Server Version#: 1.28.1.6041
Player Version#: 4.88.1
Whenever a backup is made of the database, Plex becomes unresponsive and Plex Web (and all other clients) can no longer connect to it. In Plex Web, the icon at the top that normally shows the number of streams playing and displays Activity, no longer displays anything (if I click on it, it only shows a smaller version of the dropdown with no information in it). And in the list of available servers, it shows that the server is unavailable or cannot connect.
This started happening fairly recently, and I think it has to do with the size of my database. Still, I think Plex should do a better job of being able to work with a large library, rather than just becoming unresponsive or not accepting connections.
Hey Chuck, thanks for your reply. This seems like something else though. I have the maintenance window happening overnight, and I’ll wake up in the morning and Plex has been unresponsive and unavailable for several hours.
Also, if I trigger a backup manually via the API, sometimes the backup happens without problem, and other times it will go unresponsive for an hour or more.
When the database is locked, I assume that will cause the behavior I was describing? How long should this last, even for a very large database?
@ChuckPa so this morning, I woke up to a notification that Plex had stopped responding at 6:04am. This was almost 2 hours later, and it was still down. So I had to do a restart of the service to get it back up again.
I am including the logs here, I downloaded them after the restart. I don’t know if that is helpful but maybe you can find something helpful in them.
Thanks for those logs.
Unfortunately, due to the number of users you have and all the notifications which are flying around in there, the 6:04 am time slot rolled off the backend.
The oldest timestamp is 6:28
So we can get a wider window?
Stop Plex
Edit Preferences.xml in any Linux text editor.
On the last line, before the closing /> , add
LogNumFiles="10"
It will look like this: SomePref="value" LogNumFiles="10" />
Save and restart PMS.
We might need to increase it again but, for now, this will increase the depth from 20 minutes to 40 minutes
Hi Chuck, thanks for this. I’m not in front of my computer but I’ll do this later today.
In the meantime, can you search the logs for a database backup, or do you need a specific time stamp? The reason I ask is, I know I did another attempted backup later in the morning. I’m just not sure if I did it before or after downloading these logs.
Either way, if this doesn’t help you find it, I’ll do another attempt in a little bit and send you the results. Thanks!
Hi Chuck, see if this is any better. I did another backup attempt, and after it went non-responsive, I let it sit for a few minutes, then restarted and downloaded the logs immediately. So this should include the backup attempt, which was done at around 2:05.
Also, I have recently done a manual vacuum of the database, in fact as early as a day or two ago. I can do it again, but I don’t think it will make much of a difference.
I don’t want to scare you, but this is a truly massive database. I know some that have larger libraries that don’t have problems, but I know this is one of the larger libraries I’m personally aware of.
The number of items isn’t scaring me. You should see my QA test library
How many files do you have indexed ?
That’s the select count(*) from media_parts;
It should be close to the find /dir1 /dir2 /dir3 -type f -print | wc -l
what does ls -lah "/var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Databases" look like?
I do see the backup running (which locks the DB).
Are you using the “Backup Database” function of Plex/web which exports the DB and downloads it to you?
40 minutes
