Dynamic backend changes without restarting

Does haproxy currently have the ability to be able to alter the backends (add servers, remove servers, etc) without having to restart the process and potentially dropping existing connections? I've read online about various techniques to get around dropped connections like starting a new

Re: Capture entire HTTP request (all headers one shot)

Hello, Just wondering if anyone has other ideas I can try? On Tue, Sep 20, 2016 at 4:31 PM, Manas Gupta wrote: > Hi Willy > Other captures work. > > Here are the examples > https://gist.github.com/manasg/7f4d674a5e07b140e170eadad6858a24 > > Length 0 in req.payload did not

Re: mode tcp and http type acls and keepalive

On Wed, Sep 28, 2016 at 2:51 AM Lukas Tribus wrote: > Hi David, > > > Am 28.09.2016 um 00:40 schrieb David Birdsong: > > I managed to disable keepalive properly and now I see the problem > > completely resolve itself. > > > > Is there some way to use mixed-content frontends w/ a

Re: Rate limiting using HAProxy

AFAIK also those requests responded by HTTP status code 429 are counted against requests count. Is there any delay between test #2 and #3? Mirek Svoboda | +1 (845) 584-1854 | Skype: xsvobo10 On 27 September 2016 at 17:48, Samrat Roy wrote: > Hello Sir, > > I am

Need help with configuration its not working on a new Archlinux VPS

This is the haproxy.cfg file I have been running for years on an Ubuntu 12 VPS, I just installed this on an Archlinux VPS and its not working. Note:     acl has_path path / reqirep ^([^\ :]*)\ /(.*) \1\ /ww/\2 if has_path This is because its a Wt app and needs a url that starts off with a ww

Re: PEM file question

Hello Robert I have seen your question about the requirement of the private key in haproxy and your suggestion that the webserver does not require a private key. I should correct you there. The webserver does use a private key and the intermediate chain from the key signer too. The same for

fetching and string manipulation

Hello, I want to build a stick-table with keys derived from path. Something like this: stick-table type string size 100k store gpc0_rate(60s) tcp-request content track-sc0 function(path) where function( ) does some string manipulation, for example /foo/bar123/baz ---> 123 /foo/bar456/baz

Re: Backend per-server rate limiting

> On 28 Sep 2016, at 10:49, Stephan Müller wrote: > > Hi, > > i want to configure a rate limit (say 100 http req/sec) for each backend > server like this: > > listen front > bind :80 > balance leastconn > server srv1 127.0.0.1:8000 limit 100 >

eternal sticky table

Hello! I would like to build a proxy for galera cluster which will be switch incoming connections between two servers. Both servers are equal - there is no master or slave. However, all connections must be redirected to the single server and when it goes down, they must be switched to the second

Re: mode tcp and http type acls and keepalive

Hi David, Am 28.09.2016 um 00:40 schrieb David Birdsong: I managed to disable keepalive properly and now I see the problem completely resolve itself. Is there some way to use mixed-content frontends w/ a previously opened connection and still detect HTTP (the default HTTP acl)? Provide

Case insensitive redirect maps

Hi, We're using a redirect map and converting from mixed case to lower case via:  http-request redirect location %[path,lower,map(/etc/haproxy/redirect.map)] code 301 if { path,lower,map(/etc/haproxy/redirect.map) -m found } is_url  This seems to be breaking http redirects with a source

Backend per-server rate limiting

Hi, i want to configure a rate limit (say 100 http req/sec) for each backend server like this: listen front bind :80 balance leastconn server srv1 127.0.0.1:8000 limit 100 server srv2 127.0.0.2:8000 limit 100 As far i can see rate limiting is only supported for