We have an array of squid servers acting as reverse proxy servers (web accelerators). They also work as URL rewriters, via the redirector interface, eg bouncing http to https in some cases, and mapping certain paths to different backend web servers...
I am getting lots of lines in my access.log files with "000" for the HTTP status code. It looks like it's logging this when it serves a redirect. Here are some example lines (as identified by analog which considers the lines corrupt on account of the 000 status code:)
Wed Dec 17 23:29:05 2003 161 203.45.116.35 TCP_MISS/000 0 GET http://mydomain.dom:443/act/ho/home.xml - DIRECT/163.241.210.3 -
Wed Dec 17 23:29:08 2003 22 203.45.116.35 TCP_MISS/000 0 GET http://mydomain.dom:443/act/ho/home.xml - DIRECT/163.241.210.3 -
Wed Dec 17 23:31:22 2003 6 203.12.97.112 TCP_MISS/000 220 GET http://www.mydomain.dom/ - NONE/- -
Wed Dec 17 23:31:23 2003 30 202.158.104.38 TCP_MISS/000 245 GET http://mydomain.dom/stu/dummy.html - NONE/- -
Wed Dec 17 23:32:29 2003 0 211.108.90.39 TCP_MISS/000 245 GET http://mydomain.dom/stu/dummy.html - NONE/- -
the first and second lines are the web server issueing a redirect, the rest of them are the redirector script telling squid to issue a redirect to https.
Is this supposed to happen? Any way around it so that I can get our log analysis software to read these lines?
Thanks jesse
--
::: Jesse Reynolds +61 (0)414 669 790 ::: AIM - jessedreynolds ::: ::: Virtual Artists Pty Ltd, Adelaide ::: http://www.va.com.au :::
