Urgent: ZFS problem with guest VM Drives - Loosing connection to host-frozen etc

[bitblue.lab]

I am having problems my zfs is configured by default in 2x2TB RAID mirror with 1 ssd 50GB cache zil and one 150GB Cache Read

Here is a screenshot what is happening after some time to VM-s and its increasing IO lag in Host of proxmox ve 4





proxmox-ve: 4.0-22 (running kernel: 4.2.3-2-pve)
pve-manager: 4.0-57 (running version: 4.0-57/cc7c2b53)
pve-kernel-4.2.2-1-pve: 4.2.2-16
pve-kernel-4.2.3-2-pve: 4.2.3-22
lvm2: 2.02.116-pve1
corosync-pve: 2.3.5-1
libqb0: 0.17.2-1
pve-cluster: 4.0-24
qemu-server: 4.0-35
pve-firmware: 1.1-7
libpve-common-perl: 4.0-36
libpve-access-control: 4.0-9
libpve-storage-perl: 4.0-29
pve-libspice-server1: 0.12.5-2
vncterm: 1.2-1
pve-qemu-kvm: 2.4-12
pve-container: 1.0-21
pve-firewall: 2.0-13
pve-ha-manager: 1.0-13
ksm-control-daemon: 1.2-1
glusterfs-client: 3.5.2-2+deb8u1
lxc-pve: 1.1.4-3
lxcfs: 0.10-pve2
cgmanager: 0.39-pve1
criu: 1.6.0-1
zfsutils: 0.6.5-pve6~jessie

 

[udo]

Hi,
which disk is ata1.00 which makes trouble?

dmesg | grep ata1\.00

Perhaps an consumer SSD which don't work well with multible writes/syncs without garbage collection?

Udo

 

[Erk]

Hi,
which disk is ata1.00 which makes trouble?

dmesg | grep ata1\.00

Perhaps an consumer SSD which don't work well with multible writes/syncs without garbage collection?

Udo

Are there problems with consumer grade SSDs on Proxmox? I was thinking of buying some. Specifically Samsung 850 EVO.

 

[bitblue.lab]

ata1.00 is drive in VM not in VMware Host.

Host has 2x2TB Mirrored with 1 partition SSD from one ssd as zil log and from other ssd one partition as cache with 150GB space.

Should I do anything to SSD-s when I attach them as zil and cache ? Any kind of format or alingment or something like that

 

[udo]

Hi,
if you look on the host with atop during this happens inside the VM, perhaps you see where the problem is.

Any clue in the logs?

Udo

 

[udo]

Are there problems with consumer grade SSDs on Proxmox? I was thinking of buying some. Specifically Samsung 850 EVO.

Hi,
this is not proxmox related and depends for the use cases of the SSD. If you look for an ceph journal ssd (or zfs write cache) you realy should not use an Samsung EVO drive!! (better use allways no EVO drive ;-)
For this cases use an DC grade SSD like Intel DC S3700 (or Samsung DC ???).
If you need an system-ssd without too much writes (VMs on different storage) you can use an simple SSD, but I prefer in this case also an different brand as Samsung (Intel, Corsair).


Udo