I've recently added (4 days ago) a new dedicated server that is only running NextPVR v.3.9.2 for now. It's a recent i3 with 4GB of ram running Windows 7 Ultimate. I have one computer client hooked up, plus 5 popcorn hour devices. The issue I'm having is that on a daily basis, I'll go turn on one of the TVs and the popcorn hour, when trying to link to the mvpmc folder, gives me an Out of Memory Error. This happens with all popcorn hour A-100s when it shows up on the first. I'll check the NPVR server, and it is generally at 45% memory usage. The only way I've been able to fix this is by rebooting the server (I'm not sure if just stopping and restarting NPVR would fix it also), but this is a pain with the number of recordings going on, and it's just not wife and child friendly. I don't know if it has to do with comskip running or something along those lines. Any thoughts?

I don't know how to link a file with this new version of vBulletin, so here is a link to my google drive with all of the log files: https://drive.google.com/open?id=0Bxcbws...WRoaWdUcjA

Thanks.

Quote: I'll go turn on one of the TVs and the popcorn hour, when trying to link to the mvpmc folder, gives me an Out of Memory Error

Where are you getting the 'Out of Memory' error message? On your Windows machine, or on the Popcorn Hour?

sub Wrote:Where are you getting the 'Out of Memory' error message? On your Windows machine, or on the Popcorn Hour?

On the Popcorn Hour. Rebooting the Popcorn Hour does not fix the problem. Only when I reboot the computer does the issue clear up.

If I recall this error on a PCH sometimes has nothing to with memory and more to do with losing connection to the server or the share and not being able to start the script. You could try running the NMT Syntax check entering the IP of the failing client to see if that shows anything.

Unfortunately debugging the PCH is almost impossible with multiple clients running. If you can set it to one and remember to exit to the main screen when your aren't using it that might help.

From your logs I see two potential problem

First in MVP.log the server doesn't seem to be runnon on an LAN port

2017-06-04 18:55:23.316 [INFO][6] BOOTPWorkerThread IP Address 127.0.0.1

so I'd check with netstat or a port viewer program to makes sure that your LAN ports are all open

Second, in the NRecord.log.1.txt NRecord seems to loop cycling restarting MVP proceses 0 through 9 even though you should only have six sub would need to comment on that logic

Martin

Unfortunately, it's not just one Popcorn Hour, but every single one that will show the error. I did set the client mode to 9 just because of the computer clients, but I think I goofed that up and it should have only been the PCH clients. I "only" have 5 PCH devices running currently, and can drop that to 3 with the most active devices. I'll remove 2 of them from the network, drop the client number to 3, and see if that helps.

Very strange with the server - I'll check on that also. I have it running through a pair of switches, so maybe one isn't operating correctly.

I'll post more tonight.

Thanks,
Joe

joeags Wrote:Unfortunately, it's not just one Popcorn Hour, but every single one that will show the error. I did set the client mode to 9 just because of the computer clients, but I think I goofed that up and it should have only been the PCH clients. I "only" have 5 PCH devices running currently, and can drop that to 3 with the most active devices. I'll remove 2 of them from the network, drop the client number to 3, and see if that helps.

Very strange with the server - I'll check on that also. I have it running through a pair of switches, so maybe one isn't operating correctly.

I'll post more tonight.

Thanks,
Joe

I think what Martin means is that for testing purposes, change the setting to one and turn off all PCH's but one. Then, restart the service and the PCH. That way everything in the logs about a PCH will be for that single one & it should make it easier to see what isn't working. The theory is that you could resolve the issue and then add the others back into use. Also, as he said, the NMT Syntax Check can help find the issue.

Or, maybe I am totally misunderstanding Martin.

That's it. It doesn't have to be one PCH, just one active others can be at the PCH menu. Run the syntax check with the ip of the PCH with the memory error.

Martin

Thanks guys, that makes more sense. I've rolled back to one PCH and setting one for the number of clients in the settings. I'll give this some time to see if the error creeps back in as my wife will be in full user testing mode tomorrow, and will report back.

Ok, I had 2 days free of issues with just one PCH client in operation, so I have added a second PCH client, and after an evening and morning of operation, have had no issues there either. The early results are that this issue is related to the number of clients that I had set as active in the NextPVR settings menu.