I have a python3 project which runs in a docker container environtment.
My Python project uses AWS Acces keys and secret but using a credentials file stored in the computer which is added to the container using ADD.
I deployed my project to EC2. The server has one task running which works fine. I am able to go through port 8080 to the webserver (Airflow).
When I do a new commit and push to a master branch in github, the hook download the content and deploy it without build stage.
The new code is in the EC2 server, I check it using ssh but the container that runs in the task get "stuck" and the bind volumes dissapear and they are not working until I restart a new task. The volumes are applied again from 0, and those reference to the new code. This action is fully manual.
Then, to fix it I listen about AWS ECS Blue/Green deployment, so I implemented it. In this case the Codepipeline add a build stage, but here starts the problem. If in the build I try to push a docker image to the ECR, which my task definition makes reference it fails. It fails because in the server, and in the repo (which I commit push my new code) there is no credentials file.
I tryed doing the latest docker image from my localhost, and avoiding build stage in codepipeline, and it works fine, but then when I go to the 8080 port in both working ip's I am able to get in the webserver, but the code is not there. If i click anywhere it says code not found.
So, in a general review I would like to understand what i am doing wrong, and how to fix in a general guidelines, and in the other hand ask why my EC2 instance in the AWS ECS Blue/Green cluster has 3 ip's.
The first one is the one that I use to reach server through port 22. And if there I run docker ps I see one or two containers running depending if I am in the middle of a new deployment. If here I search my new code its not here...
The other two ip's are changing after every deployment (I guess its blue and green) and both work fine until Codepipeline destroys the green one (5 minutes wait time), but the code is not there. I know it because when I click in any of the links in the webserver it fails saying the Airflow Dag hasn't been found.
So my problem is that I have a fully working AWS ECS Blue/Green deployment but without my code. Then my webserver doesn't have anything to run.
Related
I have a set of AWS Elastic beanstalk using Docker based configuration for both web server and worker server. The way we have setup is that the java process inside docker allocates 70% of the box memory when starting.
Now the first deployment works fine, but when I try to update application version with in-place Rolling update, Elastic beanstalk tries to start an additional docker container with the java process before stopping the existing one. This fails the deploy as the Java server is not able to allocate the required memory. Is there a way that I can setup AWS to kill the old docker instance before starting the new one during deployment?
I even tried Rolling with additional batch, but that one only works for the first batch and then fails for subsequent ones.
immutable updates can be the way to go for you, it basically recreates the EC2 instances completely on every deploy
Open the Elastic Beanstalk console.
Navigate to the management page for your environment.
Choose Configuration.
In the Rolling updates and deployments configuration category,
choose Modify.
Select immutable on deploy policy
Apply
you can check more on how it works here
I have created a docker image for DRUID and Superset, now I want to push these images to ECR. and start an ECS to run these containers. What I have done is I have created the images by running docker-compose up on my YML file. Now when I type docker image ls i can see multiple images running in them.
I have created an aws account and created a repository. They have provided the push command and I push the superset into the ECR for start. (Didn't push any dependancy)
I created a cluster in AWS, in one configuration step if provided custom port 8088. I don't know what and why they ask these port for.
Then I created a load balancer with the default configuration
After some time I could see the container status turned running
I navigated to the public ip i mentioned with port 8088 and could see superset running
Now I have two problems
It always shows login error in a superset
It stops automatically after some time and restarts after that and this cycle continues.
Should I create different ECR repos and push all the dependencies to ECR before creating a cluster in ECS?
For the service going up and down. Since you mentioned you have an LB associated with the service, you may have an issue with the health check configuration.
If the health check fails consecutively a number of times, ecs will kill it and re-start it.
I have a basic django/postgres app running locally, based on the Docker Django docs. It uses docker compose to run the containers locally.
I'd like to run this app on Amazon Web Services (AWS), and to deploy it using the command line, not the AWS console.
My Attempt
When I tried this, I ended up with:
this yml config for ecs-cli
these notes on how I deployed from the command line.
Note: I was trying to fire up the Python dev server in a janky way, hoping that would work before I added nginx. The cluster (RDS+server) would come up, but then the instances would die right away.
Issues I Then Failed to Solve
I realized over the course of this:
the setup needs another container for a web server (nginx) to run on AWS (like this blog post, but the tutorial uses the AWS Console, which I wanted to avoid)
ecs-cli uses a different syntax for yml/json config than docker-compose, so you need to have some separate/similar code from your local docker.yml (and I'm not sure if my file above was correct)
Question
So, what ecs-cli commands and config do I use to deploy the app, or am I going about things all wrong?
