Closed wan721 closed 2 years ago
Apparently this was explicitly raised to 120: see #663 for details. I don't think we have a way to detect AddressSanitizer at compile time, so I doubt this can be made programmatically configurable.
@wan721 did you encounter any issue due to that timeout setting being too high? I can see potential problems with bursts of requests in a short time and unlimited pool size, is that your case?
I don't think we have a way to detect AddressSanitizer at compile time
fwiw asan can be detected through macros on both gcc
and clang
@wan721 did you encounter any issue due to that timeout setting being too high? I can see potential problems with bursts of requests in a short time and unlimited pool size, is that your case?
I don't think we have a way to detect AddressSanitizer at compile time
fwiw asan can be detected through macros on both
gcc
andclang
- gcc (ref)
#ifdef __SANITIZE_ADDRESS__ // code that builds only under AddressSanitizer #endif
- clang (ref)
#if defined(__has_feature) # if __has_feature(address_sanitizer) // code that builds only under AddressSanitizer # endif #endif
Thanks for the suggestion! We will try that code.
I got a mail about a post I don't see listed here, which mentioned an Error creating thread: Resource temporarily unavailable
as one of the causes to revise this. Make sure you're setting ulimit
accordingly: https://janus.conf.meetecho.com/docs/FAQ.html#ulimit
https://github.com/meetecho/janus-gateway/blob/cffad3c54eb046ddc457e145ff899d4866d08b6b/src/janus.c#L5352
If we're not using AddressSanitizer, should we comment out this line or set it to a smaller value (in my system the default is 15 seconds).