Reduce devnet ECS cpu and memory to save cost & fix a bug in travis where aws is not installed

.travis.yml

@@ -125,8 +125,6 @@ jobs:
       - docker --version  # document the version travis is using
       - docker build -t xdc-devnet -f cicd/devnet/Dockerfile .
     script:
-      - pip install --user awscli # install aws cli w/o sudo
-      - export PATH=$PATH:$HOME/.local/bin # put aws in the path
       - docker tag xdc-devnet:latest xinfinorg/devnet:latest # Always push to the latest
       - docker push xinfinorg/devnet:latest
 
@@ -140,6 +138,8 @@ jobs:
         - unzip terraform_"$tf_version"_linux_amd64.zip
         - sudo mv terraform /usr/local/bin/
         - rm terraform_"$tf_version"_linux_amd64.zip
+        - pip install --user awscli # install aws cli w/o sudo
+        - export PATH=$PATH:$HOME/.local/bin # put aws in the path
     script:
       - echo "Merge detected, executing changes(Devnet)"
       - cd cicd/devnet/terraform

cicd/devnet/terraform/ecs.tf

@@ -21,8 +21,12 @@ resource "aws_ecs_task_definition" "devnet_task_definition_group" {
   execution_role_arn = aws_iam_role.devnet_xdc_ecs_tasks_execution_role.arn
   task_role_arn = aws_iam_role.devnet_xdc_ecs_tasks_execution_role.arn
   
-  cpu = 1024
-  memory = 2048
+  # New nodes will consume a lot more CPU usage than existing nodes. 
+  # This is due to sync is resource heavy. Recommending set to below if doing sync:
+  # CPU = 2048, Memory = 4096
+  # Please set it back to cpu 512 and memory of 1024 after sync is done to save the cost
+  cpu = 512
+  memory = 1024
   volume {
     name = "efs"