I have provided application configuration via consul's key-value store to the application containers running in ECS services.
The application reads its configuration from consul only once on start up.
When I need to change the configuration, how should I go about restarting the containers so that the application configuration is refreshed?
I am hoping to do this programmatically via the aws cli.
You don't restart containers. You can however stop the individual tasks, and ECS will respawn another instance of your task somewhere on the cluster.
Update:
As #Aidin mentioned, you can achieve it via the AWS CLI by forcing a new deployment like so:
aws ecs update-service \
--service <service name> \
--cluster <cluster name> \
--force-new-deployment \
[--profile guestapi-dev]
Note that this does not work on services with a CodeDeploy deployment controller.
Original answer:
I faced the same challenge, and what I did was follow this guide (using the old or new console depending on your service). I don't know if this can be done via the CLI, but it does actually "restart the service" in that it re-spawns new task(s) for your service and kills the old one(s).
In summary:
In the old console:
Find the service in the AWS console (ECS -> Cluster -> Service).
Click Update in the top right corner.
Check the ‘Force new deployment’ box.
Skip the other configurations and click Update Service.
In the new console:
Find the service in the AWS console (ECS -> Cluster -> Service).
Click Edit in the top right corner.
Expand Deployments options
Check the ‘Force new deployment’ box.
Click Update.
The service will re-deploy. You should be able to see the existing task(s) running, the new task(s) provision and lastly the old task(s) disappear.
this worked for me:
aws ecs list-tasks --cluster my-cluster-name | jq -r ".taskArns[]" | awk '{print "aws ecs stop-task --cluster my-cluster-name --task \""$0"\""}' | sh
Go to ECS dashboard. Just stop the running task from your ECS service from aws console. it'll spawn a new task and terminate the old one.
In conclusion, you cannot simply stop and start a container within the same task. You just start a new task. AWS should do a rolling bounce, so it will not give you a downtime, and new task will stay as long is is passing the health check
None of the two existing solutions on this question are satisfying. I don't have a full answer (yet), but I can A) tell you what I found, and B) tell you what is the "correct" architecture to handle this issue.
What I found
I was under the impression that SSHing into the instance and then simply docker restart <container-id> should work.
In fact, it initially seemed like it did. But, it turned out that I was wrong and it was just a can of worms waiting there for me! Doing so results in the container starting with no IAM role/credentials properly to talk to the other AWS services. My story in detail is on this Github issue of ecs-agent. It took me 10+ hours to find out that was the culprit. Apparently, containers will be in proper condition only if the ecs-agent starts them, and not you start/restart them.
What's the right way?
I believe the mentality and philosophy behind ECS/Tasks are that they want to take full control of the layer of abstraction between you and the running environment of the containers. You just say "Hey I want 3 of these user-avatar-uploader-to-s3 containers running" and it does that job for you. But you are not much welcome to meddle in the way they are doing their business!
However, if you want the containers to be configurable and pass certain params to it (e.g. the consul key-value pair in the original question), you are allowed to define them as Environment Variable both in the Task Definition (for each container) and in the Service/Task execution.
So, the right way would be to redo your container code to take these params (key-value pair) as Environment Variable (or from a configurable secure private S3 bucket, or AWS SecretsManager). Then put the desired values in the task/task-execution, and voila it should work. You can then change them at any time and ECS will take care of it. (Note that it will be a new container/task spinning up with the new settings, not your old one updated.)
That's it.
(I will update this answer as soon as I find how to do that emergency open-heart-surgery docker restarts.)
Related
I have running my cluster and task is running.
My need is want to update container image in running task in cluster how to do?
My Image is with latest tag and every time any new changes come will push to ecr on latest tag.
Deploying with the tag latest isn't a best practice because you loose a lot of visibility into what you are doing (e.g. scale out events where you deploy more tasks as part of a service will all end up using LATEST but will be effectively running different versions of the code, etc.).
This pontificating aside, you didn't say if you started your task(s) as standalone using the run-task API or if you started your task(s) as part of a service.
If the former, you need to stop your task and run it again. If the latter, you need to redeploy your service using the --force-new-deployment flag.
I know that there are already countless questions in this direction, but unfortunately I was not able to find the right answer yet. If a post already exists, please just share the link here.
I have several gitlab CI / CD pipelines. The first pipeline uses Terraform to build the complete infrastructure for an ECS cluster based on Fargate. The second / third pipeline creates nightly builds of the frontend and the backend and pushes the Docker Image with the tag "latest" into the ECR of the (staging) AWS account.
What I now want to achieve is that the corresponding ECS tasks are redeloyed so that the latest Docker images are used. I actually thought that there is a way to do this via CloudWatch Events or whatsoever, but I don't find a really good starting point here. A workaround would be to install the AWS CLI in the CI / CD pipeline and then do a service update with "force new deployment". But that doesn't seem very elegant to me. Is there any better way here?
Conditions:
The solution must be fully automated (either in AWS or in gitlab CI / CD)
Switching to AWS CodePipeline is out of discussion
Ideally as close as possible to AWS standards. I would like to avoid extensive lambda functions that perform numerous actions due to their maintainability.
Thanks a lot!
Ok, for everybody who is interested in an answer. I solved it that way:
I execute the following AWS CLI command in the CICD pipeline
aws ecs update-service --cluster <<cluster-name>> --service <<service-name>> --force-new-deployment --region <<region>>
Not the solution I was looking for but it works.
As a general comment it is not recommended to always push the same container tag because then rolling back to a previous version in case of failure becomes really difficult.
One suitable option would be to use git tags.
Let's say you are deploying version v0.0.1
You can create a file app-version.tf which will contain the variable backend-version = v0.0.1 that you can reference on the task definition of the ecs service.
