[vdr-plugin-satip] this and that

A new version 0.3.3 is now available containing some minor fixes.

rofafor

Thanks, we will try

Regards

Hi,

I have a problem with satip plugin 0.3.3 and OctopusNET. On VDR startup the plugin detects my OctopusNET, but looses the connection after a few minutes:

Jun 10 10:43:33 test-server vdr: [5024] SATIP: Adding device OctopusNet (192.168.1.14 DVBS2-2)
...
Jun 10 10:45:35 test-server vdr: [5024] SATIP: Removing device OctopusNet (192.168.1.14 DVBS2-2)

But satip info is still available.

root@test-server:~# svdrpsend plug satip info                                                                                                                                                 
220 test-server SVDRP VideoDiskRecorder 2.1.6; Tue Jun 10 10:59:18 2014; UTF-8
900-SAT>IP device: 0
900-CardIndex: 0
900-Stream: rtsp://192.168.1.14/?freq=12544&src=1&sr=22000&pol=h&fec=56&msys=dvbs [stream=1]
900-Signal: lock=1 strength=80 quality=66
900-Stream bitrate: 4768 kbit/s
900-Buffer bitrate: 4768 kbit/s
900-Buffer usage: 904/8192 kbit (11,1%)
900-Channel: kabel eins;ProSiebenSat.1:12544:HC56M2S0:S19.2E:22000:767=2:768=deu@3:34:0:17502:1:1107:0
900-Active pids:
900-Pid 0:  767 (3488 kbit/s)
900-Pid 1:   18 ( 944 kbit/s)
900-Pid 2:  768 ( 168 kbit/s)
900-Pid 3:    0 (   8 kbit/s)
900-Pid 4:  100 (   0 kbit/s)
900-Pid 5:   17 (   0 kbit/s)
900-Pid 6:   99 (   0 kbit/s)
900-Pid 7:   97 (   0 kbit/s)
900-Pid 8:   98 (   0 kbit/s)
900-Pid 9:   20 (   0 kbit/s)
900-Active section filters:
900-Filter 0: 46809 ( 904 kbit/s) Pid=0x12 (EIT)
900-Filter 1:  496 (   0 kbit/s) Pid=0x14 (TDT)
900-Filter 2: 7634 (   0 kbit/s) Pid=0x00 (PAT)
900-Filter 3:  382 (   0 kbit/s) Pid=0x11 (SDT)
900-Filter 4:  339 (   0 kbit/s) Pid=0x10 (NIT)
900 Filter 5:    0 (   0 kbit/s) Pid=0x66 (---)
221 test-server closing connection

Another problem is that a restart of the vdr instance never detects OctopusNET again. I have to reboot the complete system.

Any ideas?

I have a similar problem with both of my servers (Octopus T and Gbox) - detection just sometimes fails to return anything.

If detected, then they'll drop out randomly and then occasionally come back. This is the case both for the plugin and for the elgato app.

The only sure-fire workaround seems to be restarting the satip server - detection will then work. I've increased the expiry time within the plugin to try and mitigate it a bit.

However, given that it affects both servers and 2 different apps it must be something on my network, so I'm planning on changing the switch used for the 2 servers and seeing if that improves things.

Ok, the problem is my TwonkyMedia UPNP Server running on the same machine. Stopping Twonky brings back the OctopusNET and restarting VDR reconnects allways to OctopusNET. Reboot is not necessairy anymore.

It seems there some incompatibilities between satip-plugin and local TwonkyServer. But a Twonky on another machine does not have any impacts.

Hi,

it seems that there is a problem with overlapping timer on the same channels, which result in broken video data stream and loss of the 2nd recording This not happen in all cases but very often. In the logfile I found following entries

Jun 22 09:04:32 test-server vdr: [29697] switching device 4 to channel 26
...
Jun 22 09:05:00 test-server vdr: [29697] switching device 2 to channel 26
...
Jun 22 09:05:30 test-server vdr: [8646] ERROR: video data stream broken
Jun 22 09:05:30 test-server vdr: [8646] initiating emergency exit

There are 2 devices which are switching to the same channel. In all cases with overlapping timers and broken video stream, I saw this behavior. A restart of the VDR resumes the recordings.

