Wondering how to use docker swarm haproxy? Our experts have put together this guide to help you get started. Our Docker Support team is here to lend a hand with your queries and issues.

How to use docker swarm haproxy?

Use HAProxy to add routing, load balancing, and DNS service discovery to Docker Swarm

Today, let us see the steps followed by our support techs to setup swarm haproxy.

HAProxy Configuration

Prior to creating your HAProxy service in Swarm, prepare your configuration file.

Please note that you have to store this file at the same path on all of your cluster nodes that could run your HAProxy container. We use /etc/haproxy/haproxy.cfg.

Firstly, create the file and add a global section to the top:

global
log fd@2 local2
chroot /var/lib/haproxy
pidfile /var/run/haproxy.pid
maxconn 4000
user haproxy
group haproxy
stats socket /var/lib/haproxy/stats expose-fd listeners
master-worker

Here, there are two important settings.

• The first is the log fd@2 local2 line. This will provide the logs to Docker, which can process them depending on the log driver you’ve set.
• The second is the master-worker line: It allows you to reload the HAProxy configuration without needing to restart the container itself.

Next, add a resolvers section:

resolvers docker
nameserver dns1 127.0.0.11:53
resolve_retries 3
timeout resolve 1s
timeout retry 1s
hold other 10s
hold refused 10s
hold nx 10s
hold timeout 10s
hold valid 10s
hold obsolete 10s

The Docker Swarm DNS service is always available at 127.0.0.11. So, this section configures

HAProxy to direct DNS queries to there. This is essential for DNS-based service discovery.

Next, add a defaults section so that there are some sensible timeouts and other settings:

defaults
timeout connect 10s
timeout client 30s
timeout server 30s
log global
mode http
option httplog

Finally, finish with some frontend and backend sections:

frontend fe_web
bind *:80
use_backend stat if { path -i /my-stats }
default_backend be_apache_service
backend be_apache_service
balance roundrobin
server-template apache- 6 apache-Service:80 check resolvers docker init-addr libc,none
backend be_apache_service_wrong_case
balance roundrobin
server-template apache- 6 apache-service:80 check resolvers docker init-addr libc,none
backend stat
stats enable
stats uri /my-stats
stats refresh 15s
stats show-legends
stats show-node

This configuration is very basic, but take a closer look at the server-template lines in the backend sections. A server-template generates server lines.

Create an HAProxy Service

At this point, you have:

• an Apache service defined, but not yet running on any nodes
• an HAProxy configuration file that’s been copied to each node into the folder /etc/haproxy/haproxy.cfg

Now you can create the HAProxy service in Swarm. There are three ways to do this:

1. Firstly, create only one HAProxy container and let Swarm’s ingress routing mesh forward clients’ requests to it.
2. Then, create only one HAProxy container, but don’t use the ingress routing mesh; HAProxy will receive clients’ requests directly.
3. Finally, create a replica of HAProxy on each node; Each will receive clients’ requests directly.

Placement Constraints

A quick word about placement constraints: You may, as an example, have a cluster of 20 nodes and you would like to control where your HAProxy containers could be started when using --mode replicated.

In this case, you should use placement constraints.

They allow you to restrict which nodes Swarm can choose to run your containers.

Let’s say you would like to run HAProxy only on nodes 2 and 3; Run these commands on any manager:

$ sudo docker node update –label-add LB-NODE=yes dock2
$ sudo docker node update –label-add LB-NODE=yes dock3

Then, specify the constraint at service creation:

$ sudo docker service create \
–mode global \
–name haproxy-service \
–network apache-network \
–publish published=80,target=80,protocol=tcp,mode=host \
–publish published=443,target=443,protocol=tcp,mode=host \
–mount type=bind,src=/etc/haproxy/,dst=/etc/haproxy/,ro=false \
–dns=127.0.0.11 \
–constraint node.labels.LB-NODE==yes \
haproxytech/haproxy-debian:2.0 /usr/sbin/haproxy -f /etc/haproxy/haproxy.cfg -L local_haproxy

Or on an existing service with the docker service update command:

$ docker service update –constraint-add node.labels.LB-NODE==yes haproxy-service

The container will only started on nodes with a matching label.

[Need assistance with a different issue? Our team is available 24/7.]

Conclusion

To conclude, our Support Engineers demonstrated how to how to use docker swarm haproxy.