Same thing can be done for the container creation using git describe.
So, you get a new task definition for every git tag and the possibility of rolling back just by changing a value in the terraform configuration.
It is beneficial to refer to images using either digests or unique immutable tags. After the pipeline pushes the image, it could:
Grab the image's digest/unique tag
Create a new revision of the task definition
Trigger an ECS deployment with the new task definition.
As sgramo93 mentions, the big benefit is that rolling back your application can be done by deploying an older revision of the task definition.
This is almost the same question as this one, but for Fargate.
I can't find any way to just stop the cluster or the Fargate service temporarily without having to delete it or changing its task definition.
Tried to stop each task individually but as expected, Fargate provisions a new task right after.
Seems there no option in the AWS console yet - maybe a CLI option exists?
Fargate does not allow you to stop the cluster because there are no underlying EC2 instances that you control to stop. Resources are provisioned in a "serverless" way so you don't have to deal with the underlying resources.
You need to stop the individual tasks but, like you reported, you may encounter that they are replaced after you stop the running tasks that are part of a service. To ensure this doesn't happen update your services to have a "Number of tasks" set to 0. This will keep your service definition up so you don't have to delete them but it will allow you to remove any running tasks.
Hope that helps!
Found a command in the ecs-cli that does exactly what #jd-d described:
ecs-cli compose --project-name name service down --cluster-config cluster --cluster cluster
Stops the running tasks that belong to the service created with the compose project. This command updates the desired count of the service to 0.
https://docs.aws.amazon.com/AmazonECS/latest/developerguide/cmd-ecs-cli-compose-service-stop.html
It does work! but unfortunately I think it is not a complete answer as seems it only works when using ecs-cli and to manage docker compose.
I am trying to migrate a set of microservices from Docker Swarm, to AWS ECS using Fargate.
I have created an ECS cluster. Moreover, I have initialized repositories using the ECR, each of which contains an image of a microservice.
I have successfully came up with a way to create new images, and push them into the ECR. In fact, with each change in the code, a new docker image is built, tagged, and pushed.
Moreover, I have created a task definition that is linked to a service. This task definition contains one container, and all the necessary information. Moreover, its service defines that the task will run in a VPC, and is linked to a load balancer, and has a target group. I am assuming that every new deployment uses the image with the "latest" tag.
So far with what I have explained, everything is clear and is working well.
Below is the part that is confusing me. After every new build, I would like to update the service in order for new tasks with the update image get deployed. I am using the cli to do so with the following command:
aws ecs update-service --cluster <cluster-name> --service <service-name>
Typically, after performing the command, I am monitoring the deployment logs, under the event tab, and checking the state of the service using the following command:
aws ecs describe-services --cluster <cluster-name> --service <service-name>
Finally, I tried to simulate a case where the newly created image contains a bad code. Thus, the new tasks will not be able to get deployed. What I have witnessed is that Fargate will keep trying (without stopping) to deploy the new tasks. Moreover, aside the event logs, the describe-services command does not contain relevant information, other than what Fargate is doing (e.g., registering/deregistering tasks). I am surprised that I could not find any mechanism that instructs Fargate, or the service to stop the deployment and rollback to the already existing one.
I found this article (https://aws.amazon.com/blogs/compute/automating-rollback-of-failed-amazon-ecs-deployments/ ), which provides a solution. However, it is a fairly complicated one, and assumes that each new deployment is triggered by a new task definition, which is not what I want.
Therefore, considering what I have described above, I hope you can answer the following questions:
1) Using CLI commands (For automation purposes) Is there a way to instruct Fargate to automatically stop the current deployment, after failing to deploy new tasks after a few tries?
2) Using the CLI commands, is there a way to monitor the current status of the deployment? For instance, when performing a service update on a service on Docker swarm, the terminal generates live logs on the update process
3) After a failed deployment, is there a way for Fargate to signal an error code, or flag, or message?
At the moment, ECS does not offer deployment status directly. Once you issue a deployment, there is no way to determine its status other than to continually poll for updates until you have enough information to infer from them. Plus unexpected container exits are not logged anywhere. You have to search through failed tasks. The way I get them is by cloudwatch rule that triggers a lambda upon task state change.
I recommend you read: https://medium.com/#aaron.kaz.music/monitoring-the-health-of-ecs-service-deployments-baeea41ae737
As of now, you have a way to do this:
aws ecs wait services-stable --cluster MyCluster --services MyService
The previous example pauses and continues only after it can confirm that the service running on the cluster is stable. Will return 255 exit code after 40 failed checks.
To cancel a deployment, enable ECS Circuit Breaker when creating your service:
aws ecs create-service \
--service-name MyService \
--deployment-configuration "deploymentCircuitBreaker={enable=true,rollback=true}" \
{...}
References:
Service deployment check.
Circuit Breaker
I am interested in running builds/scripts on demand, inside a Docker container, on AWS or GCP.
I keep reading of the ECS service (https://aws.amazon.com/ecs/) , but I am not sure it is what I need. I surely do not need a cluster of managed EC2 instances. I also don't think Google Container Engine is the answer.
I just need to start a Docker container, run a build or any script inside it and shut it down. The lifetime of the container would be max 1h. So it's not about long running, or scaling any application. Just start, run, stop a Docker container on demand. Which AWS or GCP service is most suitable for this requirement?
Besides the service, what HTTP endpoints of it do I need to call in order to automate this process?
My application receives some bash script from the user and has to fire up a container, run the script, shut everything down when it's finished or errored and come back with the script's output. I imagine it would connect to the created/runnig instance via SSH.
Any help or hint to the proper documentation is appreciated, thanks!