My Plex Media Server for the last 3 months has been running a little strange it will work for about 1 1/2 days then crash, restart the server and same issue. My program guide data is always blank and i will have to Manually retrieve the data after a crash.
Also when transcoding a file it will say something like plex transcoder has crashed, file not found or something like that.
I’ve included a link to my logs to see if this can be fixed
Thanks in advance. dropbox.com/sh/ayc8tyusayceomo/AACYgeoMkKWIeeC87lEz4l7Va?dl=0
My Plex Media Server for the last 3 months has been running a little strange it will work for about 1 1/2 days then crash, restart the server and same issue. My program guide data is always blank and i will have to Manually retrieve the data after a crash.
Also when transcoding a file it will say something like plex transcoder has crashed, file not found or something like that.
I’ve included a link to my logs to see if this can be fixed
Thanks in advance. dropbox.com/sh/ayc8tyusayceomo/AACYgeoMkKWIeeC87lEz4l7Va?dl=0
The logs show loads of errors like this
Feb 27, 2018 20:37:45.426 [6660] ERROR - Thread: Couldn't add a new thread to the pool of size 1: boost::thread_resource_error: An attempt was made to load a program with an incorrect format
These have been seen before on systems having Bitdefender installed. Could you tell me if it is ?
If it is, please remove and use alternative security software and see if it makes a difference
If it is not installed, then we will need to get extra diagnostics and I will advise you when I get your response
I don’t have bitdefender install on my server, I just have trend micro maximum security installed as my anti virus.
Thanks
The other circumstance when this was seen before was when there was insufficient memory. One would need to monitor memory usage by Plex Media Server.exe as well as available memory within windows. The Plex Media Server.exe process, being a 32-bit process would have a limit of 2Gb memory - normally of course it uses much much less and any very high usage could be due to a memory leak
Does the server stay up running for many days between restarts ?
I see you have a Premium Music Library. Is the music structured well in folders for artists and albums ? A flat structure of hundreds of tracks could give rise to overload as it would try to identify albums by selecting combinations of tracks
If there is memory issue with usage climbing over time (eg due to memory leak) - one would need tools such as SysInternals VMMap to monitor the memory usage by the process
See Process Monitor - Sysinternals | Microsoft Learn
Looking at the logs again - I do see references to memory issues
Feb 27, 2018 21:02:57.517 [13744] DEBUG - Saving original media file to C:\Users\Owner\AppData\Local\Plex Media Server\Cache\PhotoTranscoder\cf\cf4ab35f172c00cab343d3adfcc162ff5bca3905.jpg
Feb 27, 2018 21:02:57.522 [13744] ERROR - Format [JPEG] - DIB allocation failed, maybe caused by an invalid image size or by a lack of memory
Feb 27, 2018 21:02:57.522 [13744] ERROR - Error resizing an image, we don't trust what we cached [C:\Users\Owner\AppData\Local\Plex Media Server\Cache\PhotoTranscoder\cf\cf4ab35f172c00cab343d3adfcc162ff5bca3905.jpg]
Does the server stay up running for many days between restarts ?
Yes it does its on 24/7 as my Plex server
I see you have a Premium Music Library. Is the music structured well in folders for artists and albums ? A flat structure of hundreds of tracks could give rise to overload as it would try to identify albums by selecting combinations of tracks
My music is in folders of either Artist and albums and also just albums
I’ve installed Process Monitor and also VMMap i don’t know how to use them and I’ve selected Plex media server on VMMap like in the screenshot of other post. I will run this all day and export the files tonight for you to look at.
VMMap has the option to show a timeline and auto refresh it as in my screenshot in the linked thread
You can expand / contract the timeline to allow you to zoom in or zoom out and see it from the beginning
@sa2000 said:
So what is the maximum number of days between reboots ?
and between restarts of Plex Media Server ?
