Yes. I have stop and start the uwsgi.
On Fri, Dec 6, 2013 at 6:23 PM, Roberto De Ioris <[email protected]> wrote: > >> The reload takes about 5 seconds. But I think there is nothing to do >> with that. Because the nginx error log said the connection reset while >> receiving the response. As following: >> >> 2013/12/06 17:49:46 [error] 14090#0: *63748476 recv() failed (104: >> Connection reset by peer) while r >> eading response header from upstream, client: 115.210.50.98, server: >> service.mkey.163.com, request: >> "POST /WSszq1twyG/api/v3/fetch_messages HTTP/1.1", upstream: >> "http://127.0.0.1:4004/api/v3/fetch_mes >> sages", host: "service.mkey.163.com" >> 2013/12/06 17:49:46 [error] 14090#0: *63748453 recv() failed (104: >> Connection reset by peer) while r >> eading response header from upstream, client: 221.194.31.132, server: >> service.mkey.163.com, request: >> "POST /WSszq1twyG/api/v3/check_for_client_update HTTP/1.1", upstream: >> "http://127.0.0.1:4004/api/v3 >> /check_for_client_update", host: "service.mkey.163.com" >> >> sgi >> > > > i have just noted your reload is wrong: > > it should report > > "uwsgi socket 0 inherited INET address :XXX fd XX" > > this line is not present in your logs after the reload > > Are you sure you have fully stopped the instance after the change from > http to http-socket ? (reloading is not enough) > > -- > Roberto De Ioris > http://unbit.it > _______________________________________________ > uWSGI mailing list > [email protected] > http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi _______________________________________________ uWSGI mailing list [email protected] http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi
