Can't login via web GUI, SSH works

[jkilbreth]

I've been searching the threads and nothing I've tried is working...
I'm unable to login via the GUI (the GUI works, but I get a login failure when I use the root credentials).
I can ssh just fine with those same root credentials, but I'm not sure what to try next to get the GUI login to work.



Tried this, but it didn't fix it:

systemctl restart pve-manager

systemctl restart pveproxy

Saw some thread responses request the output of systemctl status pveproxy pvedaemon...

 pveproxy.service - PVE API Proxy Server
   Loaded: loaded (/lib/systemd/system/pveproxy.service; enabled; vendor preset: enabled)
   Active: active (running) since Wed 2021-07-28 21:40:26 EDT; 5min ago
  Process: 26815 ExecStartPre=/usr/bin/pvecm updatecerts --silent (code=exited, status=1/FAILURE)
  Process: 26925 ExecStart=/usr/bin/pveproxy start (code=exited, status=0/SUCCESS)
 Main PID: 26943 (pveproxy)
    Tasks: 4 (limit: 4915)
   Memory: 142.4M
   CGroup: /system.slice/pveproxy.service
           ├─26943 pveproxy
           ├─26944 pveproxy worker
           ├─26945 pveproxy worker
           └─26946 pveproxy worker

Jul 28 21:40:25 Ugin systemd[1]: Starting PVE API Proxy Server...
Jul 28 21:40:25 Ugin pvecm[26815]: unable to open file '/etc/pve/priv/authorized_keys.tmp.26924' - Input/output error
Jul 28 21:40:26 Ugin pveproxy[26943]: starting server
Jul 28 21:40:26 Ugin pveproxy[26943]: starting 3 worker(s)
Jul 28 21:40:26 Ugin pveproxy[26943]: worker 26944 started
Jul 28 21:40:26 Ugin pveproxy[26943]: worker 26945 started
Jul 28 21:40:26 Ugin pveproxy[26943]: worker 26946 started
Jul 28 21:40:26 Ugin systemd[1]: Started PVE API Proxy Server.

● pvedaemon.service - PVE API Daemon
   Loaded: loaded (/lib/systemd/system/pvedaemon.service; enabled; vendor preset: enabled)
   Active: active (running) since Sat 2021-07-17 19:00:15 EDT; 1 weeks 4 days ago
 Main PID: 4703 (pvedaemon)
    Tasks: 4 (limit: 4915)
   Memory: 165.4M
   CGroup: /system.slice/pvedaemon.service
           ├─  432 pvedaemon worker
           ├─ 3582 pvedaemon worker
           ├─ 4703 pvedaemon
           └─25325 pvedaemon worker

Jul 28 21:23:29 Ugin pvedaemon[432]: authentication failure; rhost=10.0.0.64 user=root@pam msg=error during cfs-locked 'authkey' operation: got lock request timeout
Jul 28 21:23:35 Ugin pvedaemon[3582]: authentication failure; rhost=10.0.0.64 user=root@pve msg=no such user ('root@pve')
Jul 28 21:23:49 Ugin pvedaemon[25325]: authentication failure; rhost=10.0.0.64 user=root@pve msg=no such user ('root@pve')
Jul 28 21:24:27 Ugin pvedaemon[3582]: authentication failure; rhost=10.0.0.64 user=root@pve msg=no such user ('root@pve')
Jul 28 21:27:24 Ugin pvedaemon[432]: authentication failure; rhost=10.0.0.64 user=root@pve msg=no such user ('root@pve')
Jul 28 21:40:00 Ugin pvedaemon[3582]: authentication failure; rhost=10.0.0.64 user=root@pve msg=no such user ('root@pve')
Jul 28 21:40:18 Ugin pvedaemon[432]: authentication failure; rhost=10.0.0.64 user=root@pam msg=error during cfs-locked 'authkey' operation: got lock request timeout
Jul 28 21:40:37 Ugin pvedaemon[432]: authentication failure; rhost=10.0.0.64 user=root@pam msg=error during cfs-locked 'authkey' operation: got lock request timeout
Jul 28 21:43:55 Ugin pvedaemon[432]: authentication failure; rhost=10.0.0.64 user=jkilbreth@pam msg=no such user ('jkilbreth@pam')
Jul 28 21:45:18 Ugin pvedaemon[25325]: authentication failure; rhost=10.0.0.64 user=root@pam msg=error during cfs-locked 'authkey' operation: got lock request timeout

The "no such user ('root@pve')" concerns me...