Maximum number of days can been 15- 30 days depends on a crash, restarts of plex media server only a few days but that is only because of the crashes, i would normally have the server running 24/7 and never restart the pc only when i have an issue with plex and would either restart PMS or restart the PC
ok its in timeline view i’ll run it for about 8 hours as im going to work and see how it goes with the export to give to you tonight my time.
if you need anything else from me just let me know
Thanks.
That memory usage is reasonable and there is a very slight climb in the timeline graph
Need to watch the totals for committed and private and the timeline over a number of days - and until you get these errors in the logs
Any sign of the 2 error messages in the Plex Media Server.log and Plex Media Server.x.log files ?
Super Finder XT from FSL is an excellent and solid text search product http://freesoftland.sytes.net/products.html
Searching for DIB allocation failed
and Couldn't add a new thread to the pool of size 1
I’ve got everything loaded and just waiting for a crash and will post a screenshot of VMMap and log files and will search for the errors listed above when the crash happens, thanks for you help once again.
@puddness said:
I’ve got everything loaded and just waiting for a crash and will post a screenshot of VMMap and log files and will search for the errors listed above when the crash happens, thanks for you help once again.
Better to periodically search because it could run for some time after first logging of such an error as was seen in your logs. Your oldest log file had the issue - of course only when you have access to the logs
Might be an idea to increase the number of log files (and a bit late now to suggest it - but there was a new beta released this morning)
@puddness said:
I’ve got everything loaded and just waiting for a crash and will post a screenshot of VMMap and log files and will search for the errors listed above when the crash happens, thanks for you help once again.
Better to periodically search because it could run for some time after first logging of such an error as was seen in your logs. Your oldest log file had the issue - of course only when you have access to the logs
Might be an idea to increase the number of log files (and a bit late now to suggest it - but there was a new beta released this morning)
Will do,
I’ve downloaded the new Beta update and so far so good and I’ve also changed the log files too,
We could also increase the number of log files
To increase number of log files to 50
RegEdit
and select HKEY_CURRENT_USER\Software\Plex, Inc.\Plex Media Server
and right click on Plex Media Server and select New >/ REG_DWORD (32-bit) and create an item and >name it LogNumFiles and give it a value of decimal >50 for 50 log files
you can go back to default log files by deleting LogNumFiles
Hi @sa2000 ,
Early this morning i had my first of 3 or 4 crashes. About 5 Days up time for the Plex Media Server
I’ll upload the log files and I also did the search for
Searching for DIB allocation failed
and
Couldn’t add a new thread to the pool of size 1
@puddness said:
Hi @sa2000 ,
Early this morning i had my first of 3 or 4 crashes. About 5 Days up time for the Plex Media Server
I’ll upload the log files and I also did the search for
Searching for DIB allocation failed
and
Couldn’t add a new thread to the pool of size 1
Thanks a lot for you help so far and in the past too.
Thanks for the logs.
First error arose on Mar 03, 2018 at 08:17:35 which was about 36 hours after the launch on 28 Feb 2018 19:41
I don;t know if there was a misunderstanding but i need to relate that errorand the time to the information that VMMap captured. I need to see timeline of memory used by the Pex Media Server process and if it climbed and at what time - and what it was at the time of the first and subsequent errors in the log. To make use of the timeline, would need to know when VMMap was launched and then to have a full view of the timeline (zoomed out so we have full view) of memory used from timeline zero to current time
VMMap screen can be captured and also the vmmap data can be saved
I need to know how much memory Plex Media Server.exe was using at 08:17:35 on 3rd March and if was too high - at what points it climbed
what time was the VMMap file you uploaded created ?
If you click near the end / at the end of the timeline where it shows 300368.0 it should display in the bottom left snapshot time. Need to work out what that means - can you tell me what it shows for you when you click inside the graph area at 0 and at 300368 and at the yello peak
I expect the zero to be close to when you launched PMS which was 28 Feb at 19:41
and the 300368 to be close to 07:07 on 4th March when you captured the logs
