Re: Backend ipv6 : server always DOWN
Hi, Thanks for your reply Kévin 2014-03-10 16:17 GMT+01:00 Cyril Bonté cyril.bo...@free.fr: Hi, Le 10/03/2014 15:39, Kevin Maziere a écrit : Hi I'm trying to use haproxy with ipv6 adresses in my backend, and it always fail. Can you confirm that is a no working feature My configuration : backend ipv6 mode http balance roundrobin option forwardfor option httpchk *server test ::1:8080 weight 5 check inter 2000* #= *failed* Error : server test has neither service port nor check port. Check has been disabled Then I fix : *server test ::1:8080 weight 5 check port 8080 inter 2000* And : [WARNING] 068/144421 (3289) : Server ipv6/test is DOWN, reason: Layer4 connection problem, info: Connection refused, check duration: 0ms. 0 active and 0 backup servers left. 0 sessions active, 0 requeued, 0 remaining in queue. But my service is available on [::1]:8080 over wget/curl ... OS : Ubuntu 12.04 Version : 1.4.18-0ubuntu1.2 Do I miss something ? Yes, haproxy 1.4 doesn't support IPv6 addresses for servers. http://cbonte.github.io/haproxy-dconv/configuration-1.4.html#server -- Cyril Bonté
Re: Log format with odd number of quotes. Bug or conf error?
On Mon, 10 Mar 2014 17:30:27 -0400 Julien Vehent jul...@linuxwall.info wrote: Thanks Thierry ! It is trivial indeed, but like you I was concerned about triggering another issue by removing it... What about the second issue I'm seeing? Should request/response headers be completely absent from the logs when empty? And is there a way to force a or - sign when they are empty, to play nice with log parsers? Thanks, Julien It's a bug, indeed, and thierry's patch is the right way to fix it. -- William Lallemand
haproxy 1.5 builds for fedora/rawhide
For those interested, I have built haproxy-1.5-dev22 for Fedora. The packages are located in a copr repo since the distribution repos still contain version 1.4. The project and repos can be found here: http://copr.fedoraproject.org/coprs/rohara/haproxy/ The SRPM can be found here: http://rohara.fedorapeople.org/haproxy/haproxy-1.5-0.3.dev22.fc20.src.rpm Feel free to contact me if you have questions. Ryan
Re: Log full request/response or forward to 2 servers same request and the response
Hi Dário, please always send your email to the mailing list as-well. I managed it on java but I would like to use HAProxy due to performance and credibility... Do you think it would be hard to add this feature? I'm not sure it will be particularly easy to add something like that. HAProxy tries to be as efficient as possible, but your feature goes in the other direction. The HAProxy sends the request and waits for response or it uses async calls to backend? If the call is sync, I guess it would not be that hard... I'm still decoding the code to start contributing... If your goal is to get such a feature included in HAProxy, I suggest you discuss it here on this list with Willy et al. first, before starting to work on it. (similar to parallel proxy) Quite a task, if thats where you are heading ... Regards, Lukas
Re: Log full request/response or forward to 2 servers same request and the response
I decide to do it with Java NIO.. It's not efficient as HAProxy and C but enough for a proof of concept... Since I don't want use on production, it's enough... I couldn't agree more, it goes on totally different route. Thanks a lot, fail fast and cheap :) On 11 March 2014 23:00, Lukas Tribus luky...@hotmail.com wrote: please always send your email to the mailing list as-well. -- *Dário Nascimento* *Master Degree in Computer Science and Networks Engineering Student* *Instituto Superior Técnico - Technical University of Lisbon* *Skype: dario-nascimento darionascimento.com http://darionascimento.com*
tcp-request content track
2 related questions: I'm trying to find a way to concat multiple samples to use in a stick table. Basically in my frontend I pattern match on the request path to determine which backend to send a request to. The client requests also have a client ID header. I want to rate limit based on a combination of this pattern that matched, and the client ID. Currently the way I do this is an http-request set-header rule that adds a new header combining a unique ID for the pattern that matched along with the client-ID header. Then in the backend I have a tcp-requst content track-sc2 on that header. This works, but I'm wondering if there's a better way. Secondly, the above works, but when I do a show table mybackend on the stats socket, both the conn_cur and use counters never decrease. They seem to be acting as the total number of requests, not the number of active connections. Is this a bug, or am I misunderstanding something? -Patrick
