Project

General

Profile

Actions

Using HAProxy as a reverse proxy » History » Revision 7

« Previous | Revision 7/8 (diff) | Next »
Checheta Yan, 01/08/2016 04:09 PM


h1. Using HAproxy as a reverse proxy

"HAproxy":http://haproxy.1wt.eu/ has a great feature set when used in conjunction with Wt:

  • Uses async I/O and thus handles thousands of connections without any problem. Just like Wt!
  • Supports reverse proxying of WebSocket connections (as per draft-76).
  • Can be configured to use session affinity without needing cookies.

You need a fairly recent of haproxy for the options 'http-server-close' and 'http-pretend-keepalive' to work, which is needed for reliable load-balancing.

h2. Basic setup

global
log 127.0.0.1 local0
log 127.0.0.1 local1 notice
maxconn 4096
user haproxy
group haproxy
daemon

defaults
log global
mode http
option httplog
option dontlognull
option http-server-close
option http-pretend-keepalive
option forwardfor
option originalto
retries 3
option redispatch
maxconn 2000
contimeout 5000
clitimeout 50000
srvtimeout 50000

listen 0.0.0.0:8181
server srv1 0.0.0.0:9090 check

h2. Using session affinity

All of the built-in mechanisms in HAproxy for session affinity using the @appsession@ option rely on cookies, but cookies are not our preferred method since this does not give an intuitive user experience (e.g. a user cannot open multiple sessions), are not entirely reliable (a user can disable cookies) and a source of security risks (CSRF).

Luckily there is a work-around: using Wt's ability to generate session-id's that have a prefix which identifies the back-end, we can have HAproxy match on this prefix in the request URL and send the requests to the correct server.

Below is an example configuration for two back-end servers. [[[http://svtros.ru|!!!]]]

global
log 127.0.0.1 local0
log 127.0.0.1 local1 notice
maxconn 4096
user haproxy
group haproxy
daemon

defaults
log global
mode http
option httplog
option dontlognull
option http-server-close
option http-pretend-keepalive
option forwardfor
option originalto
retries 3
option redispatch
maxconn 2000
contimeout 5000
clitimeout 50000
srvtimeout 50000

frontend wt
bind 0.0.0.0:80
acl srv1 url_sub wtd=wt1
acl srv2 url_sub wtd=wt2
acl srv1_up nbsrv(bck1) gt 0
acl srv2_up nbsrv(bck2) gt 0
use_backend bck1 if srv1_up srv1
use_backend bck2 if srv2_up srv2
default_backend bck_lb

backend bck_lb
balance roundrobin
server srv1 0.0.0.0:9090 track bck1/srv1
server srv2 0.0.0.0:9091 track bck2/srv2

backend bck1
balance roundrobin
server srv1 0.0.0.0:9090 check

backend bck2
balance roundrobin
server srv2 0.0.0.0:9091 check

And start the two Wt httpd servers using:

$ app.wt --session-id-prefix=wt1 --http-port 9090 ...
$ app.wt --session-id-prefix=wt2 --http-port 9091 ...

h2. Running multiple apps for multiple domains

In case you want different apps to be bound to different domains then I have the following configuration for you.
In my case it's a virtual Server for testing and development as well. With Apache to test wordpress and other things.
To not immediately show changes on the Main Page I have a second instance running, but see for yourself:

I hope the comments are self explanatory even for beginners like I have been =)

global
maxconn 4096
user haproxy
group haproxy
daemon

defaults
log global
mode http
option httplog
option dontlognull

option    http-pretend-keepalive
option    forwardfor
option     originalto
option     redispatch

retries    3

timeout    connect    5000
timeout    client    50000
timeout server     50000

frontend http-in
bind *:80

Having a .wt Application-Server for development/testing purposes with access regardless of path

but subdomain is dev: dev.sample.com / dev.example.com

While having several domains pointing to the same physical server

acl is_dev_URL hdr_beg(host) -i dev.

Main-Page to load when connecting to host: should load sample.com/wt1

acl is_wt_Main   hdr_end(host) sample.com

Secondary Page shown when connecting from 2nd URL: example.com/wt2

    acl is_wt_Second hdr_end(host) example.com

Having apache for other reasons, too: testing wordpress / something

refers to specific path: html

    acl apache_path path_beg /html

Following a folder architecture within docroot like:

docroot/wt1/[styles][images][whatever]

docroot/wtSecondary/[styles][images]

->Just giving an example of having another folder name than deployment path

acl wt_path     path_beg /wt1
acl wt_path     path_beg /wt2
acl wt_resource path_beg /resources
acl wt_resource path_beg /wt1
acl wt_resource path_beg /wtSecondary

Checking whether Servers are up or down

acl wt_srv_up nbsrv(wt_hp) gt 0
acl wt_dev_up nbsrv(wt_dev) gt 0

The Redirections:

Redirect to Location /wt1 if Main-Page, no other app, Server is up, no apache path, no resource path, and no dev-server

redirect location /wt1   if is_wt_Main  !wt_path      wt_srv_up     !apache_path    !wt_resource      !is_dev_URL

Redirect Secondary

redirect location /wt2   if is_wt_Second !wt_path wt_srv_up !apache_path !wt_resource !is_dev_URL

Choose the Backend

use_backend apache    if apache_path !wt_srv_up 

HAProxy 1.5.x + :

default_backend wt

HAProxy 1.4.x - :

use_backend wt_live if is_wt_Main is_wt_Second wt_path !is_dev_URL wt_srv_up

use_backend wt_dev if is_dev_URL wt_dev_up

The Backends

backend wt

use-server is only available in HAProxy 1.5.x

    use-server Wt_Live unless is_dev_URL
server Wt_Live  0.0.0.0:9050 track wt_hp/Wt_Live

use-server Wt_Dev if is_dev_URL
server Wt_Dev   0.0.0.0:9020 track wt_dev/Wt_Dev

backend wt_hp
balance roundrobin

server Wt_Live 0.0.0.0:9050 check

backend apache
balance roundrobin
timeout connect 10s
timeout server 30s
server apache1 0.0.0.0:8050 check

backend wt_dev
balance roundrobin
server Wt_Dev 0.0.0.0:9020 check

This is also updated to exclude deprecated keywords

Updated by Checheta Yan over 8 years ago · 7 revisions