termproxy error "/usr/bin/termproxy 5900 --path /nodes/node1 --perm Sys.Console -- /usr/bin/pveceph install --version quincy"

[interdev]

Hello,
i have a 3 nodes proxmox cluster, i got the following error when trying to install ceph from the gui or when trying to access a container console

failed waiting for client: timed out
TASK ERROR: command '/usr/bin/termproxy 5900 --path /nodes/node1 --perm Sys.Console -- /usr/bin/pveceph install --version quincy' failed: exit code 1

can someone explain me what /usr/bin/termproxy does exactly and how to investigate this error

Thanks

 

[Guillermo Dewey]

Hello,
i have a 3 nodes proxmox cluster, i got the following error when trying to install ceph from the gui or when trying to access a container console

failed waiting for client: timed out
TASK ERROR: command '/usr/bin/termproxy 5900 --path /nodes/node1 --perm Sys.Console -- /usr/bin/pveceph install --version quincy' failed: exit code 1

can someone explain me what /usr/bin/termproxy does exactly and how to investigate this error

Thanks

I just had this just before my nfs connection was under heavy load. but cant find the relation

 

[fabian]

anything custom about your setup (e.g., a reverse proxy in front)? are you attempting to access the console/ceph installer while being connect to the node where the action is supposed to happen, or are you connecting using a different node in your cluster?

 

[Guillermo Dewey]

anything custom about your setup (e.g., a reverse proxy in front)? are you attempting to access the console/ceph installer while being connect to the node where the action is supposed to happen, or are you connecting using a different node in your cluster?

server is a stand alone with no cluster and access was from browser. it is a brand new install with only 5 vms and a simple nfs storage. no ceph or proxy in front.

the only particular thing that I can think of is that I was working locally on the LAN and then from home using openvpn with the same proxmox session. But I was able to work for sometime remotely. Problem ocurred and proxmox became unavailable (to admin) for 5 min until that error came into the console log and everthing became normal again.

only thing I noticed after that is that on a backup job is that Proxmox backups vm but containers generates an error. but let me post that on a different forum thread so we dont mix things

 

[fabian]

that sounds like maybe your VPN interfered somehow with the connection to termproxy. termproxy is the component responsible for handling connections to shell/console parts of PVE (host/guest).

 

[linkstat]

server is a stand alone with no cluster and access was from browser. it is a brand new install with only 5 vms and a simple nfs storage. no ceph or proxy in front.

the only particular thing that I can think of is that I was working locally on the LAN and then from home using openvpn with the same proxmox session. But I was able to work for sometime remotely. Problem ocurred and proxmox became unavailable (to admin) for 5 min until that error came into the console log and everthing became normal again.

only thing I noticed after that is that on a backup job is that Proxmox backups vm but containers generates an error. but let me post that on a different forum thread so we dont mix things

I have exactly the same problem.
But I'm pretty sure OpenVPN isn't the cause; I have several sites, which I manage via VPN, but the problem does not occur in those places... the only difference is that the problematic server is not part of any cluster... however I have made some IP or hostname change (or both) at some point in the past. Thus, I think the problem is related to the certificates; By the way, is there a way to completely reset this? (pvecm updatecerts -f seems to have no effect).

 

[fabian]

please provide "pveversion -v" and before and after the updatecerts run, "pvenode cert info"