Happy Eyeballs / RFC 8305 support
Thanks a lot for the update. I've cherry-picked e913d050 and 27c657da50 from your GitHub branch and regenerated the config file and documentation.
Thanks for the update and for testing on Windows. I think after two minor changes it can be committed. 1) Please let socket_strerror() take the buffer size as third argument instead of assuming that it's large enough. 2) The comment starting with "/* Cap the number of addresses" in rfc2553_connect_to() doesn't follow the style used elsewhere for multi-line comments. Please adjust it.
Overall the patch looks good to me and seems to work as advertised on my ElectroBSD system. I noticed that if all connection attempts fail, for the last attempt the host name is logged while for the previous attempts the attempted IP address is logged. Is this intentional? It seems inconsistent to me. The modified rfc2553_connect_to() first calculates address_count itself and then overwrites it with the value returned by order_addresses(). Are there cases I'm missing where the ordering changes the...
Happy Eyeballs / RFC 8305 support
Thanks a lot for the patch. I intend to review and test it in a couple of days.
This seems to be a different issue ("X509 certificate verification for t.me failed with error -140: ASN parsing error, invalid input"). Do your wolfSSL CFLAGS include "-DWOLFSSL_NO_ASN_STRICT"?
Error: Loading webpage certificate private key /var/lib/privoxy/certs/cd1a781f134611f5e4cc24f76562bb7d.pem