Skip to content

Latest commit

 

History

History
358 lines (228 loc) · 10.1 KB

01-functions-and-conditional-expressions.md

File metadata and controls

358 lines (228 loc) · 10.1 KB
Raw

Lab: Functions and Conditional Expressions

Help for the VSCode editor.

  1. Information only

  2. What value does floor(10.9) produce?

    Know that floor rounds_down_ to the nearest integer

    Reveal

    10

  3. What value does title("user-generated password file") produce?

    Know that title converts its input to Title Case

    Reveal

    User-Generated Password File

  4. Which type of variable does the function lookup work with?
    Reveal

    map

    We lookup a key in a map to find its value, and provide a default if the key does not exist.

  5. What type of variable is cloud_users?

    1. Navigate to /root/terraform-projects/project-sonic in the Explorer pane

    2. Inspect variables.tf to find the answer

      Reveal

      string

  6. This variable contains the names of the developers for project-sonic with the names separated by a :.
    Using this variable and the count meta-argument, create IAM users for all developers. Write the resource block in the main.tf file.

    Convert this variable from a string to a list. Do not change the variable defined in variables.tf.

    We have some work to do here, combining what has been learned in several previous course sections! We are going to draw upon the following

    • aws_iam_user resource from the Terraform with AWS section
    • count from the Working with Terraform section
    • A function that will split a string into a list on a given separator, which we may need to use more than once
    • A function that gives us the length of a list.


    • Documentation for functions
    • Documentation for aws_iam_user

      Hint: The resource created at the end of the IAM with Terraform lab is very similar to what is required here!

    1. Determine the function required to split a string given a separator

      Reveal

      split

      Therefore to get a list from the cloud_users variable, it would be

      split(":", var.cloud_users)

    2. Determine how we will set the count within the resource.

      We have the list as the expression defined above. Getting the length of this list will give us count for the number of users to create.

      Reveal 
      count = length(split(":", var.cloud_users))

    3. Determine how we will set each user's name as the count is iterated. Remember the index property of count

      Reveal

      We have to evaluate the split here as well

      name = split(":",var.cloud_users)[count.index]

    4. Put this all together into a resource block

      Reveal 
      resource "aws_iam_user" "cloud" {
    name = split(":",var.cloud_users)[count.index]
    count = length(split(":",var.cloud_users))
}

    5. Deploy

      cd /root/terraform-projects/project-sonic
terraform init
terraform plan
terraform apply
  7. What is the name of the IAM User that is created at the Index 6, of the IAM User at address aws_iam_user.cloud?

    There are multiple ways to do this, not least simply inspecting the output of the apply, but you have been asked to use the console

    Reveal 
    terraform console
aws_iam_user.cloud[6].name

    --- OR ---

    echo 'aws_iam_user.cloud[6].name' | terraform console

    braja

  8. Locate the index of the element called oni in the variable called sf.

    Find a function that return the index of an element in a list by value, and test it in the console.

    Reveal

    The index function does this.

    terraform console
index(var.sf, "oni")

    --- OR ---

    echo 'index(var.sf, "oni")' | terraform console

    7

  9. What type is the variable called media?

    Inspect variables.tf again to find the answer

    Reveal

    set(string)

  10. We have now, updated the main.tf in this configuration directory and added a new resource block to create a S3 bucket called sonic-media. Create an additional resource called upload_sonic_media to upload the files listed in the variable called media to this bucket.

    Use the following specifications

    • Use the for_each meta-argument to upload all the elements of the media variable.
    • bucket: Use reference expression to the bucket sonic-media.
    • source: Each element in the variable called media.
    • key: Should be the name of the files being uploaded (minus the path component). For example, eggman.jpg, shadow.jpg etc.
    • Do not alter the variables!


    Refer back to the lab on S3 to find the resource used to upload files. Also refer to the lab/lectures on `for_each`.

    Note also that the keys in the bucket need to be the filename only, so we will need another function to strip off the /media/ part of the pathname to get only the filename. There are two possible functions that can be used. One is to manipulate the string itself, and the other function we can use which is specifically for this purpose and would work if the paths weren't all /media/ can be found in the Filesystem Functions in the documentation.

    Reveal

    To upload files to buckets, we use aws_s3_object.

    Solving the requirements one at a time

    1. Iterate the media variable

      for_each = var.media

    2. Use reference expression for the bucket

      bucket = aws_s3_bucket.sonic_media.id

    3. source, i.e. the file to be uploaded is each element of media

      source = each.value

    4. Get the file name of the the source file. As mentioned, there's two ways

      1. String chopping to retain only the part of the string after /media/ which is character position 7 onwards. We can use substr to do this, setting the length argument to something big enough to get the longest string

        key =  substr(each.value, 7, 20)

      However that's messy and not safe.

      1. Use the basename function that knows how to extract the filename and extension from a file path

        key =  basename(each.value)

    5. Put it all together

      resource "aws_s3_object" "upload_sonic_media" {
    bucket = aws_s3_bucket.sonic_media.id
    key = basename(each.value)
    source = each.value
    for_each = var.media
}

    6. Deploy

      terraform plan
terraform apply

  11. Information only. Navigate to indicated directory in Explorer pane.

  12. What is the value of the variable called small?

    Inspect variables.tf

    Reveal

    t2.nano

  13. What is the current value for the variable called name?

    Inspect variables.tf

    Reveal

    This variable has no default. It is therefore

    undefined

  14. Create an EC2 Instance with the resource name mario_servers.

    Use the following specifications:

    • AMI: Use variable called ami.
    • Tags: Create a tag with key Name and value set to the variable called name.
    • Instance_type: Use a conditional expression so that - If the instance is created with a tag Name = "tiny", it should use the variable called small else the variable called large.
    • We will supply the variable called name using the -var command line flag.

    You will need to make use of a conditional to set the instance type. We do not know what the -var will be set to by the question's validation script! Remember that conditionals have the form:
    logical_expression ? value_when_true : value_when_false

    Reveal 
    resource "aws_instance" "mario_servers" {
    ami = var.ami
    instance_type = var.name == "tiny" ? var.small : var.large
    tags = {
        Name = var.name
    }
}

    Now run terraform init only. The validation will create the resource using the -var value of its choice.