I also doubt the internet, it is a fairly powerful rented server with a 1GB bandwidth. It may have been a one-time thing. I’ll keep checking to see if it happens again. I want to thank you again for the help.
It has happened to me again, I don’t have the logs because they rotate. What I have done has been to update the driver of the network controller since it was a quite old driver. I had the i40e 2.8.20-k and I have updated to 2.21.12. If this doesn’t work, I’ll look at how to change the driver to the e1000e
It happened to me again… around 10:45 p.m. to 10:50 p.m. I don’t know what to do now, I have updated to the Kernel and it keeps crashing… Let’s see if we can find a solution…
I’ve been checking the syslog and apparently everything just stops… I don’t understand why
EDIT: Wrong log
I attach the syslog and the dmesg in case you see something… it seems that the server does something strange to me at a certain time, everything restarts and I don’t understand why. It always happens to me at night.
I put the others dmesg
Investigating I think it’s the service unattended-upgrades… What this does is leave the machine hanging, I have stopped the process and I have uninstalled it. We’ll see if it happens again.
Isn’t that, any ideas?
The only thing I can think of at this point is running extra database steps with PMS stopped.
Here is the tool I created to help with the task.
- Stop Plex
- On the host, as root, invoke the script
- It will figure out the host and where the database files are -or- report it can’t.
- From there, using the menu options: Check, Repair, Reindex
- Exit the script and keep the temporary files for now. They can be removed later.
- Start PMS and keep an eye on it.
1 Like
Sorry for the delay, I thought I had solved it but I’m still not the same… I just launched the script to see if it solves the problem. I have launched the script and selected options, 1, 2 and 3.
You wanted options 1 (Check) , 4 (Repair), then 3 (Reindex)
We won’t need vacuum for this.
It’s true… sorry for my clumsiness. I just launched it, I will be attentive to see if it is solved.
I have the same/similar issue so I am trying this suggestion, but I am getting an error when I run the script. I downloaded the script this morning. Thanks for any help.
➜ PlexDBRepair-master whoami
root
➜ PlexDBRepair-master ll
total 44K
-rwxr-xr-x 1 root root 32K Oct 31 14:14 DBRepair.sh
-rw-r–r-- 1 root root 965 Oct 31 14:14 License.md
-rw-r–r-- 1 root root 6.2K Oct 31 14:14 README.md
➜ PlexDBRepair-master ./DBRepair.sh
./DBRepair.sh: 341: Bad substitution
➜ PlexDBRepair-master
➜ plexmediaserver uname -a
Linux hermes 5.4.0-96-generic #109-Ubuntu SMP Wed Jan 12 16:49:16 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux
SORRY. I see that now.
I was using BASH but needed to switch back to SH.
I will push an update in a few minutes (0.3.2)
1 Like
Published
2 Likes
It has happened to me again, apparently the script has not worked. I upload the logs of my Plex server in DEBUG mode and the logs of my physical server (dmesg and dmesg.0, syslog and syslog.1)
Let’s see if we can find a solution, because there is no way 
These days I have tried to look for something in the syslog and dmesg and it seemed that the system “turned off” after launching the Apt Systemd Daily script but I removed it and apparently it is not that, I also disabled hardware transcoding in case that could affect somehow and it didn’t work either. I also updated the Kernel to its most recent version (5.15) and it didn’t work either, I also have the updates up to date (apt update, apt upgrade)… I don’t know what else is wrong.
I cannot find anything wrong with any of the logs.
-
PMS is running fine then stops all by itself.
-
Without error or reporting, it stops.
-
When it restarts, there is crash report.
-
The only thing I see is the file system needs cleaning.
I see where the RAID has to adjust its configuration (modules missing).
I suggest we try this a completely different way.
- Stop Plex
- Prepare to create a new “Test” server
sudo bash
systemctl stop plexmediaserver
cd /var/lib/plexmediaserver
mv Library Library.KEEP_THIS
systemctl start plexmediaserver
-
close and reopen the browser
-
Open an incognito window to PMS to start setup
-
You will be configuring a new server – HOWEVER —
– Give the server a Friendly Name “TEST SERVER”.
– This makes certain your other server is not confused -
Add your media as you have before.
-
Run with this for a bit to see if the host or something in the PMS DATA internally.
If I do this what happens to my other server? Can I continue watching my movies and series or when I move the .Library I can’t see anything?
Will this keep or loose your watch list?
What I am suggesting does the following:
-
It takes your existing server and sets it off to the side (safely)
-
It creates a new server, with nothing carried forward, to test with.
-
Now we can test and be certain if it’s the PMS installation or the computer.
Once we know which is which, we can take the next step.
If PMS is the problem, we can extract the watch history list from the old server and bring it forward into this new server.
Your mention of watchlist has given me an idea. I will add Watchlist importing to my DBRepair.sh utility.
I cannot do this test that you mention, on my server there is always someone viewing. Another clue is that it always happens to me in the same time slot, at night from 9:00 p.m. to 11:00 p.m.
