Huge disks reads while 'idle'

Plex Media Server NAS & Devices
1

Server Version#: 1.18.6.2368
Unraid Version#: 6.8.2
Docker: plexinc/pms-docker

I’m fairly new to both Unraid, Docker, and Plex so please bear with me if all of the info to answer my question isn’t all here at the initial post.

I’m trying to figure out why I have such huge disk reads ~180MB/s for several hours at a time. See image below for the large orange rectangle.

image
image1282×390 35.5 KB

Here are some of my observations.

  • I’m 99% sure it’s Plex/Docker related because that dip in the right side of the disk usage and processor usage (~16h30) was when I ‘paused’ the docker.
  • When I restart the docker, it starts to behave normally again.
  • I’m pretty sure that it’s not related to scheduled tasks as my setting are fairly minimal AND only suppose to occur between 02h00 and 05h00.
  • It’s not a parity check, I haven’t even set up my parity drive yet.
  • No one is using it at that time (no one is home), and I’ve checked with current activity through plex as well as active streams through Unraid.
  • Could this have something to do with the amount of RAM cached at that time? The time frame for when it reaches it’s max appears to be the same.

Please help. I would really like to avoid undue wear on my HDDs and heat from the extra CPU usage. TIA!

2

Based on that start time of 2am, it could be the scheduled tasks. Check under that in your PMS settings and see what you have enabled.

3

If you have any large 4K hevc movies, a low power or older cpu, and have enabled chapter/video thumbnails, it could take many hours to process.

That said, if it is “the butler” doing scheduled tasks, it should stop around the configured time.

4

That time triggered me to look at my scheduled task too. The below is what I’ve got enabled. Even if it were a scheduled task causing it, shouldn’t it stop at 05h00 as I indicated above?

  • Backup database every three days
  • Optimize database every week
  • Remove old bundles every week
  • Remove old cache files every week
5

99% of my library is 4k remuxes BUT I don’t have the thumbnails enabled. I’ve already read about how much resources that takes up, so I disabled it fairly early on. I definitely didn’t want it making thumbnails for my 8TB of movies.

And yeah, the 2AM thing triggered me to think it was a scheduled task, but the fact that it continued for a good 11h hours after it was supposed to stop, made me think it might also be something else.

6

if you can ssh into your box and run run htop or top you should be able see the exact commandline of the process, then you can watch and see if it is constantly stuck on the same file(s).

7

Sorry, missed that you had looked into that already. If you can provide your PMS logs that include that time frame, I can check into what PMS was doing.

8

This is more to diagnose it while it’s happeningm, right? I restarted my server shortly after making my original post, and it’s been running normally so far. I guess I can use this the next time it happens.

9

To view my PMS Logs, do I just go to Manage -> Console in the webGUI?, I can only see what’s happening from I open it up. So I’m probably looking in the wrong place. I restarted my server shortly after posting this so that it’d stop whatever it was doing. Would the logs for this still be there? If so, can you direct me how to locate it?

10

Settings -> Manage -> Troubleshooting -> Download logs

That will produce a zip file. All the important logs are in there. The logs can store quite a bit of info so depending on how active your server was, the info might still be there.

11

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.