Problem with inode, & can't acces yunohost if no admin connect directly to the server

[CptCurk]

What type of hardware are you using: Raspberry Pi 3, 4+
What YunoHost version are you running: latest 12.1.38
How are you able to access your server: The webadmin
SSH
Direct access via physical keyboard/screen
Are you in a special context or did you perform specific tweaking on your YunoHost instance ?: nothing specific, external drive for var & log

Describe your issue

Hello all

I got an issue. Every time my server lose internet connection or is shutdown due to I don’t know what yet, I lose access to my server and it’s not online anymore.

I have an issue that I see when connecting a screen to the rasp directly : inode problem apear and spam the consol.

I can solve that by connecting to my admin account directly on the rasp via keyboard. Then the inode problem disapear, and I have access back to my server. Usually some service are not running anymore, just run them via command line and all work… Till the next disconnect or shutdown…

Problem is that I’m not living were the server is and I can’t go every time this bug occurs. I’d love to solve this issue.
As I can’t connect yet, I don’t have any logs available.
What kind of info would I need to get to you, and with which command should run, next time I’m physicaly at my server ?

Thank you for helping if you can !

Share relevant logs or error messages

don’t have yet

[jarod5001]

@CptCurk

inode problem

This is more the result of power outage. The disk gets corrupt on power cut.

Does the external disk has good power supply? And a good data cable?

Install logwatch, it sends daily reports that may help you

Edit

@CptCurk

which command

journalctl -b -1 -n 100 --no-pager

[CptCurk]

Okay. It clearly could be the problem. Because I suspect that my hole house got some micro cut of electricity, it’s a secondary little cottage with poor cables and connections for long distances…

So I think I’ll buy a battery and plus the rasp to the battery which will be under constant load. That will 100% solves the issues of electricity microcuts.

Then I’ll see what happen. Meanwhile, I’ll run your command line tomorrow, and after that, I’ll unmount and run an fsck fix. I went there for 5 minutes, and problem got bigger as fail2ban seems to block the server. Fortunatly mail is passing through for now !

I’ll install logwatch. I have daily mails, but it’s always something like : ‘/etc/cron.daily/node_update:
/etc/cron.daily/node_update: line 3: /opt/node_n/node_update.sh: No such file or directory
run-parts: /etc/cron.daily/node_update exited with return code 127’

I got those since the launch of my server.