From 2ecb54263391bd947b4c2592a24576e1c55f1da1 Mon Sep 17 00:00:00 2001 From: Michael Barroco Date: Thu, 7 Dec 2023 16:51:29 -0500 Subject: [PATCH] Add README with debugging instructions --- deploy/operations/ci/aws-1/README.md | 39 ++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) create mode 100644 deploy/operations/ci/aws-1/README.md diff --git a/deploy/operations/ci/aws-1/README.md b/deploy/operations/ci/aws-1/README.md new file mode 100644 index 000000000..9e7b04976 --- /dev/null +++ b/deploy/operations/ci/aws-1/README.md @@ -0,0 +1,39 @@ +# AWS-1 CI deployment + +This module deploys a kubernetes cluster to AWS and provisions the dss using helm. + +## Terraform state + +The terraform backend is configured to be shared using a S3 bucket. (see [`main.tf`](./main.tf)). + +## Debugging + +In case of issue, it is possible to connect to the cluster and retrieve the terraform state to manage it +locally. + +### Connection to the cluster + +To connect to the cluster, authenticate yourself to the AWS account. +Run the following command to load the kubernetes config: +``` +aws eks --region us-east-1 update-kubeconfig --name dss-ci-aws-ue1 +``` +Call the kubernetes cluster using `kubectl` + +#### Add other roles + +Access to the cluster is managed using the config map `aws-auth`. +Its definition is managed by [`kubernetes_admin_access.tf`](./kubernetes_admin_access.tf). +Currently only the user who bootstrapped the cluster and the ones assuming +the administrator role (see [`local_variables.tf`](./local_variables.tf)) have access. + +### Run terraform locally + +In case of failure, a user with administrator role can take over the deployment by cloning this +repository and retrieving the current deployment state by running the following command: + +``` +terraform init +``` + +At this point, the user can replay or clean the deployment as if it was the CI runner. \ No newline at end of file