We’re using Gitlab as our main DevOps tool, and the infrastructure is on AWS ECS cluster (all Gitlab pipelines are pushing to the cluster instances).
Now we are planning to migrate from AWS ECS to Kubernetes (using K8s Kops on AWS, not EKS), and would like to know the IMPACT on the pipelines (how we can implement the migration with the minimal downtime on the CI process).
HI did you find out about gitlab downtime during migration from omnibus to k8s?
moving object storage from local storage to s3 doeas it require downtime?
we have arounf 300GB of uploads,artifacts, if we change the storage to s3 and run ```
gitlab-rake gitlab:artifacts:migrate - doeas this cause downtime , I did not find any instructions regarding puting offline migrated instances etc?
DO you have any answers to that?