]> git.kaiwu.me - haproxy.git/commit
projects / haproxy.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a33410c) | patch
BUG/MINOR: fix forcing fastinter in "on-error"
authorSergiy Prykhodko <pservit@gmail.com>
Sat, 21 Sep 2013 09:05:00 +0000 (12:05 +0300)
committerWilly Tarreau <w@1wt.eu>
Mon, 23 Sep 2013 06:49:50 +0000 (08:49 +0200)
commit1d57e508104ccd085c1e70d147edcba9f6960c1b
tree86a8544c4fb3236e6f743f94d81a9817a2fd1fb6tree | snapshot
parenta33410cf94db24d3ec83003a40aeec62897cca97commit | diff
BUG/MINOR: fix forcing fastinter in "on-error"

health_adjust() should requeue the task after changing its expire timer.

I noticed it on devel servers without load. We have long inter (10 seconds)
and short fasinter (100ms). But according to webserver logs, after a failed
request next check request was called with same 10s interval.

This patch should probably be backported to 1.4 which has the same feature.
src/checks.c diff | blob | history
haproxy upstream