Upgrading from 1.8.x to 1.9.x: problems with onboard computer

Good to hear that it works now. So these parameters look correct to me

MAV_1_CONFIG=TELEM 2
MAV_1_FORWARD=Disabled 
MAV_1_MODE=Onboard
MAV_1_RATE=80000 B/s 
SER_TEL2_BAUD = 921600

however in the log you sent they are applied to mav_0_* instead of mav_1_*

however in the log you sent they are applied to mav_0_* instead of mav_1_*

Yes, because it has been one of tries.

So these parameters look correct to me

No, because it doesn’t work at all. None of this parameters didn’t provide a communication.

Good to hear that it works now

I don’t think it should work this way. If you noticed, I’ve manually started mavlink instance in shell. On the next reboot it won’t work. To workaround, I can only write init-script, that will do the same things. But it sucks, actually.

The parameters listed above correspond exactly with how you started mavlink manually.

Yes, it seems so, but actually it doesn’t work.
I’ve tried to set this parameters several time, but no result.

I am working on top of 1.9 with those parameters everyday so I think there is some step missing on your side.

there is some step missing on your side

Yes, you’re absolutely right.
The problem came from the file /fs/microsd/etc/extras.txt. I think it’s been one of our experiments with configuring mavlink stream rate. The content of this file was:

mavlink stream -r 0 -s ACTUATOR_CONTROL_TARGET0 -d /dev/ttyS2
mavlink stream -r 10 -s ALTITUDE -d /dev/ttyS2
mavlink stream -r 0 -s ATTITUDE -d /dev/ttyS2
mavlink stream -r 60 -s ATTITUDE_QUATERNION -d /dev/ttyS2
...

After I removed this file, it fixed the problem.

I’m really sorry for all this noise around the problem, shame on me. :confused:
But debugging this kind of bugs is really hard, because I didn’t find anything logged on alerted.
Thank you a lot @mrivi and @Jaeyoung-Lim