Re: tune.bufsize and tune.maxrewrites questions

Hi John. Am 17-09-2015 07:03, schrieb John Skarbek: Good Morning! So recently I went into battle between our CDN provider and our application team due to some HTTP400's coming from somewhere. At first I never suspected haproxy to be at fault due to the way I was groking our logs. The end

4/8/16/24 port poe switch

Hi Dear, We supplying 4/8/16/24 port poe switch including fast, gigabit and web management. PoE Splitter (indoor and waterproof) 12V/9V/5V/0.5A/1A/2A/3A Any interesting, Call me directly! Kate Shenzhen T-Stone Technology Co., LTD Add: 5/F,BlockB,Huaning Indurial Area, Dalang,Longhua,

Re: sticky session not always sticky

Hi Willy, thank you for your quick response. I've changed the parameter but this didn't make any difference. I still see a few connections changing backend servers without me being able to explain this. Do you have any other pointers as to where i can look ? New config : backend

HAProxy + MySQL

Hi haproxy support, Do you know any big companies that uses HAProxy together with MySQL as their main Load Balancer for their busy websites? Thank you. Regards, *Louie* *Sys. Ad. on duty* *Sinonet Technology Inc.*

Suggestions for a UDP/Websockets Load balancer

Hello experts, I am hoping to anyone could share tips on how to create a UDP/Websockets Lua based script to load balance traffic to SIP backend servers. The script should be sip aware, and be able to keep, add or modify Via headers, track connections to which SIP clients is registered to . Is

UpToDate node showing oos:*****?

Hi, i'm a little confused by this, and i don't want to do anything with these systems until i've got some clarity. i'm using drbd 8.3.13 on ubuntu 12.04.3, ( i know i should upgrade, but that isn't an option right now). the metadata partition is on /dev/sda2, the os is installed on /dev/sda1

Re: Suggestions for a UDP/Websockets Load balancer

Websockets uses TCP, and thus can be load balanced like any other TCP service. As for the sip/UDP part, haproxy doesn't support UDP, so that won't be an option. That said, there is still hope. SIP is already capable of load balancing and failure itself via the use of DNS SRV records. You lose

Re: sticky session not always sticky

Hi Willy, i'll attach one session out of the logfile that goes wrong. As the session begins, the first request has NI as cookie state which is logical because it's a new session. Then you see some activity during the session where the cookie state has VN as value. The last entry the cookie

Re: sticky session not always sticky

now with logfile On Thu, Sep 17, 2015 at 4:07 PM, Yves Van Wert wrote: > Hi Willy, > > i'll attach one session out of the logfile that goes wrong. As the > session begins, the first request has NI as cookie state which is logical > because it's a new session. Then you see

Re: UpToDate node showing oos:*****?

sorry, accidentally posted to wrong list, please ignore. (unless you know the answers :P ) On 17 September 2015 at 13:56, Lee Musgrave wrote: > > Hi, > > i'm a little confused by this, and i don't want to do anything with these > systems until i've got some clarity. > >

Re: sticky session not always sticky

Hi Yves, On Thu, Sep 17, 2015 at 04:07:41PM +0200, Yves Van Wert wrote: > Hi Willy, > > i'll attach one session out of the logfile that goes wrong. As the session > begins, the first request has NI as cookie state which is logical because > it's a new session. Then you see some activity during

Re: tune.bufsize and tune.maxrewrites questions

Certainly, ``` [~]$ haproxy -vv HA-Proxy version 1.5.14 2015/07/02 Copyright 2000-2015 Willy Tarreau Build options : TARGET = linux26 CPU = generic CC = gcc CFLAGS = -m64 -march=x86-64 -O2 -g -fno-strict-aliasing OPTIONS = USE_ZLIB=yes USE_OPENSSL=1

Re: sticky session not always sticky

Hi Yves, On Thu, Sep 17, 2015 at 02:12:23PM +0200, Yves Van Wert wrote: > Hi Willy, > > thank you for your quick response. > > I've changed the parameter but this didn't make any difference. I still > see a few connections changing backend servers without me being able to > explain this. Do

Re: sticky session not always sticky

hi Willy, hmm maybe, but i have a second case at which i'm sure the client didn't crash because the jsessionid didn't change before and after the crash. I'll attach the logfile. At line 73 is where the trouble starts. Here the client switches from backend ias01n to ias02n. But i'm sure the

Java Users.

Hi, We maintain the newly updated list of Java Technology Users with complete contact information. We wanted to check if you would be interested in it as they are one of your major targets, this is purely a limited period scheme. Please let me know and I shall get back to

Re: Cannot enable a config "disabled" server via socket command

On 15/09/2015 08:45 πμ, Cyril Bonté wrote: > Hi, > > > Le 14/09/2015 14:23, Ayush Goyal a écrit : >> Hi, >> >> We are testing haproxy-1.6dev4, we have added a server in backend as >> disabled, but we are not able >> to bring it up using socket command. >> >> Our backend conf looks like this: >>

Frontend closes w/ chunked encoding?

