Cannot log in via GUI, ssh does not respond

amckenzie4

New Member
Jan 29, 2026
4
0
1
I just tried to connect to my proxmox system, and found that while the login page loads it won't accept my username and password. It spins for a while, then says "Login failed. Please try again." I've cleared the cache, tried a private window, and tried a different browser that's never been used with PVE before. I've confirmed I'm on PAM authentication (and also tested with Proxmox Authentication Server). The password worked this morning, so I'm reasonably confident that's right.


OK, let's try SSH: nope. SSH to the host hangs without ever prompting for a password. The system shows as online at my router, and responds to a ping. Telnet to port 22 shows that SSH is listening, it's just not prompting for a password. I've confirmed I can ssh to the backup server, so the problem isn't entirely the laptop I'm using, although I won't rule out that it's a part of the problem.

This happened once before, and I ended up hard power cycling it. Obviously, I'd prefer to avoid that. Has anyone else seen this before? If so, did you find an answer?
 
No signal on the monitor, no reaction to the keyboard.

(Sorry I didn't include that initially... I've only just gotten home where I can physically check.)
 
That's what I thought, but I was hoping this was something known.

A reboot brought it back to full functionality, but I haven't had time to go through the logs yet. The disks aren't full, but it could be an issue with a failing disk or memory. The system is only about 6 months old, so I wouldn't have expected that, but it's certainly possible.

I'll start digging into the logs tonight or tomorrow and see what I can find.

Thanks for your help!
 
That's what I thought, but I was hoping this was something known.
Imagine calling an auto shop and telling them that your car sometimes just stops running. You need to turn it off and on to get going, is this something well known?

Good luck in your research


Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
Well... there are programs I support that do similar things, where I can say "Oh, yes. That particular piece of lab equipment occasionally locks up. It's a known issue and the manufacturer doesn't have a fix" or "Yes, with that piece of custom software we've found that it crashes if Foxit is also running." (I support a lot of weird custom equipment in research labs.)

So it seemed conceivable that someone would see that set of symptoms and say "Check this log file for the following message to see if it's this known issue." Not likely, I grant you, but possible.
 
Proxmox Virtualization Environment is, indeed, a collection of custom packages. However, it runs on Debian Linux with an Ubuntu-derived kernel. Those two components, primarily the kernel, are what directly interact with your hardware. SSH is a standard Linux package that depends on the kernel for TCP/IP handling and general process execution. There is no PVE involvement here, nor is PVE packaging related to physical console access.

Based on the symptoms you described, the issue is at the OS or hardware level, not within the higher-layer PVE application stack. The fundamental components of your system are not functioning correctly. Possible causes include hardware failure, firmware/BIOS issues, PCI device problems, disk failure, or a kernel bug. It is far less likely that PVE itself has somehow locked up the kernel to the point of making the physical console completely unresponsive.

As for logs, there are two primary places one would check : "dmesg" and "journalctl". This is standard for any modern Linux system. Messages from physical console are sometimes helpful, but it sounds like you don't have those. Sometimes, with Enterprise hardware, you can find some clues in iDRAC/ILO.

PS if there was a well known way to lock up Linux Kernel by a userland application, Mr Torvalds would certainly like to hear about it.


Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
 
Last edited:
I personally would look for journalctl -u pveproxy.service limiting with --since to the time-frame where you had that error. If it was 5 minutes ago for example it is: journalctl -u pveproxy.service --since "5 minutes ago"