Segfault in haproxy v1.8 with Lua

2018-04-10 Thread Hessam Mirsadeghi
Hi, I have a simple Lua http-response action script that leads to segmentation fault in haproxy. The Lua script is a simple call to txn.res:forward(0). A sample haproxy config and the Lua script files are attached. The backend is simply an nginx instance which responds with 204 No Content. The

1.8.7 http-tunnel doesn't seem to work? (but default http-keep-alive does)

2018-04-10 Thread PiBa-NL
Hi Haproxy List, I upgraded to 1.8.7 (coming from 1.8.3) and found i could no-longer use one of our IIS websites. The login procedure thats using windows authentication / ntlm seems to fail.. Removing option http-tunnel seems to fix this though. Afaik http-tunnel 'should' switch to tunnelmode

Re: resolvers - resolv.conf fallback

2018-04-10 Thread Ben Draut
I agree. On Mon, Apr 9, 2018 at 1:35 AM, Baptiste wrote: > > > On Fri, Apr 6, 2018 at 4:54 PM, Willy Tarreau wrote: > >> On Fri, Apr 06, 2018 at 04:50:54PM +0200, Lukas Tribus wrote: >> > > Well, sometimes when you're debugging a configuration, it's nice to be >>

Re: DNS resolver and mixed case responses

2018-04-10 Thread Ben Draut
It's interesting that the default behavior of HAProxy resolvers can conflict with the default behavior of bind. (If you're unlucky with whatever bind has cached) By default, bind uses case-insensitive compression, which can cause it to use a different case in the ANSWER than in the QUESTION. (See

Re: Can you help me

2018-04-10 Thread Aleksandar Lazic
Hi. Am 10.04.2018 um 22:14 schrieb Juan Carlos Real Guevara: > Hi and thanks for your response. I think that ive found the problem, and > i look that in the first packets the ip.src is the ip of the balancer > and in the GET to download the archive is with ip.src from the client > and the

Re: Logs full TCP incoming and outgoing packets

2018-04-10 Thread florent
Hello, Thanks for answer. Yes, I would prefer to say no as well but I am not the CTO here ;) I thought about tcpdump as well even if it will kill the performance ! Anyway, I found in the ML archives some relevant informations like this one :

Re: 1.7.10 and 1.6.14 always compress response

2018-04-10 Thread Veiko Kukk
On 04/10/2018 03:51 PM, William Lallemand wrote: On Tue, Apr 10, 2018 at 03:43:12PM +0300, Veiko Kukk wrote: Hi, Hi, This happens even when either compression algo nor compression type are specified in haproxy configuration file. If you didn't specify any compression keyword in the

Site Proposal

2018-04-10 Thread Ruel Revales
Good day! I came across your rockstar site and at the same time is very informative. Can I ask about adding my fantastic site: https://www.topvpncanada.com/ to your page: http://www.haproxy.org/they-use-it.html? Hoping for your kind consideration. Ruel Revales

Re: 1.7.10 and 1.6.14 always compress response

2018-04-10 Thread William Lallemand
On Tue, Apr 10, 2018 at 03:43:12PM +0300, Veiko Kukk wrote: > Hi, > Hi, > > This happens even when either compression algo nor compression type are > specified in haproxy configuration file. > If you didn't specify any compression keyword in the haproxy configuration file, that's probably

Re: [PATCH] epoll: the listener socket use EPOLLEXCLUSIVE flag

2018-04-10 Thread Willy Tarreau
Hi, On Tue, Apr 10, 2018 at 08:38:32PM +0800, ??? wrote: > Hi, all, > haproxy still have the thundering herd problem in the multi-process > mode, the EPOLLEXCLUSIVE flag has been added since linux 4.5, which can > solve this problem Well I disagree with this approach, it will instead

1.7.10 and 1.6.14 always compress response

2018-04-10 Thread Veiko Kukk
Hi, Lets run simple query against host (real hostnames replaced). curl https://testhost01.tld -o /dev/null -vvv Request headers: > GET / HTTP/1.1 > Host: testhost01.tld > User-Agent: curl/7.58.0 > Accept: */* Response headers: < HTTP/1.1 200 OK < Date: Tue, 10 Apr 2018 12:23:44 GMT <

[PATCH] epoll: the listener socket use EPOLLEXCLUSIVE flag

2018-04-10 Thread 龙红波
Hi, all, haproxy still have the thundering herd problem in the multi-process mode, the EPOLLEXCLUSIVE flag has been added since linux 4.5, which can solve this problem 0001-epoll-the-listener-socket-use-EPOLLEXCLUSIVE-flag.patch Description: Binary data

Re: [PATCH] BUG/MEDIUM: threads: Fix the max/min calculation because of name clashes

2018-04-10 Thread Willy Tarreau
On Tue, Apr 10, 2018 at 09:29:37AM +0200, Christopher Faulet wrote: > Sorry, I sent the patch I used for HAProxy 1.8. Here is the right patch, for > the upstream. But good news for you Willy, it will be easier to backport it > in 1.8 now :) Both now merged, thanks for the backport ;-) Willy

Re: Rejected connections not getting counted in stats

2018-04-10 Thread Errikos Koen
Hey Moemen, You are right I was indeed looking at the wrong counter and had not checked the socket output. I assumed it would be available in the stats page or in the metricbeat module which I use to track stats. Thanks for pointing it out! On 4 April 2018 at 19:08, Moemen MHEDHBI

Re: Haproxy 1.8.4 crashing workers and increased memory usage

2018-04-10 Thread Cyril Bonté
Hi Robin, > De: "Robin Geuze" > À: "Willy Tarreau" > Cc: haproxy@formilux.org > Envoyé: Lundi 9 Avril 2018 10:24:43 > Objet: Re: Haproxy 1.8.4 crashing workers and increased memory usage > > Hey Willy, > > So I made a build this morning with libslz and

Re: [PATCH] BUG/MEDIUM: threads: Fix the max/min calculation because of name clashes

2018-04-10 Thread Christopher Faulet
Le 09/04/2018 à 11:52, Christopher Faulet a écrit : Hi, This patch fixes a bug affecting HAProxy compiled with gcc < 4.7 (with threads). It must be merged in 1.8. Sorry, I sent the patch I used for HAProxy 1.8. Here is the right patch, for the upstream. But good news for you Willy, it will