Minisatip tries wrong PMT #1129
Comments
|
I think problem is pmt 48 is misinterpreted as audio. But I don't know much about coding. |
|
i think this happens when the VPID (and PCR) is the same on multiple channels and only APID is different (or reverse, same APID but different VPID on different PMTs like beIN SPORTS ticari/commercial and non-commercial) as it also happens on d***t 42e as well with stingray music channels (12188 V 27500, 8PSK DVB-S2 5/6), all music channels like for instance if you load up i've included a sample log from a raspberry pi 4 with TurboSight TBS 5930 DVB-S/S2/S2x |
|
Thanks for the additional clues! |
|
So it seems that the problem is that |
|
Same issue in @surfcu 's log (PMT 1 (52) gets incorrectly marked as master for PMT 10 (48)): |
|
@catalinii what do you think would be the best fix for this? Can you elaborate what the idea behind "master PMT" is? Does it make any sense anymore? |
|
Generally for a vpid/apid, you can have multiple PMTs reporting those vpid/apids (for example the German PMT and the Austrian PMT). obviously the user owns just one card, however it is imposing to know which one of them is it. so Master PMT is the first PMT that minisatip receives and the other PMTs are linked to the master, so when the CAPMT is sent it contains all the capids and caids for all PMTs that contains the same vpid/apid to ensure it can be decrypted. The fix would be to identify why the wrong PMT is linked to the master PMT (see process_pmt) |
|
Perhaps the logic should be changed then to look for a PMT that contains both the new PMT's VPID and APID, not just either one of them? Not sure how much better that is but it should at least be a little less prone to false positives. |
|
Then again I don't know what would happen to encrypted radio channels (that presumably don't have a VPID at all) |
|
Jalle19@bdb643b trying a fix here, haven't tested it yet |
|
Seems to be that going from a random PID to a PMT is not possible to get absolutely right |
|
Try it, let’s see if it works |
|
|
|
@surfcu @iandebris can you try this branch? https://github.com/Jalle19/minisatip/tree/master-pmt2 |
|
Testing it myself at the moment, seems to work just fine |
@Jalle19 Heya, your
[global]
logfile = stdout
lb_mode = 1
lb_nbest_readers = 2
lb_force_reopen_always = 1
lb_stat_cleanup = 24
disablecrccws = 1
disablecrccws_only_for = 092B:000000;06F8:000000;0656:000000
[cache]
[dvbapi]
enabled = 1
au = 1
pmt_mode = 4
request_mode = 1
listen_port = 15012
delayer = 65
user = dvbapi
read_sdt = 2
write_sdt_prov = 1
demuxer_fix = 1
boxtype = pc
[webif]
httpport = 8888
httpallowed = 127.0.0.1,192.168.0.0-192.168.255.255,10.0.0.0-10.255.255.255,172.16.0.0-172.31.255.255,::1
[account]
user = dvbapi
uniq = 2
au = 1
max_connections = 5
group = 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16 |
|
I figured that could happen. Need to find a better way to "start" only the right PMT. |
|
@iandebris are you able to test the Bein Sports 2 vs. Ticari bein sports case? |
@Jalle19 sorry, although I don't have an access card for the provider of those channels; I can at least confirm the same PMT demux flood happens on oscam: |
|
@catalinii in |
|
The idea is that minisatip will cache all PMTs for all seen transponders, so the list of pmts (and pmt-> master pmt) needs to be correct regardless of the subscribed pids. The other reason is that the clinet can chabge channels (and subscribed pids) without chaging the transponder. Again the subscribed pids should not be relevant. PAT/PMT scanning is an expensive operation as you need to wait for them to arrive which could take more than 1s |
|
get_master_pmt_for_pid checks if vpid and apid match and assigns it as master_pmt. I think the solution would be using SID instead of master_pmt. It was discussed before for different purposes here |
|
I tried to fix this at one point but stalled because the fix wouldn't have been complete (would only have worked for channels that share either a video or audio PID, not both). Going by SID sounds like a good solution. |
|
Just clarify: A pid cannot be the video pid for 2 channels because it representa just one stream. if you want to offer access to your channel (via ssh) I can take a look |
Actually they are broadcasting same content for different packages under different channel names.
I am away for business. |
|
Try the latest binary and add argument -9 |
|
for me it did not work... without adding -9 all worked as before... when I added -9 at startup then oscam decoding wont work |
|
It's working without any issues with -9 argument now. |
|
when i start minisatio with -9 then it looks like oscam cant decode the cannel because it is using the wrong ecm pid |
|
minisatip_nok.log I add the logs with starting normally and starting with -9 |
|
@jyrts the pid that contains the ecm for you is 7301 which is part of pmt pid 312 not 412. |
|
Yes, when I add -9 on startup and change pid 412 to 312 then all working good. Case closed then? :) |
|
Does it make sense to have this as an option, or should this simply be the default behavior? |
|
What is the downside of delaying pmt scan to process_pat? |
There are 2 cases where -9 does not work when people use other clients than tvh/vdr: |
This option disables the scan and will read only pmt pids that the client requests |
According to SATIP specifications media stream object must be fully and correctly defined. |
|
To me it seems reasonable to assume that any client looking to decrypt a channel will send the PMT PID. Are there any clients that don't do that? |
|
People using playlists or anything that does not infer pids based on the list of channels from the satellite |
I can watch all Digital Platform channels on my enigma2 (Zgemma h9 twin se) box 42e.
When running Minisatip can provide all but beIN SPORTS 2 (PMT=48).
Instead when pmt=48 is requested minisatip tries Ticari beIN Sports 2 (PMT=52) and fails since it's not in the subscription.
minisatip.log
Any ideas?
The text was updated successfully, but these errors were encountered: