search

MattMills / radiocapture-rf

RF side of Radiocapture's SDR based trunked radio bulk collection system
https://radiocapture.com
GNU General Public License v3.0
66 stars 8 forks source link

Question on logs #17

Closed 24e2 closed 4 years ago

24e2 commented 4 years ago

Very cool project.

Are these quit from clients normal with incrementing client numbers?

[2020-07-07 20:21:25,209] (receiver.py:542) - frontend - INFO - quit received from client 4271
[2020-07-07 20:21:26,626] (receiver.py:518) - frontend - INFO - Released channel 9308094e-811b-4646-89e8-5a1f479bb1ca 4272
[2020-07-07 20:21:26,670] (receiver.py:568) - frontend - INFO - connect received from 4274
[2020-07-07 20:21:26,672] (receiver.py:498) - frontend - INFO - 1594178486.67 Created channel ar: 13 12500 851175000 22479 9308094e-811b-4646-89e8-5a1f479bb1ca 4274
[2020-07-07 20:21:26,800] (receiver.py:542) - frontend - INFO - quit received from client 4272
[2020-07-07 20:21:34,128] (receiver.py:518) - frontend - INFO - Released channel 9308094e-811b-4646-89e8-5a1f479bb1ca 4274
[2020-07-07 20:21:34,182] (receiver.py:518) - frontend - INFO - Released channel e262f6e2-7bcc-4871-9236-55a5140122b3 4273
[2020-07-07 20:21:34,247] (receiver.py:542) - frontend - INFO - quit received from client 4273
[2020-07-07 20:21:34,313] (receiver.py:542) - frontend - INFO - quit received from client 4274
[2020-07-07 20:21:34,635] (receiver.py:568) - frontend - INFO - connect received from 4275

I am recording my p25 P1 audio, although it feels like the end of the transmission are being cut off. Need to get out a digital radio to compare.

MattMills commented 4 years ago

Yeah, I'm not super proud of that side of the code, but individual clients are tracked on that side using an incrementing integer, so as channels are opened and closed you should see higher numbers. the ar: ## is the number of currently active receiver channels.

Receive channels will be kept around idle for a time period to minimize flowgraph lock() and unlock() activity, as that is fairly disruptive...

MattMills commented 4 years ago

Also, the P25 code should be fairly well tested, as I've run it for years on a dozen or so sites, the EDACS code is also well tested, the Motorola smartnet code is the least well tested, and might be kind of buggy on different systems (I noticed that for each new motorola smartnet system I added it would need tweaks, because of the weird way their protocol works).

Also, congrats, you are the first person who has mentioned they've gotten the code running!

MattMills commented 4 years ago

Oh also, Re: end of transmission issues, I believe I eventually figured out that was an issue in OP25 itself... and I was using a modified version of OP25 that should be available on my github profile, although it's probably very out of date by now.

I believe I figured out that the way OP25 was buffering packets was not totally compatible with the way my code worked, and would leave off little bits of the buffer without my patch.

24e2 commented 4 years ago

Cool, ill checkout your op25 stuff.. Thanks..

MattMills commented 4 years ago

FYI I also added some stuff to op25 so I could track how many errors each transmission had to try to figure out which recording was "best" when something was recorded to multiple sites, it should be pretty obvious if you want to exclude it.