Remote Scheduling Potential Conflicts

[phit03]

I know that this has been raised before but I thought that I'd revisit the subject.

RS reports a potential conflict when 2 programs on the same channel are scheduled one after the other and a third program is scheduled that covers the period at the end and start of the programs on the same channel.

I understand that the message says potential conflict but in reality to the best of my knowledge the recordings never fail.

Also, when setting up a reminder of zero minutes via the "Manual Event" button to wake up the box. This also causes a potential conflict message if 2 other programs are scheduled at this time. As far as I can remember, this did not happen with the old method of scheduling wake-up/sleep schedule entries. I've tested this and having a zero minute reminder set for the same time as 2 other recordings does not cause these recordings to fail.

Is there any chance that a setting "ignore potential conflicts when programs on same channel or zero minutes" could be introduced and the coding changed to allow these situations?

 

[Black Hole]

You shouldn't be able to schedule a reminder of zero period. I guess it's a bug in the RS. Even if you could, it would be pointless - the point of a reminder is that the box wakes up for it, then goes to sleep again afterwards (if it was woken and not already awake).

Like it or not, a reminder ties up a tuner, so there will still be conflicts.

 

[phit03]

The point of the zero minute reminder in this case is to wake the box up so it can be accessed remotely via the web interface. I'm still trying to test it to see if the box stays on after it's woken up from low-power standby (no network connectivity). It's taking a while as there seem to be all sorts of things that are stopping the box from going into this state. I'm trying to emulate the sleep/wake-up facility that used to be part of RS to wake-up the box at predetermined intervals so I can access it remotely via the internet when away from home.

 

[Black Hole]

You can't have a zero minute reminder (regardless of whether the RS allows it)! I guess it would be interpreted by the Humax as 24 hours.

If the Humax is in standby at the scheduled time for the reminder, it will wake up with the relevant service tuned. If it is already running, it will prompt the user and then change to the relevant service. At the end of the reminder period it will either go back to standby (if it had to wake up) or do nothing (if not).

The wake and sleep time options were a non-standard trick exploiting the method the Humax uses to set a timer on and timer off through the standard menus. These create individual schedule entries, which can be duplicated by direct access to the schedule database. Unfortunately, the Humax code which interprets the schedule database was not written to expect multiple timer entries, and it can get its nickers in a twist and result in misoperation. You can do this without the help of custom firmware - set an on time but not an off time (or vice versa) through the SUI and you can get the same effect: a Humax which refuses to go into standby properly.

Consequently the wake and sleep options in the RS have been withdrawn, and the best alternative is reminders - a safer mechanism. If you intend to use them for fetching instructions from the RS and subsequently integrating them into the schedule, you must time them for when the unit is likely to be in standby - a reminder that does not induce a boot cycle is of no use for transferring RS events into the schedule.

Don't forget that the disable-ota package (or the manual equivalent) already sets up a daily reminder for 0420-0440, so as long as you are not triggering events through the RS with less than 2 days notice you are (or should be!) already covered. For more info see Preventing External Events from Disturbing the CF (click).

 

[phit03]

Thanks for the feedback, my tests bear out what you've said. If I set a zero reminder and the box is in low-power standby, the box wakes up but almost immediately goes back into a low-power standby state. I guess I'll have to set "Power Saving in Standby" to off while I'm away to stop the boxes going into this state. I've still got old wake-up/sleep entries in my schedules that seem to work fine so I'll leave those as well.

This still leaves my original question about conflict message/emails unanswered though.

 

[Black Hole]

If I set a zero reminder and the box is in low-power standby, the box wakes up but almost immediately goes back into a low-power standby state.

Thanks for the data.

I guess I'll have to set "Power Saving in Standby" to off while I'm away to stop the boxes going into this state.

I can't imagine what you think this will achieve. Power Saving in Standby only controls whether the aerial pass-through amplifier is active and the clock is displayed on the front panel - nothing else at all. Please see Things Every... (click) sections 7 and 18.

Even if PSS did what you think, it would be of no use to you - in order to transfer events from RS (or the WebIF) to the schedule database the Humax must go through reboots - which means entering the complete standby disks-stopped state periodically.

I've still got old wake-up/sleep entries in my schedules that seem to work fine so I'll leave those as well.

