Hello! Try logging into the pound machine and telnetting to the backend
telnet 192.168.8.51 80 GET / See what happens. This sounds like a wrongly configured iptables or missing route. Possibly wrong reverse path routing? Try using tcpdump to see where your packets hang. tcpdump src or dst port 80 btw I suggest not to post all your config internals on a public mailing list cause everyone in the world will be able to read it and possibly use this information for an attack. This is why most admins try to hide all data about internal structure which makes it a lot harder for an attacker. regards, Felix > -----Ursprüngliche Nachricht----- > Von: mark hardwick [mailto:[email protected]] > Gesendet: Donnerstag, 8. August 2013 18:21 > An: [email protected] > Betreff: [Pound Mailing List] e500 response error read > > Hi Guys > I'm having a problem and my site is currently down. > > Pound has been working fine for me for months and I've had no reason to > change the config. > > Today, suddenly my site went down. Accessing the site flooting.com > <http://flooting.com> gives a > > > "An internal server error occurred. Please try again later." > > > However there are no error messages in the site logs. I was confused. > It turns out, it's pound that is returning the error. Pound is saying it's unable > to connect to the web servers that it's balancing requests over, however the > servers are there and I can access them directly. > > The networks are up, everything is fine. > > The error message is; > Aug 8 18:12:46 six pound: (f7282b70) e500 response error read from > 192.168.8.51:80/GET /login HTTP/1.1: Connection timed out (15.016 secs) > > > There has been no change to the pound configs for a long time. > > I've bumped the log output up, but that's the only message that is output. > which is strange. > also i have an EMERGENCY host configured (and I'd consider this an > emergency) however pound is not redirecting to the emergency page. > > From the point of view of the web servers, the request is never arriving at > the servers. > I'm dead in the water and I can't see anything that may have changed. In > desperation I've rebooted everything. That didn't help. > > Has anyone ever seen anything like this? > > Thanks > Mark > > Configs are: > > ########################################################## > ############ > ## global options: > > User "www-data" > Group "www-data" > #RootJail "/chroot/pound" > > ## Logging: (goes to syslog by default) > ## 0 no logging > ## 1 normal > ## 2 extended > ## 3 Apache-style (common log format) > LogLevel 2 > > ## check backend every X secs: > Alive 30 > > ## use hardware-accelleration card supported by openssl(1): > #SSLEngine "<hw>" > > # poundctl control socket > Control "/var/run/pound/poundctl.socket" > > > ########################################################## > ############ > ## listen, redirect and ... to: > > ## redirect all requests on port 8080 ("ListenHTTP") to the local webserver > (see "Service" below): > # Main listening ports > ListenHTTP > Address 91.187.69.155 > Port 80 > RewriteLocation 2 > > # Roundcube > Service > URL ".*roundcube.*" > BackEnd > Address 192.168.8.45 > Port 80 > End > End > > # Flooting.com <http://Flooting.com> > Service > HeadRequire "(Host: flooting.com <http://flooting.com> |Host: > www.flooting.com <http://www.flooting.com> )" > BackEnd > Address 192.168.8.51 > Port 80 > End > BackEnd > Address 192.168.8.52 > Port 80 > End > Emergency > Address 192.168.16.12 > Port 8080 > End > End > > -- To unsubscribe send an email with subject unsubscribe to [email protected]. Please contact [email protected] for questions.
