-
Moving Thanos from GCS to Huawei OBS, and fixing a multipart upload bug on the way
What broke when Thanos blocks started landing in Huawei Cloud OBS instead of Google Cloud Storage, and the objstore patch that fixed it.
-
One Go API over every SCM: go-scm in the internal developer platform
How the IDP supports Bitbucket Server, Bitbucket Cloud, and GitHub through jenkins-x/go-scm, and the Stash fixes I sent upstream.
-
One timezone for everything: k8tz and the Jakarta timestamp problem
How we used k8tz to force Asia/Jakarta onto public images like Fluentd and Elasticsearch, and the three patches I sent upstream.
-
Cutting over-provisioning with KEDA: cron, external triggers, and a Selenium Grid queue
How I used KEDA cron scalers, external triggers, and Selenium Grid queue scalers to stop paying for idle replicas in production.
-
Running a k0s cluster with the GCP cloud controller for internal tooling
Why I picked k0s for an internal utility cluster, and what wiring the GCP cloud controller manager into it actually involved.