Cannot log into WebGUI-All VM inside working

[wahmed]

Hello,
I have 4 nodes Proxmox cluster. One of the node has all the VM and the other 3 for redundancy or when needed. Since last night i cannot log into 2 of the nodes. Node 1 and 2 i can login. All the VMs are in Node 4. When i login to WebGUI Node 1 or 2, i can see Node 1 and 2 online, but node 3 and 4 offline. When i try to login Node 4, it keeps saying Login failed, please try again. All the VMs in Node 4 are working though. I can even SSH to all nodes, no problem.
Any idea what is the issue here. No recent changes has been made in the Proxmox cluster.
Thanks!

 

[wahmed]

While browsing through the forum i came across a solution provided by Dietmar in another thread. I tried it and it worked. All my nodes are now visible online. From one of the node i did

#pvecm updatecerts --force

Any reason why the cert have to be update forcefully? A known bug?

The issue for me is SOLVED.

 

[cyclones]

I think this might be a bug.

I just completed moving all of our existing production servers and VMs to a new V3-based cluster. Rather than try to upgrade in place, we decided to take a new server, install 3.0 clean and create a new cluster and then start moving VMs and doing clean installs on the existing nodes and add them to the new 3.0 cluster. Four of the servers this worked fine, but for some reason one of them didn't work properly right away.

After finishing migrating the VMs and adding all of the clusters, I ran through a few steps to make sure everything was working and setup properly. When I tried a live migrate to one of the nodes it failed. Then I took a look at the Summary tab (I was logged into the web gui from another node) and saw that I couldn't see the Summary tab - eventually it just gave me a "Connection failure (0)" error. Next I tried to create a VM directly on that node and found out that it didn't let me create a VM because it didn't appear to have access to any of the three storage options, including local. Next I tried logging into the that node's web gui and got no data received error. My next step was to ssh into the node and that still worked fine, so then I restarted the cluster service. That didn't work, so next I did pveversion -v on every node and confirmed that the output was the same on all nodes.

Next I restarted each node in sequence, but that didn't fix any thing either, so then I looked around in the forums for any ideas and saw this and tried the solution on the troublesome node and it fixed it instantly.

I've been using Proxmox VE for a while now and haven't ever had this particular problem before, so if there is any other information that I could supply that would be helpful, just ask.

 

[wahmed]

I have been using Proxmox close to 2 years now. I have done live upgrade many times, but never had any issue. 2 of the nodes i have, were setup 1.5 years ago and still going. All of the nodes are upgraded to ver 3. Currently i have doing heavy testing CEPH usage with Proxmox. Some of the issue Proxmox having probably due to my CEPH test and trial but that should not affect Proxmox since the cluster itself is untouched except attaching CEPH as RBD Shared storage with Proxmox.
The way you described your issue thats pretty much how it went for my nodes.