aweber/rabbitmq-autocluster

Can't get autoscaling=true working with IAM role

motilevy opened this issue · 3 comments

motilevy commented

I have the following IAM policy attached to a role my instances are on:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "autoscaling:DescribeAutoScalingInstances",
                "ec2:DescribeInstances"
            ],
            "Resource": [
                "*"
            ]
        }
    ]
}

Launching a container:

docker run -d --name rabbitmq  --net=host \
        -p 4369:4369 -p 5672:5672 \
        -p 15672:15672 \
        -p 25672:25672 \
        -e AUTOCLUSTER_TYPE=aws \
        -e AUTOCLUSTER_LOG_LEVEL=debug \
        -e AUTOCLUSTER_CLEANUP=true \
        -e AWS_AUTOSCALING=true \
        -e AUTOCLUSTER_DELAY=10 \
        -e CLEANUP_WARN_ONLY=false \
        -e AWS_DEFAULT_REGION=us-east-2 \
aweber/rabbitmq-autocluster

fails with the following:

=INFO REPORT==== 31-Jan-2017::03:59:53 ===
node           : rabbit@ip-10-5-11-218
home dir       : /var/lib/rabbitmq
config file(s) : /usr/lib/rabbitmq/etc/rabbitmq/rabbitmq.config
cookie hash    : iqG7DCBA+lxNNLQq/Y6efg==
log            : tty
sasl log       : tty
database dir   : /var/lib/rabbitmq/mnesia

=INFO REPORT==== 31-Jan-2017::03:59:54 ===
autocluster: log level set to debug

=INFO REPORT==== 31-Jan-2017::03:59:54 ===
autocluster: Using AWS backend

=INFO REPORT==== 31-Jan-2017::03:59:54 ===
autocluster: Delaying startup for 8744ms.

=INFO REPORT==== 31-Jan-2017::04:00:03 ===
autocluster: Starting aws registration.

=ERROR REPORT==== 31-Jan-2017::04:00:03 ===
Failed to retrieve AWS credentials: undefined

=INFO REPORT==== 31-Jan-2017::04:00:03 ===
autocluster: Setting region: "us-east-2"

=ERROR REPORT==== 31-Jan-2017::04:00:03 ===
autocluster: Error fetching autoscaling group instance list: credentials

--- snip --- 
BOOT FAILED
===========

Error description:
   {could_not_start,rabbit,
       {function_clause,
           [{autocluster,maybe_register,
                [error,aws,autocluster_aws],
                [{file,"src/autocluster.erl"},{line,111}]},
            {autocluster,init,0,[{file,"src/autocluster.erl"},{line,33}]},
            {rabbit_boot_steps,'-run_step/2-lc$^1/1-1-',1,
                [{file,"src/rabbit_boot_steps.erl"},{line,49}]},
            {rabbit_boot_steps,run_step,2,
                [{file,"src/rabbit_boot_steps.erl"},{line,49}]},
            {rabbit_boot_steps,'-run_boot_steps/1-lc$^0/1-0-',1,
                [{file,"src/rabbit_boot_steps.erl"},{line,26}]},
            {rabbit_boot_steps,run_boot_steps,1,
                [{file,"src/rabbit_boot_steps.erl"},{line,26}]},
            {rabbit,start,2,[{file,"src/rabbit.erl"},{line,585}]},
            {application_master,start_it_old,4,
                [{file,"application_master.erl"},{line,273}]}]}}

If I add AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY and nothing else to the above options it works.

I checked awscli and I am able to describe instances and auto scaling groups.

Any ideas?

motilevy commented

Also just to make sure the policy is not the issue, i created a user and attached the same policy to it, and the cluster came up like a charm.

hrzbrg commented

Which Alpine/Erlang/RabbitMQ Version are you using? I rebuild the container with Alpine 3.5, Erlang 19.1 and RabbitMQ 3.6.6. At least now the IAM role is used to retrieve information about the autoscaling group.

motilevy commented

Alpine3.5 ( Installes erlang-19.1.0-r0 ) + Rabbit 3.6.6 worked like a charm.

  • 👍1