DVR Causing Massive Transcode Cache and does not clear

Has anyone else had issues with their transcode cache directory getting consuming the entire disk space until there is none left?

I am running Windows 10, PMS 1.8 (the latest, but it started while running 1.6 so I updated in hopes of fixing), and using a HDHomeRunPrime.

I just started using the DVR a few weeks ago and seems to be working well as it records the shows and put them into the correct directory/folder as well as shows up in the Plex listing as I want them to.

However I got a notice that my hard disk space was low. After digging around my PMS transcode cache (“AppData\Local\Plex Media Server\Cache\Transcode\Sessions”) was 175GB. When I went into see what the files were it was a ton of small .ts files with about 3 seconds of the TV shows that I have set to record. Not sure how Plex is recording behind the scenes but it look like it makes these small files then joins them into a whole show.

I went ahead and stopped Plex and deleted the cache folders so I could resume using the computer since the cache sits on the Windows OS drive which is just an SSD. Then I was able to verify that the only thing that sticks in that cache directory are the DVR shows.

One other interesting thing is if I manually shut down the server it will clear the cache…however when it gets large (like 50+GB) it takes a long time for the process to actually close. However the issue is if I don’t manually close the Plex server and it just closes because of a Windows Update reboot in the middle of the night those cache files don’t seem to get deleted unless I go in and manually shutdown the server.

I really like the DVR and having it integrated right into Plex that I have been using for years, but I need to get this issue with runaway caching fixed. I have search the forums and mostly the cache issues are folks that are dealing with sync or remote directories and not having to deal with DVR.

I am not sure what other logs/info to provide…but I can provide anything that might be helpful if anyone has an idea of how to fix this.

I am having this problem as well. I use the HDHomeRun Extend with a CentOS (linux) install of Plex. It’s been driving me nuts. Like you I haven’t found anything in the forums that helps with this problem. The only fix thus far has been for me to restart the PMS whenever I go to play back a recording and the error message indicates I don’t have enough drive space for the transcoding. Why it even needs to transcode I can’t figure out. It’s getting frustrating because shows are not recording because of the drive space being consumed by the cache.

I had this same issue last week. I pointed the transcoder temporary directory to a different drive that has more storage, but that isn’t a real fix. I use the HDHomerun Extend with Win10 PMS. The cache clears on PMS restart.

I don’t know why transcoding is happening in the first place, since the Extend can transcode on the fly. I just went to the Extend webpage/transcode configuration/DLNA Formats and unticked the “Advertise native format” to see if that will help.

It doesn’t help. I already tried that.

Hmmm, okay, how about scheduling a nightly restart of the PC? There seem to be a lot of reasons for me to do this anyway.

That may work since restarting the PMS service seems to clear the transcodes. Not sure how you do it in Windows, but in Linux I can schedule a service restart with some minor scripting. Seems better to restart a service than the entire PC. Still this is far from a fix.

As a paying customer I’m a little disappointed in the “support”. This thread has been up since August and it seems nobody has a solution. No word from anyone at Plex. Am I missing something? Is there another way to get support?

I agree that a nightly restart is not really a fix. For reference, to anyone else that wants to know how to do this in Windows, if you run taskschd.msc, you can set up a scheduled restart by using the “Create Basic Task” wizard and having it run “shutdown” with the “/r” argument.

Thread response to issues seams unpredictable. I think Plex has made great strides in their product in the past year or two, but lacks product focus. IMHO, they seem to be trying to do too many things with not enough people. You may get better response if you tag one of their support people on your thread, though I don’t know if that is generally encouraged or not.

I’m also having that problem, on Nvidia Shield. I’m using a temporary share to store those files, but would be nice to have it fixed.

It will eventually fill your temporary drives as well. It seems to eventually eat all space you make available to it.

I have been running some experiments with streaming live tv on my iphone and watching the transcoder cache space and the behavior seems improved over what I saw last week. I don’t know if that is due to a fix in PMS on Win10 (two updates have come out since last week), or if changing the PMS transcoder settings affected it. What I see now is PMS appears to be cleaning up after itself better when I stop watching live tv.

Did anybody find a solution to this? It started happening to me recently. Windows 10 PC Server.

Same here… we just cut the cord because I felt that live tv/dvr in plex was stable enough. The very next day I had this problem with the transcode cache directory filling up the drive. I’m going to try setting up a nightly job that will delete that folder, but not sure how plex will react if those files just disappear.

This problem seems to have stopped for me. However I now have the 100% record bug and nothing seems to actually finish recording.

I stand corrected. I’m still having this issue it was just masked by the fact that all of my recordings wer failing due to the “100% record bug.” I guess nobody the the few of us are seeing this issue?