmox.com/wiki/DRBD9