Skip to content

Commit

Permalink
Merge pull request #5092 from ministryofjustice/update-node-upgrades
Browse files Browse the repository at this point in the history
fix: 🐛 broken page
  • Loading branch information
jaskaransarkaria authored Dec 14, 2023
2 parents 4408fe9 + 71ee0e0 commit e52c447
Showing 1 changed file with 13 additions and 12 deletions.
25 changes: 13 additions & 12 deletions runbooks/source/node-group-changes.html.md.erb
Original file line number Diff line number Diff line change
@@ -1,33 +1,34 @@
---
title: Handling Node Group and Instance Changes
weight: 54
last_reviewed_on: 2023-13-12
last_reviewed_on: 2023-12-13
review_in: 6 months
---

# <%= current_page.data.title %>
# Making changes to EKS node groups or instances types

## Why?

You may need to make a change to an EKS [cluster-node-group] or [instance-type-config]. We can't just let terraform apply these changes because terraform doesn't gracefully rollout the old and new nodes. Terraform will bring down all of the old nodes immediately, which will cause outages to users.
You may need to make a change to an EKS [cluster node group] or [instance type config]. We can't just let terraform apply these changes because terraform doesn't gracefully rollout the old and new nodes. Terraform will bring down all of the old nodes immediately, which will cause outages to users.

## How?

The method to avoid bringing down all the nodes at once is to follow these steps:

1. add a new node group with your [updated-changes]
1. add a new node group with your [updated changes]
1. lookup the old node group name (you can find this in the aws gui)
1. once merged in you can drain the old node group using the following command:
> cloud-platform pipeline cordon-and-drain --cluster-name <cluster_name> --node-group <old_node_group_name>
1. raise a new [pr-deleting] the old node group
1. once merged in you can drain the old node group using the command below:
1. raise a new [pr deleting] the old node group

> ```cloud-platform pipeline cordon-and-drain --cluster-name <cluster_name> --node-group <old_node_group_name>```

notes:

- When making changes to the default node group in live, it's handy to pause the pipelines for each of our environments for the duration of the change.
- the `cloud-platform pipeline` command [cordons-and-drains-nodes] in a given node group waiting 2mins between each drained node. This command runs remotely in concourse.

[cluster-node-group]: https://github.com/ministryofjustice/cloud-platform-infrastructure/blob/97768bfd8b4e25df6f415035acac60cf531d88c1/terraform/aws-accounts/cloud-platform-aws/vpc/eks/cluster.tf#L60
[instance-type-config]: https://github.com/ministryofjustice/cloud-platform-infrastructure/blob/97768bfd8b4e25df6f415035acac60cf531d88c1/terraform/aws-accounts/cloud-platform-aws/vpc/eks/cluster.tf#L43
[pr-deleting]: https://github.com/ministryofjustice/cloud-platform-infrastructure/pull/2663
[updated-changes]: https://github.com/ministryofjustice/cloud-platform-infrastructure/pull/2657
[cordons-and-drains-nodes]: https://github.com/ministryofjustice/cloud-platform-terraform-concourse/blob/main/pipelines/manager/main/cordon-and-drain-nodes.yaml
[cluster node group]: https://github.com/ministryofjustice/cloud-platform-infrastructure/blob/97768bfd8b4e25df6f415035acac60cf531d88c1/terraform/aws-accounts/cloud-platform-aws/vpc/eks/cluster.tf#L60
[instance type config]: https://github.com/ministryofjustice/cloud-platform-infrastructure/blob/97768bfd8b4e25df6f415035acac60cf531d88c1/terraform/aws-accounts/cloud-platform-aws/vpc/eks/cluster.tf#L43
[pr deleting]: https://github.com/ministryofjustice/cloud-platform-infrastructure/pull/2663
[updated changes]: https://github.com/ministryofjustice/cloud-platform-infrastructure/pull/2657
[cordons and drains nodes]: https://github.com/ministryofjustice/cloud-platform-terraform-concourse/blob/main/pipelines/manager/main/cordon-and-drain-nodes.yaml

0 comments on commit e52c447

Please sign in to comment.