After getting started with Terraform, this intended to introduce many concepts on a very basic level. The most noticeable thing about tf-workshop is there are design/implementation oddities almost everywhere - yet it still works.
This is a training start-point; however, the end-goal is a hidden easter egg somewhere in this repo. You'll have to follow all the instructions to find it.
- A POSIX laptop; a workstation will do though.
- A Github account.
- AWS Account Credentials
After that, follow some initial account security setup steps.
Fork this repo, clone it down and cd into the directory:
git clone git@github.com:myGithubUser/tf-workshop.git && cd tf-workshop/This is a convenience wrapper to speed-up apply, plan and destroy operations; it also removes and cleans-up nicely to reset the environment. These are the usage details:
Description: Build or Destroy systems at will.
Usage: ./test_systems.sh [OPTION1] [OPTION2]...
Example: ./test_systems.sh -v --ip-check
OPTIONS:
-t, --test 'terraform plan' operation
Initialize and plan a potential terraform build.
Example: ./test_systems.sh -t
-b, --build 'terraform apply' operation
Build a new system from the *.tf files in the current directory.
Example: ./test_systems.sh --build
-d, --destroy 'terraform destroy' operation
Destroy a system corresponding to the state files in the
current directory.
Example: ./test_systems.sh -v -d
-c, --cleanup After a glitch and manual intervention is required, reset
the state and remove the log file from the last run.
Example: ./test_systems.sh --cleanup
-ip, --ip-check Verify the current gateway address
Example: ./test_systems.sh -ip
-r, --retest Quickly rebuild a system; calls destroy, then builds again.
Example: ./test_systems.sh -v --retest
-v, --verbose Turn on 'set -x' debug output.
Initialization and planning are simple:
./test_systems.sh --test
If the plan looks good, then build it:
./test_systems.sh --build
At the end, a pub_ip will be output. Login to your system or just test that nginx is running:
curl -4 http://$pub_ip
NOTE: it does take a moment for the instance to become available. Be patient and retry if it doesn't work immediately.
When done with the inspection, log out of the instance:
admin@52.11.112.251:~$ exit
logout
Connection to 52.11.112.251 closed.
myHost:tf-workshop user$ The instance can be destroyed easily when you're done:
./test_systems.sh --destroy
So there are no lingering state files lying about, clean them up as well:
./test_systems.sh --cleanup
After that, it's time to make improvements and submit pull requests. This thing is horrible.
This is all well and good but we're still dragging the habits of the old data center into the modern cloud. To remedy that, check back to see The modern approach to deployments.
The modern approach will decrease complexity, increase security, and make (normally) complex deployments easier than you might think. Stay tuned...