Feel free to say I'm doing it all wrong. I could also use Elastic Beanstalk - the tutorials on this don't seem to use docker/docker-compose, but seem easier overall (at least well documented).
I'd like to understand why any given approach is a good way to do this.
One alternative you may wish to consider in lieu of ECS, if you just want to get it up in the amazon cloud, is to make use of docker-machine using the amazonec2 driver.
When executing the docker-compose, just ensure the remote Amazon host machine is ACTIVE which can be viewed with a docker-machine ls
One item you will have to revisit with the Amazon Mmgt Console is to open the applicable PORTS such as Port 80 and any other ports exposed in the compose file. Once the security group is in place for the VPC, you should be able to simply refer to the VPC ID on subsequent executions bypassing any need to use the Mgmt console to add the ports. You may wish to bump up the instance size from the default t2.micro to match the t2.medium specified in your NOTES.
If ECS orchestration is needed, then a task definition will need to be created containing the container definitions you require as defined in your docker compose file. My recommendation would be to take advantage of the Mgmt console to construct the definition and then grab the accompanying JSON defintion which is made available and store in your source code repository for future executions on the command line where they can be referenced in registering task definitions, executing tasks and services within a given cluster.
I am running an elasticbeanstalk application, with multiple environments. This particular application is hosting docker containers which host a webservice.
To upload and deploy a new version of the application to one of the environments, I can go through the web client and click on "Upload and Deploy" and from the file option I select my latest Dockerrun.aws.json file, which references the latest version of the container that is privately hosted. The upload and deploy works fine and without issue.
To make it simpler for myself and others to deploy I'd like to be able to use the CLI to upload and deploy the Dockerrun.aws.json file. If I use the cli eb deploy command without any special configuration the normal process of zipping up the whole application and sending it to the host occurs and fails (it cannot reason out that it only needs to read the Dockerrun.aws.json file).
I found a documentation tidbit about controlling what is uploaded using the .elasticbeanstalk/config.yml file.
Using this syntax:
deploy:
artifact: Dockerrun.aws.json
The file is uploaded and actually deploys successfully to the first batch of instances, and then always fails to deploy to the second set of instances.
The failure error is of the flavor: 'container exited unexpectedly...'
Can anyone explain, or provide link to the canonical approach for using the CLI to deploy single docker container applications?
So it turns out that the method I listed about with the config.yml was correct. The reason I was seeing a partially successful deployment was because the previously running docker container on the hosts was not being stopped by EB.
I think that what was happening was that EB is sending something like
sudo docker kill --signal=SIGTERM $CONTAINER_ID instead of the more common sudo docker stop $CONTAINER_ID
The specific container I was running didn't respond to SIGTERM and so it would just sit there. When I tested it locally with SIGKILL it would (obviously) stop properly, but SIGTERM alone wouldn't stop it.
The issue wasn't the deployment methodology but rather confusion in the output that EB generated and my misinterpretation.
Since you have asked for a link, I am providing a link which I initially used to successfully test and deploy docker using elasticbeanstalk cli.
Kindly see if this helps you as well: https://fangpenlin.com/posts/2014/11/25/running-docker-with-aws-elastic-beanstalk/
I have configured an aws asg using ansible to provision new instances and then install the codedeploy agent via "user_data" script in a similar fashion as suggested in this question:
Can I use AWS code Deploy for pulling application code while autoscaling?
CodeDeploy works fine and I can install my application onto the asg once it has been created. When new instances are triggered in the ASG via one of my rules (e.g. high cpu usage), the codedeploy agent is installed correctly. The problem is, CodeDeploy does not install the application on these new instances. I suspect it is trying to run before the user_data script has finished. Has anyone else encountered this problem? Or know how to get CodeDeploy to automatically deploy the application to new instances which are spawned as part of the ASG?
AutoScaling tells CodeDeploy to start the deployment before the user data is started. To get around this CodeDeploy gives the instance up to an hour to start polling for commands for the first lifecycle event instead of 5 minutes.
Since you are having problems with automatic deployments but not manual ones and assuming that you didn't make any manual changes to your instances you forgot about, there is most likely a dependency specific to your deployment that's not available yet at the time the instance launches.
Try listing out all the things that your deployment needs to succeed and make sure that each of those is available before you install the host agent. If you can log onto the instance fast enough (before AutoScaling terminates the instance), you can try and grab the host agent logs and your application's logs to find out where the deployment is failing.
If you think the host agent is failing to install entirely, make sure you have Ruby2.0 installed. It should be there by default on AmazonLinux, but Ubuntu and RHEL need to have it installed as part of the user data before you can install the host agent. There is an installer log in /tmp that you can check for problems in the initial install (again you have to be quick to grab the log before the instance terminates).