how to get squid 3.1.10 to stop giving IPv6 errors fetching www.google?.com

I installed squid 3.1.10 on CentOS 6.3 with the default squid.conf. When I test it out from localhost: >>> [root@33736 ~]# telnet localhost 3128 Trying ::1... Connected to localhost. Escape character is '^]'. GET [login to view URL] HTTP/1.0 >>> (followed by two carriage returns), it waits about three minutes and then outputs an error page saying: The following error was encountered while trying to retrieve the URL: [login to view URL] Connection to 2607:f8b0:4004:800::1014 failed. The system returned: (110) Connection timed out This didn't happen with Squid 2.6 which I was using previously, but it happens with squid 3.1.10, and it only happens with certainly websites -- namely, these sites all time out with errors similar to the above, when requested through squid: [login to view URL] [login to view URL] [login to view URL] but these do not: [login to view URL] [login to view URL] [login to view URL] This leads me to think that because of something to do with IPv6 in the new version of Squid, it's failing only for certain sites (presumably sites that are IPv6 enabled?). What change do I need to make to make Squid fetch IPv6-enabled sites with no error? (Assuming that's the problem.) Also note there is NO CONNECTION PROBLEM between our server and the remote servers [login to view URL], [login to view URL] and www.facebook.com. On the command line I can connect to all of them on port 80 with no problem and fetch the front pages of those sites. Also, we cannot give out access to the server in order to diagnose the problem. If you want me to run a command I can run it and tell you the output. But this looks like something that is a very common squid-related problem that has come up before, and someone can probably look at the description and tell what is wrong and how to fix it.