Search
j0ke.net Open Build Service
>
Projects
>
ha
>
glb
> glb.changes
Sign Up
|
Log In
Username
Password
Cancel
Overview
Repositories
Revisions
Requests
Users
Advanced
Attributes
Meta
File glb.changes of Package glb (Revision 17)
Currently displaying revision
17
,
show latest
------------------------------------------------------------------- Tue Feb 12 19:53:19 UTC 2013 - cs@linux-administrator.com - update to release 0.9.2 ------------------------------------------------------------------- Mon Jan 14 20:43:53 UTC 2013 - cs@linux-administrator.com - update to release 0.9.1 ------------------------------------------------------------------- Sat Dec 22 09:58:32 UTC 2012 - cs@linux-administrator.com - update to release 0.9.0 * created subpackage for libglb0 and libglb-devel ------------------------------------------------------------------- Thu Dec 6 13:55:22 UTC 2012 - cs@linux-administrator.com - update to release 0.8.2 * hotfix to slow a connection initialization issue introduced in 0.8.1 * adds a 'random' balancing policy ------------------------------------------------------------------- Tue Dec 4 20:44:22 UTC 2012 - cs@linux-administrator.com - update to release 0.8.1 * fixes connection loss when too many connections were attempted at once * significantly improves performance under certain kinds of load (like sysbench) * adds a max connections option * implements basic "source tracking" capability * adds a start/stop service script and is generally much more suitable to be run as a "service" ------------------------------------------------------------------- Fri May 18 19:02:53 UTC 2012 - cs@linux-administrator.com - update to release 0.7.5 * list of backend servers is configurable in runtime. * supports server "draining", i.e. does not allocate new connections to server, but does not kill existing ones, waiting for them to end gracefully. * on Linux 2.6 and higher glb uses epoll API for ultimate performance. * glb is multithreaded, so it can utilize multiple CPU cores. In fact even on a single core CPU using several threads can significantly improve performance when using poll()-based IO. * connections are distributed proportionally to weights assigned to backend servers.