I was looking for a solution for graceful reload of haproxy. I have a nginx server running which passes requests to Haproxy and at times I reload the Haproxy configuration. But I am observing that at time of reloading all the existing connections are getting cut off and haproxy backend queue shows 0 requests ( got from socket stats of haproxy ) .
I am new to HA Proxy and Upstart Scripting. I am using, HA Proxy version 1.4.18 2011/09/16. I am trying to write an upstart script that will keep haproxy alive in case haproxy dies. This is what I have so far:
script
if [ $(pgrep haproxy) ]; then
restart haproxy;
else
start haproxy;
fi
end script
I use haproxy to loadbalancer multiple virtual machines (web services). Each virtual machine has multiple IP addresses.
What is best practice to setup this on haproxy? Right now my configuration are each IP of virtual machine have an entry "listen" on haproxy with corresponding two real server.
Can I ask if it can be revise like multiple "listen" and only two backend server.
I am setting up HAProxy to load balance between two web servers. Some of the pages on the site require SSL. Stunnel is handling the https connections and passing them off to haproxy (Stunnel contains the cert). HAProxy will hand off requests to the web servers using http. Will containing the web servers and haproxy in an internal network be enough to be PCI compliant?
I am running a recent version of HAProxy 1.5 (dev15), and I've noticed that after HAProxy has been running for a while, its memory usage creeps up from initially 0.9% of the total memory to 1.9% and stays there even when there is no more incoming requests.
Email this page
