Redis Update Assessment

[laineymajor]

Problem Statement Our current development environment relies on Redis in Amazon Elasticache, specifically version 3. However, this version has become outdated and poses potential security and performance risks. To ensure the continued reliability and security of our development environment, we need to upgrade our Redis instance to version 6.2 or a higher compatible version.

Action Items Conduct a thorough assessment of the existing Redis configuration and dependencies within our development environment.

[laineymajor]

We are upgrading (hopefully) a shorter/simpler upgrade path! Fingers crossed. Chris continuing to research.

[flooose]

The update to redis 6 or 7 (see below) should be relatively straight-forward because backward compatibility has always been a design principle of the Redis project. The following points need to be addressed before moving to the actual upgrade.

• [x] do we need to do an incremental update or is jump to 7.x fine
  • advantages of incremental
    1. we get two versions with which to test automation
  • advantages of direct jump
    1. we're done with one go
  • [x] is direct upgrade even an option?
    • I suspect, yes
      • we're only going to 6.2, so this is irrelevant.
• [x] how do we upgrade with terraform?
  • [x] we should be able to test this locally with terraform plan
    • yes
  • We still have to figure out how get the terraforms to run on M1
    • this got resolved, but we can't use terraform for upgrades
• [x] Assess aws-cli for back-ups
  • backups are not a problem. Could look approximately like this

    aws elasticache create-snapshot \
        --cache-cluster-id dsva-vagov-dev-sidekiq-002 \
        --snapshot-name sidekiq-dev-$(date +'%Y%m%d%H%M%S')`

  • [x] still needs to be assessed because it involves creating a new cluster with the backup.

Action Items

taken from main story https://app.zenhub.com/workspaces/platform-product-team-633af4074573d06c3cda142a/issues/gh/department-of-veterans-affairs/va.gov-team/64051

• [ ] Conduct a thorough assessment of the existing Redis configuration and dependencies within our development environment.
  • [ ] Since this is a managed redis instance, the only redis config we should have to validate is redis.yml in the vets-api repo. @LindseyHattamer do you see this the same way?
• [x] Research and evaluate compatibility and potential impacts of upgrading to Redis version 6.2 or higher.
  • The update to redis 6 or 7 (see below) should be relatively straight-forward because backward compatibility has always been a design principle of the Redis project.
• [x] Create a comprehensive upgrade plan, including a rollback strategy in case of unforeseen issues.
• [ ] Schedule a maintenance window for the upgrade to minimize disruption to development activities.
• [ ] Implement the Redis upgrade as per the plan.
• [ ] Thoroughly test the upgraded environment to ensure it functions as expected.
• [ ] Update documentation to reflect the new Redis version and any changes to configurations.
• [ ] Notify relevant stakeholders, including the development team, of the upgrade schedule and potential impacts on their work.

[flooose]

This is the documentation https://vfs.atlassian.net/wiki/spaces/TT1/pages/2806808593/Redis+Update

[laineymajor]

Seems like this can be closed since we have the proper documentation... correct @flooose ?

[flooose]

I'm of the opinion that it can be closed. Is it worth getting a second developer opinion on this though? Maybe there's something I failed to consider? @Kshitiz-devops or @LindseySaari maybe?

[laineymajor]

duplicate https://app.zenhub.com/workspaces/platform-product-team-633af4074573d06c3cda142a/issues/gh/department-of-veterans-affairs/va.gov-team/53708