We've used spread (spread.org) and wackamole (backhand.org/wackamole) with great success to do HA clustering. Had a very bad time with pacemaker.
As for repos, I'd recommend compiling from source to stay current - it's a pretty simple build, not many deps. Cheers, Jeff On Jan 16, 2015 5:47 PM, "Benjamin Smith" <li...@benjamindsmith.com> wrote: > We recently switched our web cluster to use HaProxy 1.5.x from a > combination > of pound and xrctl. So far, we're loving it! > > Some questions have come up, though, that we'd like to entertain: > > 1) Logging performance data: > A) How long before the page started putting out data? (implying > that the > server side is done processing, though not necessarily) > B) How long did the whole cycle take from initial connection to > end of > download? > > 2) Is there a way to anti-DDOS based on a cookie? (Really, we want to > filter > based on user/login but that isn't actually part of the HTTP session, > that's > determined by the cookie) W > > 3) Is there a repo for 1.5.x for RHEL 6 / CentOS 6 ? (I couldn't find one) > > 4) As loads and uptimes vary on our logic servers, we have a script that > sets > the weight for each of our 5 logic servers every 5 seconds, to a weight of > 0 > if a fault is detected. Is there a way to have haproxy set the weight to 0 > if > the script doesn't report in to the load balancer after a timeout? EG: 60 > seconds. > > 5) Is there a relatively simple way to get "true HA" with a redundant load > balancer? We have two identical machines side-by-side running EL6 and > haproxy, > one is a disk dd of the other. In the past we used heartbeat with limited > success; pacemaker has been very problematic for us. For now, we're > managing > manually. > > > Ben Smith > >
