strange multipath/iscsi problem.

[mikes]

Hi.
multipath and iscsi works ok.
config:iscsi using volume directly=yes

Guest connected to multipath device is using only single path, when cache=none or directsync.
(the kvm process is realy using mpath device /proc/PID/fd shows it)

When guest is using writheback/writethrough the multipath works as expected.

main problem is: path goes down in case of cache=none and the guest will get ioerrors/crash

Thanks for any help

-------------------------------

proxmox-ve-2.6.32: 3.2-129 (running kernel: 2.6.32-30-pve)
pve-manager: 3.2-4 (running version: 3.2-4/e24a91c1)
pve-kernel-2.6.32-27-pve: 2.6.32-121
pve-kernel-2.6.32-30-pve: 2.6.32-130
pve-kernel-2.6.32-29-pve: 2.6.32-126
pve-kernel-2.6.32-26-pve: 2.6.32-114
lvm2: 2.02.98-pve4
clvm: 2.02.98-pve4
corosync-pve: 1.4.5-1
openais-pve: 1.1.4-3
libqb0: 0.11.1-2
redhat-cluster-pve: 3.2.0-2
resource-agents-pve: 3.9.2-4
fence-agents-pve: 4.0.5-1
pve-cluster: 3.0-12
qemu-server: 3.1-16
pve-firmware: 1.1-3
libpve-common-perl: 3.0-18
libpve-access-control: 3.0-11
libpve-storage-perl: 3.0-19
pve-libspice-server1: 0.12.4-3
vncterm: 1.1-6
vzctl: 4.0-1pve5
vzprocps: 2.0.11-2
vzquota: 3.1-2
pve-qemu-kvm: 1.7-8
ksm-control-daemon: 1.1-1
glusterfs-client: 3.4.2-1

edit:
the problem is simple to check. /proc/diskstats for given mpath and its blockdevices are not updated.
this is true for non multipath setup, and I guess even for non iscsi.
After you change cache=writheback/writethrough diskstats gets updated, and mpath for guest works.

 

[mikes]

no one on this? check /proc/diskstats if your numbers are too low, multipath is not working even with good setup. (this is valid only for "Use LUNs Directly" disks.)