[SOLVED] Trouble with iSCSI and cluster node inconsistency

N

naps1saps

New Member
Feb 16, 2026
3
0
1
I have 3 iSCSI targets on 2 NASs. Two of targets (one on each NAS) are connecting fine on both nodes in my cluster.

Node 01 is having issues with one new target that is on the same device as another target that is working fine.
Node 02 is fine, no issues.

Node 01 - Upon troubleshooting I found a post to delete the target configuration under /etc/iscsi/nodes which works and it will connect and create a new entry automatically.
Node 01 - After rebooting the node, the issue returns.

Node 02 - Has no configuration folders under /etc/iscsi/nodes
I am very confused by this. I saw a post where the opposite was true that the node with the missing config folders was the one having trouble, but that is not the case here and the node operates fine with a guest disk located on one of the targets.

I am at a loss for what to do to fix this problem. I appreciate any help you can give.
A small backstory, Node01 was existing and Node02 is a relatively new add along with a q-device. Both have the latest updates and kernel running.

PS - I am also getting log spam about proxmox nodes not being able to connect to 172.17.0.1,3260 which is not a valid IP on my network. I'm not sure where it's getting that for both targets on one NAS. The other NAS does not have this IP under /etc/iscsi/nodes. I think this is unrelated. I am not using multi path. I also verified I don't see any odd subnets listed under the nodes or the NAS. I would like to fix this as well.
 
Last edited:
What version of PVE are you running on each node?
In PVE9 most of the database has moved to /var/lib/iscsi
How are you managing iSCSI sessions? Are you using PVE storage pool or manual configuration?
If former, is your cluster healthy? If later, are you sure that you set proper restart options?

It is possible that you have a problem with the network where the IP connectivity on one node is delayed and breaks iSCSI start on boot. You should examine your boot log sequence.

There is some useful information in this article: https://kb.blockbridge.com/technote/proxmox-lvm-shared-storage/

In particular, you may want to use the command sequence from this part to troubleshoot your connectivity and IP reporting: https://kb.blockbridge.com/technote...rage/#connect-all-hosts-to-the-shared-storage


Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
I'm fairly new to Proxmox and didn't realize I had installed a newer major version on my 2nd node. I've upgraded the first node from 8 to 9. The iSCSI seems to be connecting so far after reboots. I got the delay dialed in. I am hoping this resolves the VM io crash/panic issue as well on my main guest machine using those iscsi drives.

Also I am really really hoping this resolves an odd networking issue on my 2nd node. I moved a VM there for the first time (it has been idle previously for a few months) and now I lose all network connection to the host and guest suddenly randomly, however the guest is still communicating with external devices receiving video streams and saving files to an SMB share. The only way to fix it is to cycle the physical ethernet connection (unplug/replug) but the issue returns either soon after or a few days later. I don't think this is a hardware issue, I've used it as a server for years. If it persists, I will try a USB NIC. Really not sure what might be going on.
 
You must log in or register to reply here.
Top Bottom
Back