Hello all. We have 2 Farms, an inner(intranet, just as a filter, no hdd caching) and an outer(DMZ, just hdd chaching) one. We use CARP to optimise the caching in the outer proxies. Now, all connect requests are routed to only one single machine. While that wasn't an issue in the last years, we have now reached a point where that single one is no longer capable to handle all the incomming requests, even after setting its carp weighting to a minimum.
To my knowledge my predecessor already brought this up years(?) ago and got the answer (from Henrik?) that this behaviour is actually wanted... Well, if that's the case I'd like to patch just our squid version. I have however some problems to find the point where I'd have to make the needed changes. Anyone able to give me some hints? regards, Oliver Baumgaertel
