High-volume HAProxy deployment tips?
I've been using HAProxy in a production environment for some time and am
generally pleased with the results. However, I'd like to get some
feedback from other users and developers on the best way to tune this
for our needs. Basically:
- We have two active clusters. One serves LAMP content with Apache and
the other serves static content (images, thumbnails, etc.) using
lighttpd. Is there any concern in running multiple HAProxy instances on
the same server? We are doing so now but last night's load test produced
mixed results regarding content from the static server.
- Our proxy server is running FreeBSD 7.0 Stable. Are there any
recommended kernel or sysctl flags we can put in place to get optimal
- Has anyone noticed performance or stability issues after a certain
level of bandwidth or number of concurrent connections?
Received on 2008/12/17 22:48
This archive was generated by hypermail 2.2.0
: 2008/12/17 23:00 CET