-
Poul-Henning Kamp authored
If the chosen backend is sick or fails to give us a connection, we rescale the keys position in the backends window onto the all the remaining backends. The upside of this is that we get a stable and predicatable hash distribution. The downside of this is that you are very unlikely to spot the healthy .weight=1 backend between two .weight=100 sick ones before your retries run out. Inspired by: patch from WaveCDN
0352ea25
Name |
Last commit
|
Last update |
---|---|---|
bin | ||
doc | ||
etc | ||
include | ||
lib | ||
m4 | ||
man | ||
redhat | ||
.gitignore | ||
ChangeLog | ||
INSTALL | ||
LICENSE | ||
Makefile.am | ||
README | ||
autogen.des | ||
autogen.sh | ||
configure.ac | ||
varnishapi-uninstalled.pc.in | ||
varnishapi.pc.in |