In our sceanrio, there is one squid server which reverse proxies connection to two other in a sequential (NOT load balanced) oder.
When one peer is down there seems to be a problem with the dead peer detection, the logfile access.log shows, that the peer is dead but tries again to contact him. This causes deleays in establishing the connections. 2013/03/10 22:33:10| Detected DEAD Parent: banetWildcardPeer_squid1 2013/03/10 22:33:10| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:13| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:16| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:18| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:19| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:22| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:29| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:32| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:35| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:33:41| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:24| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:27| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:27| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:30| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:48| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:51| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:51| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:34:55| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:35:52| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:35:54| TCP connection to 188.118.198.34/443 failed 2013/03/10 22:35:54| Detected DEAD Parent: banetWildcardPeer_squid1 Also, although configured exactly like the first squid, the second seems to have ssl errors: 2013/03/11 09:45:25| fwdNegotiateSSL: Error negotiating SSL connection on FD 13: error:1408C095:SSL routines:SSL3_GET_FINISHED:digest check failed (1/0/0) Any ideas?
