hi willy,
Our production environment is use Haproxy-1.5.12 as load balancer,
our haproxy loads a lot of configuration information, resulting in the
start up of memory to reach 300m, after running for a period of time, a
single process memory occupancy rate reached 1G,
We must reload haproxy, memo
>
> Could you please confirm that most of the CLOSE_WAIT are on the front side
> and the ESTABLISHED on the backend side ? If that's the case, can you also
> please verify if there are pending data in the send queue for CLOSE_WAIT
> sockets (3rd column in netstat) ?
>
Most of the close_wait/establi
On Thu, Apr 20, 2017 at 11:33:17PM +0200, Lukas Tribus wrote:
> Hello,
>
>
> Am 20.04.2017 um 15:05 schrieb Sander Hoentjen:
> > A new patch, that puts the order like this:
> > config:
> > crt A crt B
> >
> > [...]
> > If A contains wildcard, and B contains exact match, then wildcard is used.
>
Hello Vladimir,
On Thu, Apr 20, 2017 at 11:58:32AM +0200, Vladimír Houba ml. wrote:
> Hello,
>
> the usual setup for haproxy logging is using an udp. Though, this comes
> with the risk of dropped datagrams. It is very difficult to figure out a
> dropped packet. I think that it would be nice if ea
On Thu, Apr 20, 2017 at 11:03:42PM +0800, jaseywang wrote:
> 1. The backlog of haproxy soon become full and begin to drop new tcp
> connection since peak traffic begin, before CDN, our net.core.somaxconn is
> 1024, and use default backlog of haproxy, everything performs well. After
> using CDN, eve
On Fri, Apr 21, 2017 at 10:23:53AM +0800, jaseywang wrote:
> Hi, Willy
> Thanks for your help. We upgrade the version from 1.5.4 to 1.5.19, but
> still the same issue, and what's your recommended version we can use for
> production env?
OK nice, at least you're not facing one of the many already f
Hi, Willy
Thanks for your help. We upgrade the version from 1.5.4 to 1.5.19, but
still the same issue, and what's your recommended version we can use for
production env?
$ haproxy -vv
HA-Proxy version 1.5.19 2016/12/25
Copyright 2000-2016 Willy Tarreau
Build options :
TARGET = linux26
CPU
Hi Adam,
On Wed, Apr 12, 2017 at 3:00 AM, Adam Spiers wrote:
> Hi all,
>
> I've pored over the Configuration Manual again and again, and I'm
> still struggling to fully understand sticky counters. This paragraph
> seems to hold some important information:
>
>Once a "track-sc*" rule is execu
Hello,
Am 20.04.2017 um 15:05 schrieb Sander Hoentjen:
A new patch, that puts the order like this:
config:
crt A crt B
[...]
If A contains wildcard, and B contains exact match, then wildcard is used.
This last one is different behavior from what is implemented now.
People rely on the specif
On Thu, Apr 20, 2017 at 06:25:17PM +0200, Olivier Houchard wrote:
> Hi,
>
> In server_parse_sni_expr(), we use the "proxy" global variable when I think
> we really want to use the "px" argument, so the attached patch fixes this.
Merged, thanks Olivier!
> Hopefully one day that proxy variable wil
On Thu, Apr 20, 2017 at 07:27:47PM +0200, Nenad Merdanovic wrote:
> Gil,
>
> On 04/05/2017 10:02 AM, Gil Bahat wrote:
> > Hey Nenad,
> >
> > did anything come out of this? interestingly enough, some compliance
> > challenges I was facing brought this to mind again.
>
> Sadly I didn't have any ti
Gil,
On 04/05/2017 10:02 AM, Gil Bahat wrote:
> Hey Nenad,
>
> did anything come out of this? interestingly enough, some compliance
> challenges I was facing brought this to mind again.
Sadly I didn't have any time to work on this. I did check a bit and
found the maps interface somewhat counter-
Hi,
In server_parse_sni_expr(), we use the "proxy" global variable when I think
we really want to use the "px" argument, so the attached patch fixes this.
Hopefully one day that proxy variable will be renamed :)
Olivier
>From eb5033db545ae093f73485e4a29c112e126c159c Mon Sep 17 00:00:00 2001
From:
Now with patch attached, thanks Fred :)
On 04/20/2017 03:05 PM, Sander Hoentjen wrote:
> A new patch, that puts the order like this:
> config:
> crt A crt B
>
> if A contains wildcard, but not exact match, then wildcard is used.
> if A contains exact match, exact match is used.
> (this also means
Hello,
On Thu, Apr 20, 2017 at 11:03:42PM +0800, jaseywang wrote:
> Our haproxy 1.5.4 MS cluster performs quite well before, and the peak
(...)
> Now, the weird thing is why haproxy has so many closewait connections? and
> why the backlog queue soon becomes full? Usually so many closewait means
>
Our haproxy 1.5.4 MS cluster performs quite well before, and the peak
current connections is about 6k. Haproxy forward the request from client to
Nginx, Nginx send the request to upstream JVM servers like this:
client -> Haproxy -> Nginx -> upstream
This week we use CDN to accept request from clie
A new patch, that puts the order like this:
config:
crt A crt B
if A contains wildcard, but not exact match, then wildcard is used.
if A contains exact match, exact match is used.
(this also means that if A contains both wildcard and exact match, exact
match is used.)
If A contains wildcard, and B
Hello HAProxy ML,
On 04/10/2017 05:00 PM, Baptiste wrote:
On Mon, Apr 10, 2017 at 2:30 PM, Willy Tarreau mailto:w...@1wt.eu>> wrote:
On Mon, Apr 10, 2017 at 10:02:29AM +0200, Frederic Lecaille wrote:
> With server templates, haproxy could preallocate 'server' objects which
> would
Hello,
the usual setup for haproxy logging is using an udp. Though, this comes
with the risk of dropped datagrams. It is very difficult to figure out a
dropped packet. I think that it would be nice if each message could have a
counter that would make it possible to detect that some messages were
d
19 matches
Mail list logo
