3
==========================================================
4
An Introduction to boto's Elastic Load Balancing interface
5
==========================================================
7
This tutorial focuses on the boto interface for Elastic Load Balancing
8
from Amazon Web Services. This tutorial assumes that you have already
9
downloaded and installed boto, and are familiar with the boto ec2 interface.
11
Elastic Load Balancing Concepts
12
-------------------------------
13
Elastic Load Balancing (ELB) is intimately connected with Amazon's Elastic
14
Compute Cloud (EC2) service. Using the ELB service allows you to create a load
15
balancer - a DNS endpoint and set of ports that distributes incoming requests
16
to a set of ec2 instances. The advantages of using a load balancer is that it
17
allows you to truly scale up or down a set of backend instances without
18
disrupting service. Before the ELB service you had to do this manually by
19
launching an EC2 instance and installing load balancer software on it (nginx,
20
haproxy, perlbal, etc.) to distribute traffic to other EC2 instances.
22
Recall that the ec2 service is split into Regions and Availability Zones (AZ).
23
At the time of writing, there are two Regions - US and Europe, and each region
24
is divided into a number of AZs (for example, us-east-1a, us-east-1b, etc.).
25
You can think of AZs as data centers - each runs off a different set of ISP
26
backbones and power providers. ELB load balancers can span multiple AZs but
27
cannot span multiple regions. That means that if you'd like to create a set of
28
instances spanning both the US and Europe Regions you'd have to create two load
29
balancers and have some sort of other means of distributing requests between
30
the two loadbalancers. An example of this could be using GeoIP techniques to
31
choose the correct load balancer, or perhaps DNS round robin. Keep in mind also
32
that traffic is distributed equally over all AZs the ELB balancer spans. This
33
means you should have an equal number of instances in each AZ if you want to
34
equally distribute load amongst all your instances.
38
The first step in accessing ELB is to create a connection to the service.
39
There are two ways to do this in boto. The first is:
41
>>> from boto.ec2.elb import ELBConnection
42
>>> conn = ELBConnection('<aws access key>', '<aws secret key>')
44
There is also a shortcut function in the boto package, called connect_elb
45
that may provide a slightly easier means of creating a connection:
48
>>> conn = boto.connect_elb()
50
In either case, conn will point to an ELBConnection object which we will
51
use throughout the remainder of this tutorial.
53
A Note About Regions and Endpoints
54
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
55
Like EC2 the ELB service has a different endpoint for each region. By default
56
the US endpoint is used. To choose a specific region, instantiate the
57
ELBConnection object with that region's endpoint.
59
>>> ec2 = boto.connect_elb(host='eu-west-1.elasticloadbalancing.amazonaws.com')
61
Alternatively, edit your boto.cfg with the default ELB endpoint to use::
64
elb_endpoint = eu-west-1.elasticloadbalancing.amazonaws.com
66
Getting Existing Load Balancers
67
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
69
To retrieve any exiting load balancers:
71
>>> conn.get_all_load_balancers()
73
You will get back a list of LoadBalancer objects.
75
Creating a Load Balancer
76
------------------------
77
To create a load balancer you need the following:
78
#. The specific **ports and protocols** you want to load balancer over, and what port
79
you want to connect to all instances.
80
#. A **health check** - the ELB concept of a *heart beat* or *ping*. ELB will use this health
81
check to see whether your instances are up or down. If they go down, the load balancer
82
will no longer send requests to them.
83
#. A **list of Availability Zones** you'd like to create your load balancer over.
87
An incoming connection to your load balancer will come on one or more ports -
88
for example 80 (HTTP) and 443 (HTTPS). Each can be using a protocol -
89
currently, the supported protocols are TCP and HTTP. We also need to tell the
90
load balancer which port to route connects *to* on each instance. For example,
91
to create a load balancer for a website that accepts connections on 80 and 443,
92
and that routes connections to port 8080 and 8443 on each instance, you would
93
specify that the load balancer ports and protocols are:
98
This says that the load balancer will listen on two ports - 80 and 443.
