Hey,
good news! HA-Proxy version 1.5-dev25-a339395 2014/05/10 is running
since yesterday morning with a 1/4 of our traffic and without coredumps,
infinite loops or other problems. We have nbproc and kqueue enabled and
everything seems to behave fine.
In peak times we had 11000 sockets open and a connection/s rate of about
1300. The Load of the machine with 12 cores was about 2-2.5 - so lots of
headroom.
It seems the fixes worked as anticipated :)
We'll slowly push more traffic over the next couple of days.
I just wanted to give you a quick update as promised.
Kind regards,
John
Willy Tarreau wrote:
Hi John,
On Fri, May 09, 2014 at 11:54:56AM +0200, John-Paul Bader wrote:
Hey Willy,
I have just applied the patch and will run another test after lunch.
Since we're testing with live traffic I can't leave it unattended :)
Just out of curiosity, is this a bug that affects also Linux or is it
FreeBSD specific?
It's generic. And after I found it, I searched how I could reproduce it
and found it could happen on out of memory condition. So I started
"haproxy -m 8" on linux and injected traffic. Bingo, segfault and exact
same trace as yours :-)
We were recently stumped to discover a crash of haproxy on an ALOHA,
and we couldn't explain it, it was in relation with a high memory
usage with SSL. Now I know where it was, so we'll issue an update :-)
Thanks again for all your traces and tests!
Willy
--
John-Paul Bader | Software Development
www.wooga.com
wooga GmbH | Saarbruecker Str. 38 | D-10405 Berlin
Sitz der Gesellschaft: Berlin; HRB 117846 B
Registergericht Berlin-Charlottenburg
Geschaeftsfuehrung: Jens Begemann, Philipp Moeser