I put here the detailed log. I use OctopusNet Firmware 1.0.29 and 4 Tuner (2 Cards) und SatIP-Plugin 0.3.3. VDR 2.0.6 und 2.1.6 produce the same error.

Does anyone have the same problem? Any ideas to solve this?

I made further tests:

Disabling EIT scanner by setting the EPGScanTimeout to 0 seams to solve the problem. I had no more broken video streams for several days.
But, I made a little patch in the eitsanner.c which prevents EIT scanning while recording. This not solve the problem.

BTW: vtuner with satip daemon works w/o problems.

Guys, check your logs, maybe you have the same problem but never noticed that. Because a restart by emergency exit will fix the problem for the current recording and only recording is affected, not live TV

This looks like a bug in the cSatipDevice::SetChannelDevice(), where the device selection is trying to avoid already active ones.

Thx for response. If I can do some tests, let me know.

Rolf: you mentioned in another thread that you are working on a bugfix release. So here some hints and wishes for this release:

* Can you please add Telestar Digibit R1 to session id quirk? Adding this line in server.c works for me

strstr(*descriptionM, "DIGIBIT")

* The problem concerning broken video streams described above can easily reproduced by programming timers on each tuner on different channels with exactly the same start time
e.g. on a 4 tuner system
timer1 - channel1 - start 10:00
timer2 - channel2 - start 10:00
timer3 - channel3 - start 10:00
timer4 - channle4 - start 10:00

* Have a look at the latest Octopus firmware release 1.0.35. I had a lot of broken video stream errors, the plugin is not usable with that firmware. So I switched back to 1.0.29

Hi,

it seems, that my setup isn't also running as it should. If some time is gone after (re)start of the VDR instance a starting timer will result in a restart of the VDR instance caused by a 'ERROR: video data stream broken'. This can be also within a running recording. Directly after the VDR restart I can start up multiple timers on up to 3 transponders without problems. But if some time is gone, a new timer or a channel switch will result in trouble.
For debugging I have done a Packet capture and syslog split while a new recording is startet resulting in restarting the VDR instance. Maybe this will help, to get this error fixed. My setup is the experimental Server from my signature (unstable-yaVDR 2.1.6 with satip-plugin 0.3.3 on Triax TSS400 with SW 0.5.12).

ByE...

Hi,

i have also the "ERROR: video data stream broken" issue. But i have vtuner/satip at work. VDR also does a Emergency exit. This appears as described above when a recording ist started.

Jochen

I dont have video data stream broken with vtuner/satip. Maybe it is device specific? But i dont have emergency exit enabled.

The satip-plugin i couldnt test yet.

Hi.

The vtuner/Satip combo has worked without any problems, even with HD channels on my Triax TSS-400. But with this package I've got scrambled recordings, if some other (K)VM works hard on the network (using NFS share), for example cutting recordings or converting to H264 while a recording was running. So I had to switch to OpenVZ using VDR with satip-plugin. I didn't got the vtuner/Satip combo running on my Promox kernel 2.6.31, so I have go with the satip-Plugin.

ByE...

A new version of the plugin is in development. I think rofa will release it in the next days or weeks.

Any active GSSBOX users having some spare time to test a new version and willing to give me a hand? It looks like that I've broken something, but I just don't have any idea what it could be...

I'm only having a TSS400. I think it's another plattform. What do you want to test?

ByE...

Hi, I have a GSS box and can test something.

Quote from ullek

Hi, I have a GSS box and can test something.

Quote from atl

I'm only having a TSS400.

Great! Please, check your PM for further instructions.