That's a head-in-the-sand approach. You have been warned, on your own head be it.

 

[phit03]

"I can't imagine what you think this will achieve. Power Saving in Standby only controls whether the aerial pass-through amplifier is active and the clock is displayed on the front panel - nothing else at all."

Well, there is one other thing that it affects and that's the ethernet interface. This is the bit that's of importance to me as I want to be able to access the box(es) when I want to when away from home.

As well us using RS and sometimes web-if to schedule new programs while away, I use the web-if download function to download them onto a PC in Spain, usually overnight, to watch them there. It's a real pain if I can't access the boxes to do this. If you can think of a better solution for this please let me know.

I've also had the wake-up/sleep entries in the schedules of my boxes for some considerable time now (certainly since they first became available via RS) and can't remember any problems that this has caused (that's not to say that there haven't been any). Generally, access to the boxes while I'm away from home works well with few problems and I'm happy to take the risks associated with this approach.

 

[Black Hole]

Well, there is one other thing that it affects and that's the ethernet interface. This is the bit that's of importance to me as I want to be able to access the box(es) when I want to when away from home.

If you think the Ethernet, or any processing come to that, will be live when in standby with Power Saving in Standby = Off, you are very much mistaken. My post above is explicit, as are the references I have pointed you to,

 

[phit03]

So when I use the remote control to turn off a box and the display goes blank and I use the web-if to access that box via the ethernet interface and the bar at the top says "The System is in Standby", what sort of standby is this? Is this a deferred standby because the custom firmware is doing some background processing and it will go into real standby when that processing is complete?

 

[Black Hole]

Not the CF, no - but the Humax standard firmware takes time to tidy up and shut down, during which period the WebIF is still active. The message is slightly misleading, maybe it should say "preparing for standby".

If you put the unit into standby while you are observing, you will see that although the front panel goes dead straight away, there is a considerable time before the hard drive clicks off - at which point the Ethernet port activity lights go out and power drops at the USB ports. This time can be anything from a few seconds to 30 seconds or more.

If you power it up again before the shutdown is complete, or if it never powers down properly at all (see Things Every... section 18, "Delinquent Half-Awake"), you will see that it comes up without the usual start-up delay and Humax splash screen on the TV - indicating there has not been a boot and therefore processes that rely on the boot cannot happen. Boots are vital to the operation of the RS, and the reason you cannot risk using wake and sleep timers - if they go wrong the unit will go into Delinquent Half-Awake and any RS events will fail to be transferred to the schedule.

If your "The system is in standby" state (with the WebIF still active) is a permanent situation rather than temporary, you are suffering the consequences of using wake and sleep timers that I was warning about. Under those circumstances, scheduling by RS or WebIF will not work unless you force reboots by power cycling or system reset. (Note that there are other causes of the Delinquent Half-Awake state, not just mismatched or excess on-off timers.)

 

[Trev]

A useful side-effect of the delinquent state is that network access remains available (but not DLNA), when otherwise the HD/HDR-FOX should be in standby.

But this is not what phit03 is describing in Post #9. Or is it?

 

[Black Hole]

I don't know whether it is or not - the OP has yet to report whether the 'WebIF up in standby' state is transient or permanent. The point is that he came into this thinking that the 'up in standby' state is a result of having Power Saving in Standby = Off (which it isn't), and that it would allow the 'FOX to continue communication with RS, (which it will) ...but being up in standby also means there are no reboots to trigger transfer of RS events into the schedule, so although the availability of network services can be described as a useful side-effect in some circumstances, it is definitely not useful from the point of view of remote scheduling via WebIF or RS.

 

[phit03]

Yes, you're right Black Hole, I had read the Wiki in the past and knew about the ethernet interface not being available in standby, but I'd forgotten. The box may have been in Delinquent Half-Awake mode but I still have several old style sleep entries in the schedule that are probably forcing a shutdown, as well as using "Schedule Reboot" via the RS when I add new entries and hence I haven't been having schedule update problems. I don't have time in the next couple of days to test this further but I've ordered some IR over IP boxes to see if I can remotely control the boxes via the internet using the remote control. If I can, I'll remove all the extraneous stuff from the schedule and use this method (IR over IP) with a strategically placed webcam to see what's happening, to wake the boxes from standby remotely to activate the ethernet interface. This should allow me to add new programs quickly either via RS or the Web-if and download video over the internet using the Web-if.

