Web GUI goes down, reboot requires manual fsck

[smibrandon+proxmox]

Hello all--

Newbie here (with both Proxmox and Linux, in general); I know just enough to be dangerous lol. Please be patient with me!

Anyways, I'm running my server. And after three weeks or so of uptime, I'll not be able to make any changes in the web GUI due to write-protection errors (yet no issues with containers or VM's writing to their own storage). Then, a few days later, the web UI goes down (I'm still able to SSH in, though). I'll reboot the server, and immediately, I get the command line telling me the file system requires a manual fsck. When I do this, it fixes the issues, reboots file, GUI is available, and no read/write issues.

In a nutshell, after a few weeks of uptime, almost systematically:
1) Read-only drive issues. then a couple days later,
2) GUI down. I'll reboot to fix,
3) manual fsck required

Note: I've searched extensively for each of these issues individually, but the threads are just FAR above my head to be able to gain anything from them. Plus, given the consecutive issues, I think it's a safe assumption they're related.

Thanks in advance!

 

[apoc]

I'll have a guess: you are booting from a USB stick? Is that correct?

 

[smibrandon+proxmox]

I'll have a guess: you are booting from a USB stick? Is that correct?

Negative. I have everything running on a 1-year old, 1TB HDD (yes, HHD, not SSD)

 

[bbgeek17]

At the time the system goes read-only there is a chance that something is written to log (journalctl). Of course there is also a chance that it cant write anything because the disk is ro.
You could change your syslog to ship the system log to a remote system or a different disk as part of troubelshooting, so you have things to review.
Without any logs, an educated guess is that you have bad hardware...
You also want to check the physical console of the server - it may have some info.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[apoc]

Negative. I have everything running on a 1-year old, 1TB HDD (yes, HHD, not SSD)

It was a try.
The symptom you describe sounded exactly what I was experiencing back at the time when I was booting from USB thumb drives.
At some time these went RO.
The system worked a long time until it ran out of memory. Then the UI stopped working and SSH + reboot was my only way to get it back.
Fsck on reboot was necessary, and sometime even failed to repair.

So your system is different and a view into the logs could reveal something. Without that there is still some chance that the system has a defect. It could be the disk, memory or something else.

Can you share more details about your setup?