terminate called after throwing an instance of 'std::out_of_range'

[mincequi]

terminate called after throwing an instance of 'std::out_of_range'

[mincequi]

@detroitgoose,

Ok, having a stack trace would not help, if the tools runs without fault. It depends a little bit on your environment, but checkout some documentation on how to enable "core dumps" (sorry, the wording stack trace might have been misleading).
https://wiki.archlinux.org/title/Core_dump

If the tool crashes, you get a core file. Together with cornrow's executable you can trace back the issue:
gdb executable core-file

[detroitgoose]

Adding cornrow config if pertinent:

[bluetooth_source]
codecs = [ 0x0 ]        # As defined in the A2DP specification (SBC (0x0) is mandatory)
sample_rates = [ 44100, 48000 ]

[airplay_source]
name = "KitchenAirplay"
 port = 0              # port to listen on (0 for auto select)
 buffer_time = 2000    # in ms (defaults to 2000ms)

#[alsa_sink]
#device = "default"

[tcp_sink]
host = "127.0.0.1"
port = 4953

[mincequi]

Hmm, this map out-of-range is suspicious.
But in any case: your config file has an error: [tcp_sink]`<- this symbol might cause an error when parsing the config file.

Maybe you can produce a stack trace?

[detroitgoose]

@mincequi

your config file has an error: [tcp_sink]`<- this symbol might cause an error when parsing the config file.

Oops, that was a result of my poor github markup. I've edited the post to show the config as it actually is: without the backtick.

Maybe you can produce a stack trace?

I tried sudo strace -s 99 -ffp 766, 776 being the current process ID, but I'm not sure how helpful the data is. I get the following response which repeats ad infinitum:

[pid   766] clock_gettime64(CLOCK_MONOTONIC, {tv_sec=15991, tv_nsec=591824800}) = 0
[pid   766] clock_gettime64(CLOCK_MONOTONIC, {tv_sec=15991, tv_nsec=591944428}) = 0
[pid   766] epoll_wait(10, [], 128, 0)  = 0
[pid   766] clock_gettime64(CLOCK_MONOTONIC, {tv_sec=15991, tv_nsec=592170463}) = 0
[pid   766] poll([{fd=3, events=POLLIN}, {fd=6, events=POLLIN}, {fd=7, events=POLLIN}, {fd=8, events=POLLIN}], 4, 1) = 0 (Timeout)

This was done while the program is running without fault.

If you could tell me the proper way to do a stack trace, I'll run it next time the error occurs.

[detroitgoose]

@mincequi

I enabled core dumps 2 weeks ago, but it hasn't crashed since.

Not sure what happened but I'll mark this closed. Thanks again for such seamless BT/AP integration