Skip to content

Instantly share code, notes, and snippets.

@MrHassanMurtaza
Forked from j-mprabhakaran/AWSDevOpsExamTips
Created August 28, 2019 21:31
Show Gist options
  • Save MrHassanMurtaza/ee586352bf3599c72715da4426a901f2 to your computer and use it in GitHub Desktop.
Save MrHassanMurtaza/ee586352bf3599c72715da4426a901f2 to your computer and use it in GitHub Desktop.
AWSDevOpsExamTips
Autoscaling.
◾Understand autoscaling inside and out.
◾Make sure you know the differences between deploying with and without Cloudformation (as there are some deployment methods you can do with Cloudformation that you can’t without) and the command or JSON syntax to perform them.
◾Understand lifecycle hooks and the autoscaling stages they run in
◾Understand launch configurations and how they work and are updated (hint: You must replace them)
◾Understand why you may need to place an instance into STANDBY state
•Elastic Beanstalk.
◾You’ll need to know this to an advanced level.
◾Understand what stacks Beanstalk supports natively and how you would deploy a stack it doesn’t (hint: Use Docker).
◾Understand the differences between a single Docker deployment v’s multiple Docker deployments.
◾Understand how .ebextensions work and the syntax of each of the files and how you’d make custom modifications to your beanstalk container.
◾Make sure you know the “eb” command line and can spot “fake” commands
•Opsworks
◾As with Elastic Beanstalk you’ll need to know this to a solid depth
◾Understand Opsworks concepts (stacks/layers/applications) and the properties of each
◾Understand that Opsworks uses chef and make sure you know the different deployment lifecycle states and when they are each run
◾Understand the differences between Windows and Linux stacks
◾Understand how Opsworks actually works and the different style (and limitations) of autoscaling it uses
◾Understand Opsworks monitoring and how this differs slightly from native Cloudwatch metrics for EC2 instances
•Deployment Methodologies◾You’ll need to understand the different deployment methods for blue/green and the pro/cons of each. Simply watch this video and understand it. Its all you’ll need: Deep Dive into Blue/Green Deployments on AWS
•Cloudformation
◾You won’t need to memorise every single command in Cloudformation but ensure you understand the different sections of the templates and the syntax
◾Understand fully cloudformation::init and cfn-init
◾Understand Creationpolicy and cfn-signal
◾Understand Waitcondition and handles and the difference between when to use these v’s Creationpolicy
◾Understand how to use Cloudformation with auto-scaling deployments (memorise the commands for this section)
◾Know what each of the fn::* commands is used for within a template
◾Understand stack policies, when to use them and what they do
•Cloudwatch
◾Know what Cloudwatch is used for and how it works
◾Make sure you know the difference between a namespace, metric and a dimension and how they relate to each other
◾Understand Cloudwatch logs. Specifically know what the following are: log event, log stream, log group, metric filter, retention settings
◾Remember that Cloudwatch only keeps data for 14 days but Cloudwatch logs keeps data indefinitely (by default)
◾Understand how you can set up alerting within Cloudwatch and which endpoints you can send them to
•Cloudtrail
◾Make sure you know what Cloudtrail is, what it can be used for and how it can be setup (including how to configure it globally)
◾Understand the default encryption for the logs in S3 (SSE-S3) and that they can use KMS if desired and why you may want to do this
◾Know hash verification, integration with SNS and Cloudwatch logs and how to set each of those up
◾Realise you can have up to 5 trails per region
For example on the CloudFormation. I built templates from scratch. Like create an autoscaling group and launch configuration.
Ran it, updated it with things like wait handlers, Update Policies, Custom Lambda functions, parameters, added nesting etc…
This was key for the exam. Know what really happens.
I’m my opinion you have to learn it by doing it. This is not a memorization test.
I also created my own OpsWorks and ElasticBeanstalk sample configurations and applications.
Created my own chef recipes, with Berkshelf, loaded them in Git and learned the hard way what works and what doesn’t.
Created my own docker application and deployed then redeployed with OpsWorks.Cloned swapped everything in the process to see what
would happen.
This way you have real world experience and won’t get tied up on some tricky wording in the test.
One other thing and I stress this the most. Do everything from the cli. The GUI is nice but if you want to know exactly whats
required do it from the CLI. For example:
Create Launch Configuration
aws autoscaling create-launch-configuration --launch-configuration-name my-lc --image-id ami-e13739f6 --instance-type t2.micro
Create Autoscaling Group
aws autoscaling create-auto-scaling-group --auto-scaling-group-name my-asg --launch-configuration-name my-lc --max-size 5
--min-size 1 --availability-zones "us-east-1a"
This is exactly what the guys did in the lessons but I took it to the next step by running every command I could find for every
major technology in the material.
65% of the test is on CI/CD and supporting concepts:
Autoscaling, Launch Configurations, CloudFormation, Opsworks, and Elastic Beanstalk.
You need to know the dozens of different scenarios and how to deploy these technologies.
Although the other content was helpful I spent most of my time on the 65% concepts.
I burned too much time studying deep dives on things DynamoDB, EBS. Really good things to know but I had one question related to
these technologies.
Also I had at least a dozen question related to Data Pipeline. No clue what that was. I think it was predecessor to Kinesis.
This needs to be addressed in the lessons.
============================================================
1. Know the following services inside and out. In addition to watching the lectures, I read through the user guides and followed along with the tutorials. It's a good idea to practice in the console AND the CLI.
•CloudFormation
•OpsWorks
•Elastic Beanstalk
•Auto Scaling
•Elastic Load Balancing
•CloudWatch
•CloudWatch Logs
2. Be familiar with the following services. Data Pipeline and CloudSearch aren't discussed in the course, so be sure to cover them yourself.
•CloudTrail
•Data Pipeline
•Kinesis
•CloudSearch
•SWF
•SQS
3. You need general knowledge of the following services. Hopefully you know these from the associate exams, but it doesn't hurt to review.
•EC2
•S3
•IAM
4. Watch the DynamoDB and Delegation & Federation lectures once, but don't spend additional time on them. There are limited questions on these and you don't need in-depth knowledge to answer correctly.
5. Read the following whitepapers.
•Blue/Green Deployments: https://d0.awsstatic.com/whitepapers/AWSBlueGreen_Deployments.pdf
•Strategies for Managing Access: https://d0.awsstatic.com/whitepapers/strategies-for-managing-access-to-aws-resources-in-aws-marketplace.pdf
=====================================================================
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment