Problem with recordings

[peanut]

I have 2 DVB cards one setup for TVNZ and the other setup for other channels and got 2 analog cards setup with all analog channels on both
I had 2 dvb recordings start at 8.30pm to 9.30pm and one at 9.30pm to 11.30pm
Started witching one 8.30 recording at 9.30 with nmt (was still recording) at 10.00pm npvr (in pending) said 3 recordings were still recording (only one should have been) couldn’t stop 2 recordings so went to computer and tried there couldn’t stop there
Checked the files the 2 at 8.30 recordings wasn’t increasing and the 9.30 recording was 0kbs
So copped logs (logs 1) then stopped recording service. In pending the 2 8.30 recordings said they failed (but they were 90% there). The 9.30 recording said it was still recording so copped logs (logs2)
So restarted computer

[Jaggy]

sub have you got any thoughts on these recording services stopping working problems?

I have just read too many reports of it happening to too many different users & it is the biggest/main reason for me not to switch my production server over to NextPVR at this stage, also I know peanut's server PC is just about identical to mine as I sold him most of the hardware & originally helped him setup for GB-PVR.

[sub]

I can see it tried to stop a device, but the device failed to stop and it all went down hill from there. I'm not sure why this happens to some users. I've never been able to reproduce it here. We've seen this in logs occasionally over the last few years, with both GBPVR and NPVR.

[peanut]

the srevic pc been running sice the last reinstall for the last 2ish years with gbpvr and havnt had this problem.
I offen witch programs wile there are still recording and han't had this problem befour

[sub]

Sorry, I'm not sure what to suggest. It definitely looks like the same problem I'm refering to.

011-01-27 21:35:00.218 [INFO][7] No more streams active. Stopping device.
2011-01-27 21:35:00.218 [DEBUG][7] About to request async graph stop
2011-01-27 21:35:00.218 [DEBUG][14] Graph stopping... (async)
2011-01-27 21:35:08.218 [DEBUG][7] Async stop taking way too long

[Jaggy]

sub I have just got off the phone to peanut & something he said made me start thinking that "may" shed some light on the problem....

The (or a) problem actually occurred long before it tried to stop the device as the two recordings from 8:30 to 9:30 (+5minutes pre & post padding) both were cut short by approx 10 minutes & the 9:30 to 11:30 (also with 5 minutes padding) didn't record at all, so I would say the problem actually started at 9:25 when it tried to start that third recording (that I think should have been on a different device).

2011-01-27 21:21:14.171 [DEBUG][7] - saved 0. skipped 12
2011-01-27 21:25:00.312 [DEBUG][7] About to start recording (1007): F:\GBPVR-R\Recordings\Army Of Darkness\Army Of Darkness_20110127_21302320.ts
2011-01-27 21:25:00.312 [INFO][7] DigitalRecorder.StartStream(F:\GBPVR-R\Recordings\Army Of Darkness\Army Of Darkness_20110127_21302320.ts)
2011-01-27 21:25:00.328 [INFO][7] DigitalRecorder.StartStream() allocated handle: 0x4
2011-01-27 21:25:00.328 [DEBUG][7] Temp at 27/01/2011 9:25:10 p.m.
2011-01-27 21:25:00.328 [DEBUG][7] Started recording (1007:F:\GBPVR-R\Recordings\Army Of Darkness\Army Of Darkness_20110127_21302320.ts)
2011-01-27 21:25:00.421 [DEBUG][7] No C:\Documents and Settings\All Users\Application Data\NPVR\Scripts\ParallelProcessing.bat
2011-01-27 21:35:00.187 [DEBUG][7] Stopping recording (1164). Past end time of recording.
2011-01-27 21:35:00.203 [INFO][7] DigitalRecorder.StopStream() handle: 2
2011-01-27 21:35:00.218 [INFO][7] No more streams active. Stopping device.
2011-01-27 21:35:00.218 [DEBUG][7] About to request async graph stop
2011-01-27 21:35:00.218 [DEBUG][14] Graph stopping... (async)
2011-01-27 21:35:08.218 [DEBUG][7] Async stop taking way too long
2011-01-27 21:35:08.281 [DEBUG][7] Removing filter MPEG-2 Sections and Tables
2011-01-27 22:07:21.421 [DEBUG][16] CancelRecording(1164)
2011-01-27 22:07:33.500 [DEBUG][16] CancelRecording(1164)
2011-01-27 22:07:43.375 [DEBUG][16] CancelRecording(1167)

[sub]

sub I have just got off the phone to peanut & something he said made me start thinking that "may" shed some light on the problem....

The (or a) problem actually occurred long before it tried to stop the device as the two recordings from 8:30 to 9:30 (+5minutes pre & post padding) both were cut short by approx 10 minutes & the 9:30 to 11:30 (also with 5 minutes padding) didn't record at all, so I would say the problem actually started at 9:25 when it tried to start that third recording (that I think should have been on a different device).

If you're able to find a reproducable set of steps, I can advise some steps to collect extra logs to investigate.

[Jaggy]

Damn the timing stinks as I would have liked to see if we could get to the bottom of this, peanut says it has happened a number of time so the chances are it is reproducible but he is going on holiday today so it will have to wait unless someone else can reproduce the same problem & get the logs.

[Jaggy]

peanut asked me to post this message as he is very busy at the moment, we have been testing his failed recording problem to try to eliminate some things & get a re-creatable scenario. It is fairly easy to recreate as it only fails when it uses both his DVB-S cards at once. Yesterday I had him do a number of single recordings (that all worked fine) & then a dual card recording that failed with a crashdump file created in temp.....attached along with the logs.

peanuts message to me...

Hi got test results
tv3 started at 11am to 12
tv3 '' at 12.30 to 1pm
tv2 '' at 1.30 to 2.30
tv1 '' at 3 to 4
all work good GB was doing a manual record at the time 11am to 5pm all work will

tv2 and tv3 together 5.30 to 6pm
failed got a message that there was a log in c:\temp
also got this message as well

nRecord exe (at top )
the instruction at 0x5e0f95b0 referenced memory at 0x5e0f183c.
the memory could not be written
click ok to terminate the program

[sub]

This crash dump shows it crashing in the Microsoft psisdecd.dll. We see this occasionally for XP users.

If you search you'll find other discussion about this DLL, and suggestions for trying other versions of it. Very important to note that Microsoft protects this DLL, so if you think you've replaced the file with a different version, you'll probably find it's reverted to the original version. From memory, you have to boot in 'safe mode' to replace the file.