NEW: Channels DVR backend for Nvidia SHIELD

[shepherdr]

I have a Pro and used ES File Explorer to create a ChannelsDVR folder in the / folder. I then set channels to record in /storage/emulated/0/ChannelsDVR and it works fine. I can see the recordings and database etc in ES File Explorer so they are not contained in the storage area for the app. Sorry if this is already known.

[hippykiller2]

Wander what would happen if you developed it and put the Channels app on the Sheild too? Go on, give it a go, you did it for Roku. Could be a nice all in one if the shield can handle live tv and simultaneous recordings, an all in one solution

[tmm1]

Looks like the directory picker was simply broken. It has been fixed in v2017.06.07.0238.

[tmm1]

Side question: anyone else running Plex and Channels on the same box? It seems like running with both of them on keeps restarting Channels so I haven’t been getting new guide updates or able to record much of anything. I turned the Plex Server off and Channels has now been running for 4 hours and successfully recorded shows (it wasn’t previously).

Now that comskip and dir picker are fixed, I’m going to look into this one. Any specific steps to reproduce? I’ll try starting Plex and then starting Channels afterwards…

[timstephens24]

I haven’t see any way to reproduce. It’s actually down right now after working this morning to do the comskip recordings. I am using Plex on my Shield to play music right now…

[tmm1]

Can you try to install android tools and run

adb connect x.x.x.x
adb logcat

to see if its showing any errors related to Channels DVR.

[DebbieFL]

FYI:

Today, I did total reset of Shield to factory settings. Reset did load version 5.2 on the Shield.

I was hoping to be able to disable Shied Plex App and Shield Media Server (on the Shield) and just enable run for Channels DVR; however, I wasn’t able to set viewing on local (allowing PC map drive) unless PMS was enabled. This is somewhat troubling to me, as it appears that local storage is inexplicably tied to PMS.

Will follow up with results for Channels DVR on the Shield, in the near future. Currently, only using Qnap as our only Plex server…with no problems. Haven’t setup Plex as media server on Shield, although problem with missing several recordings appears to have been resolved (noted on tests prior to reset).

If anyone has been able to utilize the Shield as a stable Channels DVR server and/or has a different set up, I’d be very interested in replicating your set up and utilization.

[timstephens24]

Can you try to install android tools and run

Hadn’t looked into this until now, but I’m getting a connection refused from the Shield IP on port 5555. Is there something I need to do on the Shield to let me run the commands?

[tmm1]

Oh yea you need to enable developer mode

[timstephens24]

Just did that, but still not letting me connect. Jumped over during a commercial break from the Apple TV haha.

[DebbieFL]

No problems incurred for first 3 recordings (comskip ok, also).

Only anomalies noted were buffer warnings, but didn’t affect recordings:

2017/06/11 21:00:37 [DVR] Waiting 59m22.720685513s until next job 1497232800-ch431 The $100,000 Pyramid
2017/06/11 21:00:59 [DVR] Waiting 59m0.668598542s until next job 1497232800-ch431 The $100,000 Pyramid
2017/06/11 21:01:12 [DVR] Waiting 58m47.400005474s until next job 1497232800-ch431 The $100,000 Pyramid
2017/06/11 22:00:00 [DVR] Starting job 1497232800-ch431 The $100,000 Pyramid on ch=[431]
2017/06/11 22:00:00 [TNR] Opened connection to 1318CAEB for ch431
2017/06/11 22:00:00 [DVR] Starting job 1497232800-ch434 FOX 4 News at Ten on ch=[434]
2017/06/11 22:00:00 [DVR] Waiting 59.515061091s until next job 1497232860-ch429 Talking Dead
2017/06/11 22:00:00 [TNR] Opened connection to 1318CAEB for ch434
2017/06/11 22:00:08 [WRN] Buffer for 1318CAEB ch431 is more than 50% full (clients=1, len=16777684)
2017/06/11 22:00:12 [WRN] Buffer for 1318CAEB ch431 is more than 75% full (clients=1, len=25165868)
2017/06/11 22:00:12 [WRN] Buffer for 1318CAEB ch434 is more than 50% full (clients=1, len=16777684)
2017/06/11 22:00:15 [WRN] Buffer for 1318CAEB ch431 is more than 95% full (clients=1, len=31877468)
2017/06/11 22:00:16 [WRN] Buffer for 1318CAEB ch431 is more than 99% full (clients=1, len=33219788)
2017/06/11 22:00:18 [DVR] Recording for job 1497232800-ch431 from 1318CAEB ch431 into “TV/The $100,000 Pyramid/The $100,000 Pyramid S02E01 2017-06-11 Leslie Jones vs LL Cool J and Tom Bergeron vs Jennifer Nettles 2017-06-11-2200.mpg” for 59m59.996165414s
2017/06/11 22:00:18 [DVR] Recording for job 1497232800-ch434 from 1318CAEB ch434 into “TV/FOX 4 News at Ten/FOX 4 News at Ten 1999-08-30 2017-06-11-2200.mpg” for 59m59.51496406s
2017/06/11 22:01:00 [DVR] Starting job 1497232860-ch429 Talking Dead on ch=[429]
2017/06/11 22:01:00 [TNR] Opened connection to 1318CAEB for ch429
2017/06/11 22:01:00 [DVR] Recording for job 1497232860-ch429 from 1318CAEB ch429 into “TV/Talking Dead/Talking Dead S07E18 2017-06-11 Fear the Walking Dead 303 2017-06-11-2201.mpg” for 58m59.99429153s
2017/06/11 23:00:00 [TNR] Closed connection to 1318CAEB for ch429
2017/06/11 23:00:00 [TNR] Closed connection to 1318CAEB for ch431
2017/06/11 23:00:00 [TNR] Closed connection to 1318CAEB for ch434