We've noticed that our front-end connections to haproxy are closing after talking to a backend running php-fpm. The php-fpm backend is not sending a content-length header, but is using chunked encoding which encodes lengths of the chunks and should be enough to keep the connection alive for

Re: [PATCH] DOC: add references to rise/fall for the fastinter explanation

Hi Pieter, On Thu, Sep 17, 2015 at 09:56:20PM +0200, PiBa-NL wrote: > Sorry, new 2nd patch mail, mixed the words around.. Now it should be good. > > Hi Willy, > > Attached a patch for the doc/configuration.txt to add references for > rise/fall to the fastinter explanation as these are related

Re: [ANNOUNCE] haproxy-1.6-dev5

On 14/09/2015 01:37 μμ, Willy Tarreau wrote: > Hi all, > > we've fixed several bugs since -dev4 so in order to encourage people to > safely test the code, here comes -dev5. > > This has been running few days now and I haven't noticed any problems. I am going to push it *slowly( to other

[PATCH] DOC: add references to rise/fall for the fastinter explanation

Hi Willy, Attached a patch for the doc/configuration.txt to add references for rise/fall to the fastinter explanation as these are related but no links where present. Hope its ok to merge like this? I kept the within 80 characters with, but had to shift around the text a little.. Thanks,

[PATCH] MINOR: support cpu-map feature through the compile option USE_CPU_AFFINITY on FreeBSD

Hi Willy, Added support for cpu-map feature on FreeBSD. It works, but only as long as the as the number of CPU's doesnt rise to high <=64. While the cpuset_t supports up to 256 bits. Maybe a better solution would be to create a separate function like done here

Re: Cannot enable a config "disabled" server via socket command

On Thu, Sep 17, 2015 at 9:42 PM, Pavlos Parissis wrote: > On 15/09/2015 08:45 πμ, Cyril Bonté wrote: >> Hi, >> >> >> Le 14/09/2015 14:23, Ayush Goyal a écrit : >>> Hi, >>> >>> We are testing haproxy-1.6dev4, we have added a server in backend as >>> disabled, but we are

Re: Cannot enable a config "disabled" server via socket command

On 17/09/2015 09:53 μμ, Baptiste wrote: > On Thu, Sep 17, 2015 at 9:42 PM, Pavlos Parissis > wrote: >> On 15/09/2015 08:45 πμ, Cyril Bonté wrote: >>> Hi, >>> >>> >>> Le 14/09/2015 14:23, Ayush Goyal a écrit : Hi, We are testing haproxy-1.6dev4, we have

Re: [PATCH] MINOR: support cpu-map feature through the compile option USE_CPU_AFFINITY on FreeBSD

Hi Pieter, On Thu, Sep 17, 2015 at 09:52:42PM +0200, PiBa-NL wrote: > Hi Willy, > > Added support for cpu-map feature on FreeBSD. > It works, but only as long as the as the number of CPU's doesnt rise to > high <=64. While the cpuset_t supports up to 256 bits. We're limited to 64 as well on

Re: Backend connection resets in http-heep-alive mode

I've started experiencing the same problem, with no HTTP mode set (so the default of http-keep-alive). Did you ever figure out the reason for this? Thanks, Jacob

Re: Frontend closes w/ chunked encoding?

Update: Someone pointed out to me that the requests to haproxy are forced to HTTP/1.0, but the response is HTTP/1.1 w/ chunked encoding. So question is now, if haproxy will accept the chunked encoding to keep alive the frontend connection when there isn't a content-length header, and if it will

Re: Need Help

But how it will be conditional Mean suppose if I want anything that will access help that should go to TCP rest should work with normal http or HTTPS How can I configure this one On Sep 18, 2015 7:05 AM, "Jeff Palmer" wrote: > You can have haproxy listen in both http mode, and

Need Help

Hi, I want to setup haproxy in way there that will work on both http and https and also tpc but that will be conditional mean if any perticular link will come that will go via tcp So can you help me how can i setup this -- Regards Nitesh Kumar Gupta

Re: Need Help

You can have haproxy listen in both http mode, and tcp mode. You'll just need to setup multiple frontends. something like: frontend myhttpservice bind *:80 mode http ... frontend mytcpservice bind *:3306 mode tcp ... If I misunderstood the request, please feel free to post

Re: Need Help

Can you give us an example scenario? Tell us the actual services you want to use on the frontends and backends. and how you would envision your conditions to work? On Thu, Sep 17, 2015 at 9:39 PM, Nitesh Kumar Gupta wrote: > But how it will be conditional > Mean

Re: Need Help

We have 4 servers where tomcat is running server1 server2 server3 server4 and what i want the normal request will serve by the load balanced server via http or https server1 server2 server3 but if any request coming to access /portal that should routed through mode tpc with port 80 or 443 to

Re: Need Help

Oh, I see now. so you want this to all be over the same port? If so, your frontend would have to be in mode tcp, and you'll lose some of the benefits of http mode. With that said, the following should work: frontend foo bind *:80 mode tcp tcp-request inspect-delay 5s tcp request

Re: Need Help

Thanks will try and let you know if it works On Sep 18, 2015 7:44 AM, "Jeff Palmer" wrote: > Oh, I see now. so you want this to all be over the same port? > > > If so, your frontend would have to be in mode tcp, and you'll lose > some of the benefits of http mode. With

Re: Need Help

Please do, I'd be interested in knowing if it worked in reality, the way I suspect it'd work in theory! On Thu, Sep 17, 2015 at 10:29 PM, Nitesh Kumar Gupta wrote: > Thanks will try and let you know if it works > > On Sep 18, 2015 7:44 AM, "Jeff Palmer"

Re: Cannot enable a config "disabled" server via socket command

It'd be interesting to know the complete semantics of the feature you are implementing. I know you understand that our use case is a valid one. And we are open to explore alternative approaches to achieve the same. Till then we will be running haproxy with the flag change reverted and wait for