Output of systemctl status pve-manager...

pve-guests.service - PVE guests
   Loaded: loaded (/lib/systemd/system/pve-guests.service; enabled; vendor preset: enabled)
   Active: active (exited) since Sat 2021-07-17 19:00:40 EDT; 1 weeks 4 days ago
 Main PID: 6706 (code=exited, status=0/SUCCESS)
    Tasks: 0 (limit: 4915)
   Memory: 0B
   CGroup: /system.slice/pve-guests.service

Jul 17 19:00:18 Ugin pve-guests[7290]: start VM 104: UPID:Ugin:00001C7A:00002B0D:60F36102:qmstart:104:root@pam:
Jul 17 19:00:20 Ugin pvesh[6706]: Waiting for 15 seconds (startup delay)
Jul 17 19:00:35 Ugin pvesh[6706]: Starting CT 100
Jul 17 19:00:35 Ugin pve-guests[17922]: starting CT 100: UPID:Ugin:00004602:000031B3:60F36113:vzstart:100:root@pam:
Jul 17 19:00:35 Ugin pve-guests[7286]: <root@pam> starting task UPID:Ugin:00004602:000031B3:60F36113:vzstart:100:root@pam:
Jul 17 19:00:36 Ugin pvesh[6706]: Starting VM 101
Jul 17 19:00:36 Ugin pve-guests[18572]: start VM 101: UPID:Ugin:0000488C:00003218:60F36114:qmstart:101:root@pam:
Jul 17 19:00:36 Ugin pve-guests[7286]: <root@pam> starting task UPID:Ugin:0000488C:00003218:60F36114:qmstart:101:root@pam:
Jul 17 19:00:40 Ugin pve-guests[6706]: <root@pam> end task UPID:Ugin:00001C76:00002B0A:60F36102:startall::root@pam: OK
Jul 17 19:00:40 Ugin systemd[1]: Started PVE guests.

PVE version:
pve-manager/6.2-4/9824574a (running kernel: 5.4.34-1-pve)

Retroactive thanks for all the great you people (community and admins/mods) do..
Thanks in advance for the help I know is coming

 

[oguz]

hi,

The "no such user ('root@pve')" concerns me...

it shouldn't, since there's no such user. you probably want to log in as root@pam -- for that you need to pick Linux PAM standard authentication when logging in via GUI.

but this error message seems to be relevant:

Jul 28 21:23:29 Ugin pvedaemon[432]: authentication failure; rhost=10.0.0.64 user=root@pam msg=error during cfs-locked 'authkey' operation: got lock request timeout

there was a thread before with similar symptoms [0]

you can try my workaround from there:

systemctl stop pve-cluster
rm -f /var/lib/pve-cluster/.pmxcfs.lockfile
systemctl start pve-cluster

hopefully that fixes your problem.

PVE version:
pve-manager/6.2-4/9824574a (running kernel: 5.4.34-1-pve)

don't forget to upgrade to PVE7 [1] soon!

[0]: https://forum.proxmox.com/threads/login-failure.71684/
[1]: https://pve.proxmox.com/wiki/Upgrade_from_6.x_to_7.0

 

[mhakim]

thanks @oguz
it's worked like a charm.

 

[Heliart]

Let me add to this, I too have been having this login issue every few weeks, since moving to a new 7.1 install.
The suggested fix above works when this happens wichout having to reboot the server - so very happy with that.

 

[syd_d]

I came home from holiday and switched on my home lab and had exactly this problem. Found this post and looking at the solution here remembered I have a cluster (i stuck pve on an old laptop, which I'd forgotten to switch on, and created a cluster). So before doing anything I switched on the laptop and lo and behold no problem logging in. I also noticed that containers and VM's (not on the laptop) that were meant to autostart hadn't; they did once the laptop came up.
I have a question though. What if the laptop should die and I'd rebooted my main pve? Presumably I'd be unable to login through the GUI, would the above work with the node missing?

 

[quasimodem]

login to the shell, make sure the firewall is disabled and not running.

 

[vaschthestampede]

I had the same problem, what caused it was a cable, used for ceph and as a secondary interconnect between two nodes.

Once the connection was re-established it was possible to enter a WebGUI.

 

[bummer]

you can try my workaround from there:

systemctl stop pve-cluster
rm -f /var/lib/pve-cluster/.pmxcfs.lockfile
systemctl start pve-cluster

hopefully that fixes your problem.

This helped me too!
Thank you!