[DebbieFL]

Fyi:

Recordings are directed to USB3 external, direct connected to Shield.

[TeddyR]

Ok… finally decided to try it out on the shield…

DVR: 2017.06.09.2116

I did disable Bonjour and setup remote access manually for now. Using SE5.2 with an adopted 3tb drive.

Some items to note:

• The directory picker still does not work correctly. had to use “/storage/emulated/0/DVR” to create the folder.
• The service seems to stop and restart on its own. At this time the web ui shows that the service has been up for 34 mins, even though the device has been on and sleep disabled for several hours.
• Known item with the Shield (happens with kodi as well); the number of cores fluctuates between 2-4 cores [usually shows 3. but has shown only 2 or all 4] currently showing 4.

[tmm1]

The directory picker still does not work correctly. had to use “/storage/emulated/0/DVR” to create the folder.

Currently only shows external storage options. The “emulated/0” trick is a good one for adopted storage, so I’ll have to add support for that as well.

The service seems to stop and restart on its own. At this time the web ui shows that the service has been up for 34 mins, even though the device has been on and sleep disabled for several hours.

I saw this in my log this morning as well:

2017/06/14 10:06:27 [IDX] Pruned 4598 expired airings from USA-CA04487-X in 16.657126036s.
2017/06/14 10:08:06 [SYS] Starting Channels DVR v2017.06.09.2116 (android-arm64) in /data/data/com.getchannels.dvr/files/channels-dvr/data

Looking at adb logcat from the same time, I see this:

06-14 10:08:04.251   687  1480 I ActivityManager: Process com.getchannels.dvr:channels_dvr (pid 3419) has died
06-14 10:08:04.251   687  1480 D ActivityManager: cleanUpApplicationRecord -- 3419
06-14 10:08:04.251   687   702 I libprocessgroup: Killing pid 3475 in uid 10094 as part of process group 3419
06-14 10:08:04.251   687  1480 W ActivityManager: Scheduling restart of crashed service com.getchannels.dvr/.DVRService in 1000ms
06-14 10:08:04.292   687   702 I libprocessgroup: Killing pid 3475 in uid 10094 as part of process group 3419
06-14 10:08:04.333   687   702 I libprocessgroup: Killing pid 3475 in uid 10094 as part of process group 3419
06-14 10:08:05.264   687   700 I ActivityManager: Start proc 6177:com.getchannels.dvr:channels_dvr/u0a94 for service com.getchannels.dvr/.DVRService

So it definitely seems it was killed and restarted for some reason. However, I see that DVR wasn’t the only one this happened to… almost all the apps/services running died and were restarted. For example, same thing happened to HDHR DVR:

06-14 10:08:16.241   687  1479 I ActivityManager: Process com.silicondust.view:hdhr_svc_proc (pid 3646) has died
06-14 10:08:16.241   687  1479 D ActivityManager: cleanUpApplicationRecord -- 3646
06-14 10:08:16.241   687  1479 W ActivityManager: Scheduling restart of crashed service com.silicondust.view/com.silicondust.hdhomerun_record.RecordService in 29010ms
06-14 10:08:45.268   687   700 I ActivityManager: Start proc 7586:com.silicondust.view:hdhr_svc_proc/u0a85 for service com.silicondust.view/com.silicondust.hdhomerun_record.RecordService

Apparently this is just normal Android behavior, according to https://stackoverflow.com/questions/14747780/android-process-produces-logcat-has-died-message-very-often/14748123#14748123

[timstephens24]

[quote=“tmm1, post:154, topic:1643”]
Apparently this is just normal Android behavior[/quote]
My issue stemming from this is that it is keeping the guide data from getting downloaded.

[tmm1]

My issue stemming from this is that it is keeping the guide data from getting downloaded.

Is it getting killed during guide download?

[timstephens24]

Yeah, that was my main issue. It’d download a little bit then get killed. I could get a few days by going in and scanning but then it’d get killed and restarted while it was scanning.

Also had this issue I didn’t catch till today since work had me in all night: https://stephens.cc/logs/channels.log

[tmm1]

Hrm, seems like your guide database might be corrupted from one of the earlier crashes. Can you delete/reinstall the app and restore from backup? (Or shutdown and manually wipe the USA* directories in /data/data/com.getchannels.dvr/files/channels-dvr/data)

[timstephens24]

How do I get to those folders?

[tmm1]

Not sure. Might not be possible without rooting the system. They should get deleted if you delete the DVR app.