99
Connections on 80 will use an HTTP load balancer to forward connections to port
100
8080 on instances. Likewise, the load balancer will listen on 443 to forward
101
connections to 8443 on each instance using the TCP balancer. We need to
102
use TCP for the HTTPS port because it is encrypted at the application
103
layer. Of course, we could specify the load balancer use TCP for port 80,
104
however specifying HTTP allows you to let ELB handle some work for you -
105
for example HTTP header parsing.
108
Configuring a Health Check
109
^^^^^^^^^^^^^^^^^^^^^^^^^^
110
A health check allows ELB to determine which instances are alive and able to
111
respond to requests. A health check is essentially a tuple consisting of:
113
* *target*: What to check on an instance. For a TCP check this is comprised of::
117
Which attempts to open a connection on PORT_TO_CHECK. If the connection opens
118
successfully, that specific instance is deemed healthy, otherwise it is marked
119
temporarily as unhealthy. For HTTP, the situation is slightly different::
121
HTTP:PORT_TO_CHECK/RESOURCE
123
This means that the health check will connect to the resource /RESOURCE on
124
PORT_TO_CHECK. If an HTTP 200 status is returned the instance is deemed healthy.
125
* *interval*: How often the check is made. This is given in seconds and defaults to 30.
126
The valid range of intervals goes from 5 seconds to 600 seconds.
127
* *timeout*: The number of seconds the load balancer will wait for a check to return a
129
* *UnhealthyThreshold*: The number of consecutive failed checks to deem the instance
130
as being dead. The default is 5, and the range of valid values lies from 2 to 10.
132
The following example creates a health check called *instance_health* that simply checks
133
instances every 20 seconds on port 80 over HTTP at the resource /health for 200 successes.
136
>>> from boto.ec2.elb import HealthCheck
137
>>> conn = boto.connect_elb()
138
>>> hc = HealthCheck('instance_health', interval=20, target='HTTP:8080/health')
140
Putting It All Together
141
^^^^^^^^^^^^^^^^^^^^^^^
143
Finally, let's create a load balancer in the US region that listens on ports 80 and 443
144
and distributes requests to instances on 8080 and 8443 over HTTP and TCP. We want the
145
load balancer to span the availability zones *us-east-1a* and *us-east-1b*:
147
>>> lb = conn.create_load_balancer('my_lb', ['us-east-1a', 'us-east-1b'],
148
[(80, 8080, 'http'), (443, 8443, 'tcp')])
149
>>> lb.configure_health_check(hc)
151
The load balancer has been created. To see where you can actually connect to it, do:
153
>>> print lb.dns_name
154
my_elb-123456789.us-east-1.elb.amazonaws.com
156
You can then CNAME map a better name, i.e. www.MYWEBSITE.com to the above address.
158
Adding Instances To a Load Balancer
159
-----------------------------------
161
Now that the load balancer has been created, there are two ways to add instances to it:
163
#. Manually, adding each instance in turn.
164
#. Mapping an autoscale group to the load balancer. Please see the Autoscale
165
tutorial for information on how to do this.
167
Manually Adding and Removing Instances
168
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
170
Assuming you have a list of instance ids, you can add them to the load balancer
172
>>> instance_ids = ['i-4f8cf126', 'i-0bb7ca62']
173
>>> lb.register_instances(instance_ids)
175
Keep in mind that these instances should be in Security Groups that match the
176
internal ports of the load balancer you just created (for this example, they
177
should allow incoming connections on 8080 and 8443).
181
>>> lb.degregister_instances(instance_ids)
183
Modifying Availability Zones for a Load Balancer
184
------------------------------------------------
186
If you wanted to disable one or more zones from an existing load balancer:
188
>>> lb.disable_zones(['us-east-1a'])
190
You can then terminate each instance in the disabled zone and then deregister then from your load
195
>>> lb.enable_zones(['us-east-1c'])
197
Deleting a Load Balancer
198
------------------------