[SOLVED] Problem during Migration with gluster filesystem.

[fiona]

Hi,

Could you please tell how can I get the patch?

the fix is included in versions pve-qemu-kvm>=6.2.0-6. You can check with pveversion -v if a newer version is installed. Otherwise, check for updates. If there are no updates, check your package repository configuration (for the currently installed version of Proxmox VE).

 

[imran.tee]

Hi,

the fix is included in versions pve-qemu-kvm>=6.2.0-6. You can check with pveversion -v if a newer version is installed. Otherwise, check for updates. If there are no updates, check your package repository configuration (for the currently installed version of Proxmox VE).

thanks for you reply. I am checking

 

[andy77]

Hi, would just like to add that the problem is also recreatable with "ZFS over iscsi" LUNs.

 

[fiona]

Hi,

Hi, would just like to add that the problem is also recreatable with "ZFS over iscsi" LUNs.

please post the full migration log, any messages from /var/log/syslog around the time the issue happens and the output of pveversion -v and qm config <ID> for the affected VM.

 

[andy77]

Dec 12 14:14:20 node1 pmxcfs[2381]: [status] notice: received log
Dec 12 14:14:30 node1 QEMU[38582]: kvm: ../block/io.c:2847: bdrv_co_pdiscard: Assertion `num < max_pdiscard' failed.

This is happaning on a "zpool trim" command inside a VM, where the VM has a "ZFS over iSCSI" Disk attached.

pve-qemu-kvm: 5.2.0-6

With pve-qemu-kvm: 7.1.0-4 I cannot reproduce the issuce, so it seems to be fixed with > 6.2

 

[fiona]

pve-qemu-kvm: 5.2.0-6

With pve-qemu-kvm: 7.1.0-4 I cannot reproduce the issuce, so it seems to be fixed with > 6.2

Well, that's good then
Proxmox VE 6.x has been end-of-life for about half a year now, so I'm afraid no fix is getting backported there.

 

[andy77]

Proxmox VE 6.x has been end-of-life

Sure, just wanted to comment that it also occours with "ZFS over iSCSI" for other who may have that problem.