No, I didn't report it upstream. We didn't change that much of the default configuration, so it was easy for me to use the default settings from 3.2.x and adjust only some settings like acl's. I've planned to upgrade the system to openSUSE 13.2 in the next days. Lets see wether this issue is still present. The issue itself isn't that critical, because squid gets restarted by pacemaker. The critical thing is, that squid needs about 15-25 minutes to rebuild the store (~300GB) and doesn't respond to requests in this time. This behaviour wasn't present in squid 2.x (requests have been answered while the store was rebuilt). I couldn't find this change in behaviour in the changelog.