All media server v.1.18.x updates cause server unreachable/unavailable after update

I’ve rolled back to 1.17 and i’m still crashing, can anyone help by looking at my logs? Its only certain content that I’ve watched before that’s crashing it.

Plex Media Server Logs_2020-01-18_14-28-07.zip (892.6 KB)

that failed for me so I fully removed the application reinstalled and rebuild libraries from scratch… remote access is still shaky or fails.

Like others I can’t upgrade to v1.18.x which makes me worry about the changes announced for TVDB integration for matching shows etc that specifies that after Jan 31 for this feature to work you will require v1.18.x

How can you use the Repair a Corrupt Database procedure if Plex is run on a Synology NAS? Is there an equivalent of SQLLite 3 Tools that can run on a NAS?

Doing the database rebuild as mentioned by others above seems to have worked. I was able to upgrade to the latest PMS after doing the rebuild. So far everything seems to be working as should.

1 Like

FYI. Latest update to this thread could be the root cause.

https://forums.plex.tv/t/http-localhost-32400-web-index-html-is-unavailable-but-plex-works-on-devices/

Day 5 and all is still well. Longest any version of 1.18.x has lasted on my server!

1 Like

Yeah, that’s my thread. I’ve been dealing with this for months. Well, as soon as 1.18 came out. I have a zillion audio files. Over 500,000 music, almost 200,000 in audio books. You name it. I always thought that was the problem. Well I guess it kind of is. I could remove the music and audio book libraries and 1.18.x would work find. But once you started adding the music back in you would get to a point where it would just no longer be accessible via WEB. I could still hit it on Apple TV or iOS, but not web. I’, just glad the engineers now see it as a problem and are working on it.

Exactly the same symptoms I was experiencing. I also have a HUGE audio library. Mine would bomb out any time a media scan started. Like yours, it would be mostly accessible from players, but I couldn’t access the actual server. Sometimes it would overload the processor so badly that my server wouldn’t be accessible to anything. It would not even be available to remote desktop session into.

Thought I would let you know that I just tested a new build this morning and it seems to have fixed the issue. So it won’t be too long before it gets into production.

1 Like

Thank you @tramp78 for all your help providing all the diagnostics that enabled us to find this bug.

The issue relates to database accesses and the more libraries you have the worse it gets.

3 Likes

I hope this is a solid fix, because I’m getting failures now with 1.17, rendering Plex useless to me. Irony of ironies, with my systems mainly down right now I’m running a very reduced set of libraries.

If you’re impatient, I rebuilt my instance from scratch on the latest 1.18 version, and its working fine. There must of been something in my 1.17 config or DB that screwed an inplace update to 1.18.

It has to do with the length of time 1.18 takes to poll a library. If you have a lot of libraries the time stacks up and it times out.

How do I dowload 1.17? It looks like this is the same issue I’m having, but they don’t provide links to older version, what piece of crap company doesn’t do that?

Look in your Plex directories for the updates folder. The older update packages persist until manually removed.

Keep in mind the following announcement:

To go back to an older version, you need to stop the process for the current version, uninstall it, and then install the old one. Your database will still be there. You don’t have to do an uninstall to go to a newer version, but you do to go to an older version.

The path to the updates
C:\Users[local user]\AppData\Local\Plex Media Server\Updates

What a pita…ok looks like I may actually be ok now. Plex running, data not lost.

PITA? Oh yeah. Trying doing it for a month on and off. LOL