aws.ec2 | AWS EC2 Client Package | AWS library

I think I got this to work now - by using a different module.

I guess this module only works with the Classic LB (or something like that).

By using the community.aws.elb_target module I got it to work with the AWS ALB we were using. Sorry for creating confusion. Should probably adjust title

This is a limitation of CloudTrail, so at this time it's not be possible to pass that information from CloudTrail if it exceeds 100KB.

Potential work-around that may be useful to others with this message is to create an EventBridge rule to track EC2 instance state changes. So instead of monitoring the api call runinstances look for instances changing into the state running triggering from that as this should have a smaller response.

https://docs.aws.amazon.com/AmazonCloudWatch/latest/events/CloudWatch-Events-tutorial-CloudWatch-Logs.html

ec2_instance returns a list of instances, in this case you have just one instance. Try as below:

I fixed this by adding the following.

Add string parameter under This project is parameterized as shown below.

Next under Generic Webhook Trigger I have added Header parameters

Now, we can directly pass this param in the build command using $param.

The curl command is now modified to below.

It depends on the types of S3 events you are trying to trigger from.

The link you provided shows that you can use CloudTrail events to trigger CloudWatch event rules. This would allow you to trigger on events such as creating/removing buckets.

If you want to trigger off of things like objects being added/deleted from a bucket, then you will not be able to do it without writing some additional resources. You can create S3 events to trigger SQS, SNS or a Lambda function. My suggestion would be to implement a lambda function that accepts an S3 event and kicks off your state machine as desired. Then create an S3 event to trigger your lambda when the desired object event occurs.

I have just encountered your situation:

As Chris Williams mentioned in a reply to your question, you cannot use the public IP address when binding the listening socket on the EC2 instance. Change that to use the private IP and it should work.

Also make sure you enable traffic for the port you are binding from AWS security groups.

I'd suggest to use Auto Scaling groups instead of EC2 Fleet, since balancing instances across AZs is a key requirement. Using AWS SDK this should be a straightforward change.

The other thing I'd suggest, is to use Spot instances instead of On-Demand or mix of both to optimize the cost. This is possible through the MixedInstancesPolicy property, and as a best practice you can use multiple instance types and capacity-optimized as SpotAllocationStrategy to allow ASG to pick the capacity from the most available instance pools.

Also ASG will be able to keep number of instance at the desired capacity by replacing any interrupted Spot instances.

Instead of making subnet_list as a list of dictionaries, just make it a list of strings. This will make your life easier.

There were 2 issues as discovered through the comments.

The first was that the RunInstances task was not including the MinCount and MaxCount properties which led to no instances being launched.

Once this was fixed the next issue was a permissions issue due to the lack of permissions to run ec2:RunInstance or e2:CreateTags.

It is worth stating the best practice with permissions is to scope down to the minimal permissions that you require to successfully run.