 

[Trev]

Novel approach, but at least you will be able to see it happening 'in real time'.

 

[phit03]

In all the foregoing in this thread, my original first part of the question about the issuance of recording conflict messages from RS has been overlooked.

RS reports a potential conflict when 2 programs are scheduled where one finishes at the same time another starts and a third program is scheduled that covers the period at the end and start of those programs. For example a program is set to record from 20:00 to 21:00 on BBC1, another set to record from 21:00 to 22:00 on BBC2 and a third to record from 20:30 to 21:30 on ITV. For the purposes of this illustration let's assume that all 3 programs are on different multiplexes. I suppose you could say that there is an overlap of the programs on BBC1 and BBC2 for 1 minute at 21:00 and that because the third program is also recording at this time there is a need to record from all 3 multiplexes. But in these circumstances I've never had a recording fail although if a program is late or early you may lose the start or end of it. However, most of the time the programs do not start or end at the scheduled time and therefore nothing is lost.

I have to add that the conflict message also occurs if the back to back programs are on the same channel and in this case there is no real overlap.

The TV Diary under the Web-if handles this as I would expect and doesn't flag a conflict when this occurs. I would expect the RS to act the same way.

Is there any chance that RS could be changed to match the way that TV Diary operates or an option made available to suppress the message if the conflict is caused by 2 back to back recordings?

 

[phit03]

Just an update on post 13 where I mentioned trying out IR over IP boxes I can report that I purchased a pair of boxes from Keene Electronics (KEENE IR ANYWHERE IR OVER IP MODULES (PAIR)) and can report that they work with the HDRFOX-T2 boxes. I am able to remote control the boxes over an internet connection using the original remote controls.
Setup was straightforward apart from my BT Homehub 5 not seeming to like the default port number when I set-up the port forwarding.

 

[Black Hole]

How does this work - do you have a static IP address, or some kind of dynamic IP service, or is it mediated by a dedicated web service somewhere?

The problem is that without further help you don't get any feedback on what effect the IR commands have had. I know you mentioned using a web cam, and there is limited information via the WebIF status if you were to create an external link through to it.

With an external link to the WebIF you could do most things through the virtual remote control - except turn the box on!

I am curious that you think it is worth this effort and expense to be able to schedule a recording for immediate action - instead of just planning ahead. And it's "only telly" - and catch-up services are available.

Update: http://www.keene.co.uk/keene-ir-anywhere-ir-over-ip-modules-pair.html

OK, so they use Dynamic DNS.

 

[af123]

They look nice. I didn't find anything that cheap when I was looking last year and I ended up rigging up an IR blaster to a Raspberry Pi I have on the shelf on the other side of the room to the Humax. Works well and since I already had the Pi it only cost pennies.

 

[phit03]

As Black Hole mentioned you can do just about everything you need to via the Web-if and RS. The problem I've had in the past when accessing the boxes from abroad is that you've fewer options if a box fails to respond. In the past I've resorted to powering the boxes down and back up remotely to resurrect them but I'm not too comfortable with this approach. Using a webcam and remote control I can see what is happening to the box in real time. Why do it when I could view them via catch-up services? I've found proxy servers a bit flakey when trying to access UK catch-up services from Abroad. If I had the time and skills I might have a go at a DIY solution as AF123 has but I'm impatient and it's easier to buy an off the shelf solution and as AF123 mentioned, the price of this sort of thing has come down over the past year or so.

Yes, the remote box uses Dynamic DNS to access my UK router. The DDNS setup is a bit confusing though and it took me a while to sort it out. They uses UDP to communicate but if the packets get dropped you just try again. So far I've only shutdown and restarted the boxes via remote control as this is really all I need to do. I've tested this out using NAT reflection on my LAN and also took it to my brother-in-law's house and used it from there without having to change the remote configuration.

 

[prpr]

Yes, the remote box uses Dynamic DNS to access my UK router. The DDNS setup is a bit confusing though and it took me a while to sort it out. They uses UDP to communicate but if the packets get dropped you just try again.

So do you just set up a UDP port-forward on your router then?