[mythtv-users] DVB-S2 and DVBT2 via SAT>IP

[Ken Smith]

Hi Everyone,

I've been experimenting with minisatip on a system that has a TBS Dual 
S2 card, a TBS Dual T2 card and a Hauppauge DVB-T card. I've been 
collecting logs and wireshark captures, with various combinations of 
enabled interfaces, to find out why DVB-S and DVB-T don't work 
together.  I'm finding that I can get DVB-T to work or DVB-S, but not 
both together. My aim is to have the host system be a SAT>IP server and 
run MythTV backend on a VM. It should simplify backend upgrades.

In one of my tests I had two DVB-S2 interfaces enabled and one DVB-T2. I 
used MythTV to run a DVB-S2 channel scan.

minisatip logs the DVB-S2 request from MythTV like this:-

minisatip[733468]: [02/02 18:32:38.786 main]: detect_dvb_parameters 
(S)-> 
fe=1&src=1&freq=11428.00&pol=h&ro=0.35&msys=dvbs&mtype=qpsk&sr=27500&fec=23&plts=auto

A subsequent channel scan of the T2 interface is logged by minisatip 
like this

[02/02 18:48:09.572 main]: copy_dvb_parameters -> src=1, fe=1, 
freq=474000, fec=0 sr=0, pol=0, ro=3, msys=3, mtype=0, plts=2, 
bw=8000000, inv=2, pids=, apids=NULL, dpids=NULL x_pmt=NULL

wireshark captured this request as

rtsp://...:554/?fe=1&freq=474.00&bw=8&msys=dvbt&tmode=auto&mtype=auto&gi=132&fec=auto

This second scan of the T2 source fails with "no channels found".

Notice that in both the S2 scan and the T2 scan, MythTV requests fe=1.

I initially asked about this on the minisatip github and the suggestion 
from there is that MythTV should omit specifying fe= parameter in its 
SAT>IP requests. Instead the msys= parameter be used to identify the 
type of tuner being addressed. I'm aware that there is also a session 
id. I'm not fully familiar with the SAT>IP Specification, however, I do 
see the fe= parameter used in many examples in that Specification.

I've kept all the logs and wireshark captures in case they are useful. I 
could re-run some of my tests, possibly also using dev-fe-tool to track 
which tuner is actually getting addressed.

Although to me it seems suspicious that, although fe=1 appears in both 
the S2 and T2 scans, that does not explain why a system with just two 
S2, or just two T2 tuners is working.

Thoughts???


Thanks

Ken






-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.