Proxmox 7 Status?

[dkowis]

I saw a single commit for proxmox 7, and found no issues filed, except one.

Thought I'd start a thread for proxmox 7 status.

How is it?

[TheGrandWazoo]

I have been using the plugin for 6.x to 7.0 update and all 7.0 update. I have had no issue with Proxmox to TrueNAS versions 12.0-U3 - 12.0-U6.
I have not done a 7.0 to 7.1 yet but most likely I will just provide a patch update to remove fuzz patches.
Also I am testing out a non-git based repo update that the patches are included with the update. Also, testing to remove the librest-client-perl dependency because there is already a rest client with the Proxmox package.

[CMAPTu]

I am running 3-Node PVE 7.1-1 Cluster with installed FreeNAS ZFS over iSCSI interface on each of them. One of new features of 7.1-1 realese is new scheduler for backups planning - "pvescheduler". This is new daemon pvescheduler.service, it parses jobs from /etc/pve/jobs.cfg. If I run vzdump manualy from CLI or from Web-UI, it works fine. But if it is runned by new service, i get following messages in it's log:

file /etc/pve/storage.cfg line 21 (section 'zfspool') - unable to parse value of 'freenas_password': unexpected property 'freenas_password'
file /etc/pve/storage.cfg line 22 (section 'zfspool') - unable to parse value of 'freenas_use_ssl': unexpected property 'freenas_use_ssl'
file /etc/pve/storage.cfg line 23 (section 'zfspool') - unable to parse value of 'freenas_user': unexpected property 'freenas_user'
INFO: starting new backup job: vzdump 101 103 --quiet 1 --storage pbs --prune-backups 'keep-last=14' --compress zstd --mailnotification always --mode snapshot
INFO: skip external VMs: 103
INFO: Starting Backup of VM 101 (qemu)
INFO: Backup started at 2022-01-19 09:40:02
INFO: status = running
INFO: VM Name: dc1-spb
INFO: include disk 'virtio0' 'zfspool:vm-101-disk-0' 50G
ERROR: Backup of VM 101 failed - freenas: unknown iscsi provider. Available [comstar, istgt, iet, LIO] at /usr/share/perl5/PVE/Storage/ZFSPlugin.pm line 35.
INFO: Failed at 2022-01-19 09:40:02
INFO: Backup job finished with errors
TASK ERROR: job errors

[TheGrandWazoo]

Not sure about this. It seems the package or commands to apply the patches to ZFSPlugin.pm did not take. Is the patch/package applied to ALL nodes in the cluster? Not seeing the FreeNAS in the iscsi provider tells me it is not installed.
I need to look at my nodes and see if I am using 7.1 and see if the patches are not working.

[CMAPTu]

Yes, patch/package applied to all nodes via your repo:

echo "deb http://download.proxmox.com/debian/pve bullseye pve-no-subscription" > /etc/apt/sources.list.d/pve-no-subscription.list
wget http://repo.ksatechnologies.com/debian/pve/ksatechnologies-release.gpg -O /etc/apt/trusted.gpg.d/ksatechnologies-repo.gpg
echo "deb http://repo.ksatechnologies.com/debian/pve stable freenas-proxmox" > /etc/apt/sources.list.d/ksatechnologies-repo.list
apt update
apt upgrade
apt reinstall pve-manager pve-docs libpve-storage-perl
apt install freenas-proxmox

Workaround to make backups execute by schedule is to manually add strings to file /etc/pve/vzdump.cron as it was in proxmox before 7.1.

P.S. I applied your package BEFORE adding nodes to HA Cluster. Do not know, if it matters.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[dkowis]

I don't think it's resolved. Stale maybe, but it takes time... Go away stalebot.

[TheGrandWazoo]

I have been running a three node Proxmox 7.1-10 and 7.1-11with TrueNAS-Core and TrueNAS-Scale as the NAS and backing up to a TrueNAS-Core using NFS with the new mechanisms you have described and I have not had any issues. Backup jobs finish a I get a report of successful every time.
It just seems like something is not patching correctly due to this line
ERROR: Backup of VM 101 failed - freenas: unknown iscsi provider. Available [comstar, istgt, iet, LIO] at /usr/share/perl5/PVE/Storage/ZFSPlugin.pm line 35.

[TheGrandWazoo]

Closing due to no activity from original requester. If you are still having issues, please reopen or open a new issue.

[CMAPTu]

Hello, I've got one more problem after adding VM to HA. When I try to start it, I get such error:

task started by HA resource agent
TASK ERROR: freenas: unknown iscsi provider. Available [comstar, istgt, iet, LIO] at /usr/share/perl5/PVE/Storage/ZFSPlugin.pm line 35.

It seems to me that tis problem is the same as I reported pefore.

Please, do not close it due to no activity. Or, maybe mention it as a problem to fix in upcoming builds.

[CMAPTu]

Sorry for bothering you all this time. Restarting pve-ha-lrm solved all the problems.

[TheGrandWazoo]

So if in an HA cluster, does the pve-ha-lrm need to be restarted for it to work? I have not had to do that but if I need to add to the documentation then please let me know.

[TruckeeAviator]

Sorry to add to an old thread. I encountered the same problem as the OP. I am running PVE 7.3-4, TrueNAS 13.0-U3.1.

I applied the patches to each of the 5 nodes in the cluster and restarted the pve-ha-lrm, pve-ha-crm.

I found that in order for pvescheduler.service to complete the task automatically, without error. pvescheduler.service needs to be restarted on all nodes systemctl restart pvescheduler.service. This is only a problem for me since the nodes are never shutdown, otherwise a reboot would solve this problem.

Hope this helps someone.

[TheGrandWazoo]

@TruckeeAviator - Thank you for the additional information. I will update the README.md to reflect your recent find and fix.