On 24/05/2012 16:06, Mladen Turk wrote: > On 05/24/2012 04:47 PM, Mark Thomas wrote: >> On 24/05/2012 05:19, Mladen Turk wrote: >>> On 05/23/2012 10:07 PM, ma...@apache.org wrote: >>>> Author: markt >>>> - apr_time_t *socket_ttl; >>>> - apr_interval_time_t max_ttl; >>>> + apr_time_t *socket_last_active; >>>> + apr_interval_time_t default_timeout; >>> >>> Think we should move both socket_last_active and socket_timeout >>> to tcn_socket_t. They are always bound to socket and we won't have >>> to allocate two additional arrays and copying won't be needed since >>> they will be part of apr_pollfd's client_data >> >> Makes sense to me although I'm unlikely to get to it any time soon. >> > > I'll see if I'll catch some time tomorrow. > >> On a related note, thoughts on a 1.1.24 release? We are going to need >> that for the 7.0.28 release (although that is still a little way off >> with 15+ open bugs to fix first). >> > > Next week?
That works for me. I should be able to run some basic tests with my modified WebSocket coed by then to check that I actually get infinite time outs (not much point releasing if that doesn't work). If I don;t shout, assume all is OK. Mark --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
