Failed to be awake for recording - why?

[McBainUK]

My NextPVR PC is set to sleep after 15 minutes of inactivity. It has always woken for each recording successfully before, but this time it failed to wake for 2 recordings starting at 8pm. From the NPVR.log I can see that the PC went to sleep at 19:53:59.306. I thought, being so close to recording time, that it should have been stopped from doing this by NextPVR. Did something go wrong this time?

[sub]

I don't know. Maybe a case of bad timing? I can see the app had asked Windows to be awake at 19:53, which is was just before the recording was about to kick in. It looks like you machine was already awake, but went to sleep soon after.

2013-02-27 19:35:02.735 [DEBUG][7] Requesting wake up for 'RecordingOID1378' at 10/03/2013 19:53

[jcjefferies]

This is a problem that comes up occasionally but has not happened to me recently. What happens is the computer starts the process of going into sleep mode which takes perhaps 5-20 seconds. Any "Wake" which happens during those few seconds is lost! My old desktop PC took over 30 seconds to sleep and I would often go to use it and wonder why it is not responding until I noticed that it was going into sleep so would have to wait until it finished before I could wake it again. Someone suggested that to cover this situation NextPVR should generate a second "Wake" around 30 seconds after the first.

[Lao Pan]

Had this happen to me 2 weeks ago - missed recording a program for the first time in well over a year - unfortunately it was one of "HER" programs and of cause "It does it all the time" :eek:
On checking it was just bad timing we had gone to bed about 25 mins (sleep setting) before the program started not the 15 mins I had thought - and left puter running.
Maybe a wakeup call before the padding and one on the start time - just in case - would be an enhancement.

Another timing problem that has carried over from Vista to W8 is that when the puter wakes up in the night to do an epg update it doesn't always stay awake long enough to complete it. I think it is the unattended wake timer that is not disabled for the epg update - it must be for recording programs.

[McBainUK]

Had this happen to me 2 weeks ago - missed recording a program for the first time in well over a year - unfortunately it was one of "HER" programs and of cause "It does it all the time" :eek:

I feel your pain. I wrote the OP while she was watching over my shoulder. :eek:

[johnsonx42]

What seems to cause this most often is that nothing prevents the computer from going to sleep between the scheduled wakeup and the beginning of recording. That is, if the computer is already awake, but the sleep timer happens to expire in the two minutes right before the recording starts, it will go to sleep and not start the recording. Also as jcjeffries observes, it can't wake-up while it's already going to sleep.

One would think that Windows would be smart enough not to do this: if the sleep timer expires and Windows is preparing to sleep, shouldn't it make sure there's no wakeup request in the next few minutes? Also, shouldn't every wakeup request reset the sleep timer? If the PC is set to sleep after 15 minutes, and the timer has reached 14 minutes when there's a wakeup request, the sleep timer should be reset to zero. It's not like sleep mode is some fancy new thing, all of these possible failures are well known by now yet the Windows devs seem to have done nothing about this.

I've suggested before that NRecord should issue a second wake request for all recordings at recording time, so that if the PC does sleep during the two minutes prior to recording start it will wake up again; the recording will a few seconds short in this case. So it would look like this:

Code:

2013-02-27 19:35:02.735 [DEBUG][7] Requesting wake up for 'RecordingOID1378' at 10/03/2013 19:53 & 19:55

A more perfect solution would be for NRecord to stop the PC from sleeping starting 3 minutes prior to the recording, so that if the PC is already awake leading up to recording time then it will be prevented from sleeping. This would probably be harder to implement though, whereas adding the second wakeup request should be trivial and would cover 90% of the potential problems.

(and yes, Windows and NPVR know which shows are HERS and will conspire to make sure that any odd failures will happen during those shows. it will record YOUR shows perfectly all week, only adding to her ire.)

[ACTCMS]

(and yes, Windows and NPVR know which shows are HERS and will conspire to make sure that any odd failures will happen during those shows. it will record YOUR shows perfectly all week, only adding to her ire.)

My experience exactly - how does NPVR know that? - I've searched through npvr.db3 and config.xml and I just can't see anything even remotely related to who the recording is for...

[imilne]

I got bit by this one a couple of weeks ago too, but NPVR must have been extra faulty as it was one of my shows, not the missus'. Luckily it was on iPlayer the next day anyway. It was the same thing though...incredible bad timing meaning it went to sleep in that tiny window between when the wake-up would have been (had it been asleep) and the start of the recording.

Iain

[johnsonx42]

haha, are you sure YOUR show isn't just a bit girly?

[gEd]

Had this happen to me 2 weeks ago - missed recording a program for the first time in well over a year - unfortunately it was one of "HER" programs and of cause "It does it all the time"

+10

or HER show runs late by 60 seconds so she misses the last 2 sentences of the show, thereby rendering the entire recording pointless...
I think I would actually pay for iPlayer, given the number of times it has dug me out of trouble.