Ciao, i have a deployment system based on: Ubuntu 14.04.3 LTS lighttpd/1.4.33 Gemstone 3.1.0.6 and Seaside 3.0.1.3 ( based on gsDevKitHome ) daemontools service
All works fine but sometimes, when i submit request from the web browser, the system answer: 500 - Internal Server Error The lighttpd error.log file report : Some records of error.log is added but without any web error. Some, for example at: 2016-12-14 09:20:31: (mod_fastcgi.c.2540) unexpected end-of-file (perhaps the fastcgi process died): pid: 0 socket: tcp:127.0.0.1:9065 2016-12-14 09:20:31: (mod_fastcgi.c.3326) response not received, request sent: 1214 on socket: tcp:127.0.0.1:9065 for /management?_s=ewL9oIqahJYTg2aG&_k=FueEPpFpqVc4Wwx5, closing connection coincides with a web browser 500 - Internal Server Error exception Some considerations? Thanks, Dario -------------------------------------------------------- The lighttpd.conf is:
Where "/etc/lighttpd/gestionale9060.conf" is:
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Dario, If the gem processes associated with the port haven't actually died, then I would think that the seaside server is closing the socket without sending a response... As you note, an internal error in the seaside server could end up closing the socket if there was no valid http response to be be returned ... In the case of the errors that are not associated with any internal error, those do appear to be mysteries, but I am curious what happens when a process hits the http listening port and posts an empty request ... in skimming through the code, there appears to be enough error handling and logging to cover all cases, but I'm not sure what happens when an empty request hits the system ... Dale On 12/14/2016 02:45 AM, Trussardi Dario
Romano via Glass wrote:
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Ciao, thanks.
The behavior is very bizarre.
Both in queuing error in the lighttpd error.log file for example: 2016-12-15 14:09:20: (mod_fastcgi.c.2540) unexpected end-of-file (perhaps the fastcgi process died): pid: 0 socket: tcp:127.0.0.1:9062 2016-12-15 14:09:20: (mod_fastcgi.c.3326) response not received, request sent: 1145 on socket: tcp:127.0.0.1:9062 for /management?_s=K2697sy5zYUSCk-Z&_k=C8jOCY2QikgBu4nO&340, closing connection 2016-12-15 14:41:49: (mod_fastcgi.c.2540) unexpected end-of-file (perhaps the fastcgi process died): pid: 0 socket: tcp:127.0.0.1:9062 2016-12-15 14:41:49: (mod_fastcgi.c.3326) response not received, request sent: 1145 on socket: tcp:127.0.0.1:9062 for /management?_s=K2697sy5zYUSCk-Z&_k=TIWpMrUDoxvi2xDC&244, closing connection It is the fact of when the browser displays the "relative ??? " HTTP 500 error.
For now I could not understand what can cause these illogical behaviors.
Any idea on how to investigate to understand the source of the problem. Thanks, Dario
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
On 12/15/2016 07:58 AM, Trussardi Dario Romano via Glass wrote:
> The behavior is very bizarre. > Both in queuing error in the lighttpd error.log file for example: > > 2016-12-15 14:09:20: (mod_fastcgi.c.2540) unexpected end-of-file > (perhaps the fastcgi process died): pid: 0 socket: tcp:127.0.0.1:9062 > 2016-12-15 14:09:20: (mod_fastcgi.c.3326) response not received, request > sent: 1145 on socket: tcp:127.0.0.1:9062 for > /management?_s=K2697sy5zYUSCk-Z&_k=C8jOCY2QikgBu4nO&340, closing connection > 2016-12-15 14:41:49: (mod_fastcgi.c.2540) unexpected end-of-file > (perhaps the fastcgi process died): pid: 0 socket: tcp:127.0.0.1:9062 > 2016-12-15 14:41:49: (mod_fastcgi.c.3326) response not received, request > sent: 1145 on socket: tcp:127.0.0.1:9062 for > /management?_s=K2697sy5zYUSCk-Z&_k=TIWpMrUDoxvi2xDC&244, closing connection > > > It is the fact of when the browser displays the "relative ??? "HTTP 500 > error. > For now I could not understand what can cause these illogical behaviors. > Any idea on how to investigate to understand the source of the problem. I don't have a clue as to what the problem is, but if I were faced with this I'd run Wireshark (or if the machine is headless, tshark or tcpdump) and try to capture the exact exchange of IP datagrams between Lighttpd and the gem when this happens. Regards, -Martin _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Free forum by Nabble | Edit this page |