Hi together,

 

i'm trying to find out why the Plex Media Server is  dying on some unRaid boxes.

If you experience this problem, please let me know on which CPU your BOX is running.

 

I'm running on AMD and I have it. But two users I know, are running on Intel and they don't have it.

 

thanks for your interest

 

regards, spidi

 

I am running the PMS on an AMD Athlon II Neo 1.3 GHz (hp microserver) and I also have the problem.

Same here but it takes more than 8 hours before drop HP microserver 5gb ram

 

Sent from my LG-P990 using Tapatalk

I don't think that I have ever experienced this on my unRaid server with an AMD X2 250. How do I check for this?

 

Using "top" command, I get MEM 1813152k total,  1726044k used.

I've experienced the same problem on my test server running an AMD CPU.  Haven't tried it on my production server (which has an Intel CPU), as I'm still running unRAID 4.7.

Kevin

@SCSI

 

ps auxw | grep -i Plex\ Media\ Server | grep -v grep | grep -v com.plexapp 

 

if the 5th value is increasing, then you have the Problem.

Ok I don't think I have this problem. I ran the command after 30 minutes and the same info came up. Still using 9.3.5

 

nobody   28036  0.0  1.3 240000 24860 ?        Sl   Oct30   1:12 ./Plex Media Server

 

I then accessed the Plex Manager Web GUI and git below. I check again after 10 minutes the the value was the same.

 

nobody   28036  0.0  1.4 241024 25608 ?        Sl   Oct30   1:12 ./Plex Media Server

 

Update: Just checked again and values went lower.

nobody  28036  0.0  1.3 240000 24204 ?        Sl  Oct30  1:12 ./Plex Media Server

 

I experience the problem (I was the one that originally posted the thread over in the Plex forums).  Intel i3.

 

I'm no expert, but there's got to be some reason why some people experience the problem and others don't.  The people who aren't experiencing the problem... are you running any other Python apps (SAB or Sickbeard, perhaps)?  I noticed that PMS comes with its own Python, so maybe that could be a cause?  

 

On the other hand, I did notice that when I launched Plex via telnet, I wasn't having the increasing memory issue.  When I put all the goodies back in my go script, the memory leak started back up again.  I've been meaning to try launching from telnet again, but haven't done it as a true test.  I know that sounds strange, but I did let it run for a while--and was transcoding--for about 2 hours with no issue, top running the entire time in another session.

i used to have this problem, but without really changing anything (other than updating to the latest unraid beta) it stopped.

 

well, there was one more thing i changed (but i don't think this would cause such a memory leak.) before i had the library saved in /mnt/usenet/.apps/plex and now its in /mnt/usenet/apps/plex (removed the dot from the apps dir).

(don't know, if the problem stopped before this change or after.)

 

btw. i have a intel celeron

Thats a possibility, ( with the dot i mean)

 

Maybe thats why my new plugin version is working for me.

When I set the link to the Library, plex hasn't to deal with the dot.

Thats a possibility, ( with the dot i mean)

 

Maybe thats why my new plugin version is working for me.

When I set the link to the Library, plex hasn't to deal with the dot.

 

Your new version of the plugin its also working for me without the memory leak. I have an AMD.

My library is located at /mnt/cache/.Plex/Library, and spidi's plugin seems to have stopped the memory leak!