I have this strange issue since recently (don't know when that started) accessing my vms console on my other nodes when connected on the gui from the node1. My cluster has 3 nodes in total and the following is working :
-Accessing vm console of node1 from node1 gui (local)
-Accessing vm console of node3 from node1 gui
What is not working :
-Accessing vm console of node1 from node2 gui
-Accessing vm console of node1 from node3 gui
-Accessing vm console of node2 from node1 gui
-Accessing vm console of node2 from node3 gui
So yes I can access VMs of node3 from node1 but not the opposite.
Thats the kind of logs I get : (it seems to connect and then half sec after it disconnect)
tail -f /var/log/auth.log
May 2 14:04:27 pve2 sshd[5910]: Accepted publickey for root from 10.32.50.8 port 38288 ssh2: RSA SHA256:ilA1TWTUx+7uZvgzRyAIYUAlOPp2y4W7DJRN5LecPvg
May 2 14:04:27 pve2 sshd[5910]: pam_unix(sshd:session): session opened for user root(uid=0) by (uid=0)
May 2 14:04:27 pve2 systemd-logind[566]: New session 11 of user root.
May 2 14:04:29 pve2 sshd[5910]: Received disconnect from 10.32.50.8 port 38288:11: disconnected by user
May 2 14:04:29 pve2 sshd[5910]: Disconnected from user root 10.32.50.8 port 38288
May 2 14:04:29 pve2 sshd[5910]: pam_unix(sshd:session): session closed for user root
May 2 14:04:29 pve2 systemd-logind[566]: Session 11 logged out. Waiting for processes to exit.
May 2 14:04:29 pve2 systemd-logind[566]: Removed session 11.
tail -f /var/log/syslog (I know the time is different but I just retried and added this log to this thread):
May 2 14:18:14 pve2 pmxcfs[9345]: [status] notice: received log
May 2 14:18:14 pve2 pmxcfs[9345]: [status] notice: received log
May 2 14:18:14 pve2 systemd[1]: Started Session 16 of user root.
May 2 14:18:16 pve2 systemd[1]: session-16.scope: Succeeded.
May 2 14:18:16 pve2 systemd[1]: session-16.scope: Consumed 1.137s CPU time.
May 2 14:18:16 pve2 pmxcfs[9345]: [status] notice: received log
I can ssh from all the 3 nodes to all the 3 nodes.
Any idea where I should investigate ?
Time is synched between the 3 nodes with the same ntp server
pvecm status
Cluster information
-------------------
Name: Maison
Config Version: 5
Transport: knet
Secure auth: on
Quorum information
------------------
Date: Mon May 2 14:21:07 2022
Quorum provider: corosync_votequorum
Nodes: 3
Node ID: 0x00000002
Ring ID: 1.6d3
Quorate: Yes
Votequorum information
----------------------
Expected votes: 3
Highest expected: 3
Total votes: 3
Quorum: 2
Flags: Quorate
Membership information
----------------------
Nodeid Votes Name
0x00000001 1 10.32.50.6
0x00000002 1 10.32.50.7 (local)
0x00000003 1 10.32.50.8
-Accessing vm console of node1 from node1 gui (local)
-Accessing vm console of node3 from node1 gui
What is not working :
-Accessing vm console of node1 from node2 gui
-Accessing vm console of node1 from node3 gui
-Accessing vm console of node2 from node1 gui
-Accessing vm console of node2 from node3 gui
So yes I can access VMs of node3 from node1 but not the opposite.
Thats the kind of logs I get : (it seems to connect and then half sec after it disconnect)
tail -f /var/log/auth.log
May 2 14:04:27 pve2 sshd[5910]: Accepted publickey for root from 10.32.50.8 port 38288 ssh2: RSA SHA256:ilA1TWTUx+7uZvgzRyAIYUAlOPp2y4W7DJRN5LecPvg
May 2 14:04:27 pve2 sshd[5910]: pam_unix(sshd:session): session opened for user root(uid=0) by (uid=0)
May 2 14:04:27 pve2 systemd-logind[566]: New session 11 of user root.
May 2 14:04:29 pve2 sshd[5910]: Received disconnect from 10.32.50.8 port 38288:11: disconnected by user
May 2 14:04:29 pve2 sshd[5910]: Disconnected from user root 10.32.50.8 port 38288
May 2 14:04:29 pve2 sshd[5910]: pam_unix(sshd:session): session closed for user root
May 2 14:04:29 pve2 systemd-logind[566]: Session 11 logged out. Waiting for processes to exit.
May 2 14:04:29 pve2 systemd-logind[566]: Removed session 11.
tail -f /var/log/syslog (I know the time is different but I just retried and added this log to this thread):
May 2 14:18:14 pve2 pmxcfs[9345]: [status] notice: received log
May 2 14:18:14 pve2 pmxcfs[9345]: [status] notice: received log
May 2 14:18:14 pve2 systemd[1]: Started Session 16 of user root.
May 2 14:18:16 pve2 systemd[1]: session-16.scope: Succeeded.
May 2 14:18:16 pve2 systemd[1]: session-16.scope: Consumed 1.137s CPU time.
May 2 14:18:16 pve2 pmxcfs[9345]: [status] notice: received log
I can ssh from all the 3 nodes to all the 3 nodes.
Any idea where I should investigate ?
Time is synched between the 3 nodes with the same ntp server
pvecm status
Cluster information
-------------------
Name: Maison
Config Version: 5
Transport: knet
Secure auth: on
Quorum information
------------------
Date: Mon May 2 14:21:07 2022
Quorum provider: corosync_votequorum
Nodes: 3
Node ID: 0x00000002
Ring ID: 1.6d3
Quorate: Yes
Votequorum information
----------------------
Expected votes: 3
Highest expected: 3
Total votes: 3
Quorum: 2
Flags: Quorate
Membership information
----------------------
Nodeid Votes Name
0x00000001 1 10.32.50.6
0x00000002 1 10.32.50.7 (local)
0x00000003 1 10.32.50.8
Last edited:
