| """Put ECS Instances in DRAINING state so that all ECS Tasks running on them are migrated to other Instances. | |
| Batches into chunks of 10 because of AWS api limitations (An error occurred InvalidParameterException when | |
| calling the UpdateContainerInstancesState operation: instanceIds can have at most 10 items) | |
| """ | |
| def put_container_instances_in_draining_state(ecs_client, cluster_name, container_instance_arns): | |
| batch_size = 10 | |
| n_batches = math.ceil(len(container_instance_arns)/batch_size) | |
| for i in range(0, len(container_instance_arns), batch_size): | |
| logger.info('Putting batch %d/%d of container instances %s in cluster %s into DRAINING state', i+1, n_batches, container_instance_arns, cluster_name) | |
| ecs_client.update_container_instances_state(cluster=cluster_name, containerInstances=container_instance_arns[i:i + batch_size], status='DRAINING') |
| #!/usr/bin/env python3 | |
| # | |
| # Script to replace EC2 instances in an ECS cluster's auto-scaling group after | |
| # changing the AMI or instance type in the launch configuration. It | |
| # checks for instances with the incorrect AMI or type, scales up the | |
| # auto-scaling group with replacement instances, then drains the tasks | |
| # from the old instances. | |
| # | |
| # Usage: aws-vault exec profile-name -- python3 replace_ecs_cluster_instances.py --group=asg-name --cluster=ecs-cluster-name --count=3 | |
| # |
| #!/bin/bash | |
| # deploy serverless stack | |
| serverless deploy \ | |
| --stage "dev" \ | |
| --region "us-east-1" \ | |
| --service-name "my-service-name" \ | |
| --slack-channel "#slack-alerts-example" \ | |
| --webhook-url "https://hooks.slack.com/services/12345/12345/abcdefg" \ | |
| --ecs-cluster-arn "arn:aws:ecs:us-east-1:123456:cluster/test-cluster" |
The standard way of understanding the HTTP protocol is via the request reply pattern. Each HTTP transaction consists of a finitely bounded HTTP request and a finitely bounded HTTP response.
However it's also possible for both parts of an HTTP 1.1 transaction to stream their possibly infinitely bounded data. The advantages is that the sender can send data that is beyond the sender's memory limit, and the receiver can act on