Jump to content

Very slow reponse times when loading images and media (Docker)


dotcom

Recommended Posts

I have a quite beefy server, 2 cpu 8 cores each 128GB RAM running docker. This has been running completely fine and very fast for a while. About 2 weeks ago I upgraded to (current) latest docker image and zero hiccups, still speedy as can be.

 

About 3 days ago randomly the ui would be slow and larger media items would take time. I've done network tests and they seem to be fine, doesn't appear to be server. Load is nothing.

 

example of request log from embyserver.txt

 

Observations
 

Pings on my network to the machine are consistently under 10ms.

A secure copy wirelessly to my laptop is about 44MB/s consistently

 

 

Seems that a restart of the container does speed it up for a short time but rather quickly it bogs down again. Any suggestions where to look further? I don't see any errors or other issues.

TIA

partial-embyserver.txt

Edited by dotcom
Link to comment
Share on other sites

Also attached the partial log from before I kept, thought I attached it in the first place but I forgot to click upload.

Edited by dotcom
Link to comment
Share on other sites

I started another container from scratch using a different volume for /config and everything is snappy again. I would like to troubleshoot my old config to see why this happened in the first place in case it happens again. Would also be nice to have my customizations and watch history back too.

Link to comment
Share on other sites

No, I’m talking a docker volume. Just a different directory on my main system. I only have one 24 TB raid 5 volume that reads and writes around 300mbps. My guess is there is an issue with the Database or something causing the slowness. If I mount my old config in the same container slowdown happens within the first couple minutes of starting. Nothing in any log to indicate why.

Link to comment
Share on other sites

Happy2Play

Attached. If I restart the container things seem decently snappy for a few minutes then everything slows down post 5-10 mins

This log shows the database being purged at "2020-01-24 04:49:55.501" on version 4.4.0.8.  Looks like 542 items being removed along with the perspective /config/metadata/library/ information and possibly re-added but can't tell with the log provided.  The slow response appears to happen right after.

 

Don't know what is changing with your setup that would cause this purge though.

 

 

Anything I can try on this or to increase verbosity?

You can enable Debug logging.   Dashboard-Logs

 

Note you have to restart your server to start or stop this logging level, when enabled or disabled.

Link to comment
Share on other sites

Here is an updated log with debug logging turned on. I removed the old log, started the container, clicked around then waited a few minutes. at 2020-01-29 03:22:12.818 things really slow down with a response clicking on an episode at over 102 seconds. At 2020-01-29 03:25:48.476 I click on a different episode of the same series which takes 78 seconds for just the first response back (nothing but spinner is showing still) then around 2020-01-29 03:27:57.172 everything else for the episode loads. Playback is also affected for higher bitrate shows, as well as downloads.

I don't care too much about this anymore because I mapped a different directory to config which is very snappy and happy again. I just lose all my watch history and customizations.

I'm happy to help narrow this down so it doesn't happen to me or someone else again if you'd like.

embyserver.txt

Edited by dotcom
Link to comment
Share on other sites

Happy2Play

Don't know what is going on with your server, but it is still purging items from the database.

2020-01-29 18:42:12.353 Debug BaseItem: Removed item:

2020-01-29 18:42:12.727 Info App: Deleting path

4.5 hours later still

2020-01-29 23:20:24.731 Info App: Deleting path
Link to comment
Share on other sites

It’s always doing that, some sort of bug was hit they caused this and has it in a loop. My other container works just fine on the same machine with a different config path

Link to comment
Share on other sites

Should I just call this a fluke and kill the old one off entirely? I am only continuing this rabbit hole in hopes of helping others that run into the same issue.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...