Re: [vdr] Sat>IP-plugin: how to configure 2 DD Octopus NETs with different satellites

[Patrick Boettcher]

On Tue, 19 Jul 2016 02:37:42 +0200
"fnu"  wrote:

> Hi,
> 
> I remember that Rolf always does suggest to use "OctopusNet" or
> "minisatip" as device name.

Yep, that's for the quirks, I think. I simply copied the string as I
found them in the README.

> The other stuff, hmm, I don't know, as the plugin follows quite
> straight the
> SAT>IP specifications. Your add-on is maybe the point to discuss with
> SAT>the  
> original author Rolf.

In the spec I found the description of the fe and src argument. They say
that, AFAIU, the src-arg depends on the configuration of your
satip-environment and that it is the client-side which should be
configured to use it correctly. I assume it is the same for the fe-arg.

I would be delighted if Rolf joins this thread. For the moment I
continue to explore things.

Thanks,
--
Patrick.

___
vdr mailing list
vdr@linuxtv.org
https://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr

Re: [vdr] Sat>IP-plugin: how to configure 2 DD Octopus NETs with different satellites

[fnu]

Hi,

I remember that Rolf always does suggest to use "OctopusNet" or "minisatip"
as device name.

So your definition is recommended to look like this:

--server=192.168.1.21|DVBS2-4|OctopusNet1;192.168.1.20|DVBS2-2,DVBT2-2|Octop
usNet2

The other stuff, hmm, I don't know, as the plugin follows quite straight the
SAT>IP specifications. Your add-on is maybe the point to discuss with the
original author Rolf.

Regards
fnu

-Ursprüngliche Nachricht-
Von: Patrick Boettcher [mailto:patrick.boettc...@posteo.de] 
Gesendet: Montag, 18. Juli 2016 17:43
An: fnu <v...@auktion.hostingkunde.de>
Cc: 'VDR Mailing List' <vdr@linuxtv.org>
Betreff: Re: [vdr] Sat>IP-plugin: how to configure 2 DD Octopus NETs with
different satellites

Hi fnu,

Thank you for your answer.

On Mon, 18 Jul 2016 10:29:58 +0200
"fnu" <v...@auktion.hostingkunde.de> wrote:

> Dear Patrick,
> 
> changes in "sources.conf" are IIRC only necessary if you have a DiSEqC 
> configuration controlled by VDR. But using a/the generic sources.conf 
> doesn't harm VDR operation, I guess most users do have one without 
> DiSEqC.

That seems to be true. Before looking at the code of the satip-plugin I
thought that I could limit magically the created (virtual) satipDevices to
the Sources assigned to them via the diseqc.conf.

> You may use a workaround by using CAID field for specific adapter:
> 
> https://www.linuxtv.org/vdrwiki/index.php/Syntax_of_channels.conf#CA_-
> _Conditional_access

Yeah, no, I don't like this.
 
> But I do not now if it is possible to cover more than 4 devices and no 
> good idea how to tie a SAT>IP device always to the same number.

As said above, I dug into the code and saw that nothing can be done with
standard vdr-config-files. Neither diseqc.conf nor sources.conf can help.

However, I added some code and now it works as I wanted it to work.

I did this by adding a Source-member to the satipFrontend created by
satipServer and in satipFrontend::Assign() I additionally check whether the
requested source can be delivered by this frontend.

In my environment, where I have one VDR which will use all satip-devices,
this is exactly what I need.

Now, how to get that correctly done?

Automatic detection? I haven't seen any option on the Octopus-device that
would help the satip-plugin to identify automatically the source/input when
it does the discovery.

So, I'd opt for the command-line-parameter-option: I'd add an additional
parameter to the plugin which will assign the correct source to the
satipFrontend. Which would only work if the -s-option is given.
Maybe it could be even part of the -s-option.. Let's try:

  --server=192.168.1.21|DVBS2-4|Octo1;192.168.1.20|DVBS2-2,DVBT2-2|Octo2

becomes

  --server=192.168.1.21|DVBS2-4|Octo1|S19.2E,S19.2E,S19.2E,S19.2E; \
192.168.1.20|DVBS2-2,DVBT2-2|Octo2|S28.2E,S28.2E,T,T

Looks good to me. I'll try that. 

This will surely work, however, right now, the satip-plugin does not use the
fe=-parameter the RTSP-URL for that. So even with this syntax, it won't be
possible to mix different satellites on _one_ Octopus NET.

Like '[..]Octo2|S28.2E,S19.2E,S28.2E,S19.2E' wouldn't work without changing
the URL-creation-method. It even looked like that the URL is created before
the satipFrontend is chosen. That would have to be changed.

best regards,
--
Patrick Boettcher


___
vdr mailing list
vdr@linuxtv.org
https://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr

Re: [vdr] Sat>IP-plugin: how to configure 2 DD Octopus NETs with different satellites

[Patrick Boettcher]

Hi fnu,

Thank you for your answer.

On Mon, 18 Jul 2016 10:29:58 +0200
"fnu"  wrote:

> Dear Patrick,
> 
> changes in "sources.conf" are IIRC only necessary if you have a
> DiSEqC configuration controlled by VDR. But using a/the generic
> sources.conf doesn't harm VDR operation, I guess most users do have
> one without DiSEqC.

That seems to be true. Before looking at the code of the satip-plugin I
thought that I could limit magically the created (virtual)
satipDevices to the Sources assigned to them via the diseqc.conf.

> You may use a workaround by using CAID field for specific adapter:
> 
> https://www.linuxtv.org/vdrwiki/index.php/Syntax_of_channels.conf#CA_-_Conditional_access

Yeah, no, I don't like this.
 
> But I do not now if it is possible to cover more than 4 devices and
> no good idea how to tie a SAT>IP device always to the same number.

As said above, I dug into the code and saw that nothing can be done with
standard vdr-config-files. Neither diseqc.conf nor sources.conf can
help.

However, I added some code and now it works as I wanted it to work.

I did this by adding a Source-member to the satipFrontend created by
satipServer and in satipFrontend::Assign() I additionally check whether
the requested source can be delivered by this frontend.

In my environment, where I have one VDR which will use all
satip-devices, this is exactly what I need.

Now, how to get that correctly done?

Automatic detection? I haven't seen any option on the Octopus-device
that would help the satip-plugin to identify automatically the
source/input when it does the discovery.

So, I'd opt for the command-line-parameter-option: I'd add an
additional parameter to the plugin which will assign the correct source
to the satipFrontend. Which would only work if the -s-option is given.
Maybe it could be even part of the -s-option.. Let's try:

  --server=192.168.1.21|DVBS2-4|Octo1;192.168.1.20|DVBS2-2,DVBT2-2|Octo2

becomes

  --server=192.168.1.21|DVBS2-4|Octo1|S19.2E,S19.2E,S19.2E,S19.2E; \
192.168.1.20|DVBS2-2,DVBT2-2|Octo2|S28.2E,S28.2E,T,T

Looks good to me. I'll try that. 

This will surely work, however, right now, the satip-plugin does not
use the fe=-parameter the RTSP-URL for that. So even with this syntax,
it won't be possible to mix different satellites on _one_ Octopus NET.

Like '[..]Octo2|S28.2E,S19.2E,S28.2E,S19.2E' wouldn't work without
changing the URL-creation-method. It even looked like that the URL is
created before the satipFrontend is chosen. That would have to be
changed.

best regards,
-- 
Patrick Boettcher

___
vdr mailing list
vdr@linuxtv.org
https://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr