Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://github.com/kubernetes/k8s.io/issues/1307 below:

Migrate k8s-gubernator:builds dataset to wg-k8s-infra · Issue #1307 · kubernetes/k8s.io · GitHub

Part of migrating away from gcp-project k8s-gubernator: #1308

This dataset: https://console.cloud.google.com/bigquery?p=k8s-gubernator

It's used by:

• kettle, which writes to it (migration ref: Migrate kettle to k8s-infra #787)
• go.k8s.io/triage, which uses results from queries against it (migration ref: Migrate triage dashboard to wg-k8s-infra #1305)
• metrics jobs, which run queries against it and store results in gcs (migration ref: Migrate test-infra metrics to wg-k8s-infra #1306)

It's publicly readable (though compute time for queries will be charged to their own GCP account), but not publicly writable

It's not quite clear to me what strategies are available to us:

• stop-the-world: stop writing data, copy to new dataset, point everything at new dataset
• sync-the-world: replicate data from old dataset to new dataset, point reporting tools to replicated dataset, then point kettle at new dataset
• backfill-the-world: cut everything over to the new dataset, then backfill (day, then week, then all)

Given that this is a public artifact I'm inclined to suggest we home this under the kubernetes-public gcp project. We should examine billing briefly to understand if there are associated network/compute costs we would want billed elsewhere.

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4