Figure out what's required to re-enable https://kubernetes.io/docs/concepts/workloads/controllers/cron-jobs/#new-controller after last night's incident
Description
Description
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Open | None | T327025 Upgrade Toolforge Kubernetes to version 1.26 | |||
| Open | None | T316107 Upgrade Toolforge Kubernetes to version 1.25 | |||
| Open | None | T307651 Upgrade Toolforge Kubernetes to version 1.24 | |||
| Resolved | taavi | T298005 Upgrade Toolforge Kubernetes to version 1.23 | |||
| Resolved | taavi | T286856 Upgrade Toolforge Kubernetes to latest 1.22 | |||
| Resolved | rook | T308172 Upgrade PAWS to Kubernetes 1.21 | |||
| Resolved | taavi | T282942 Upgrade Toolforge Kubernetes to latest 1.21 | |||
| Resolved | BUG REPORT | None | T308189 Toolforge jobs stopped getting scheduled around the same time as the Toolforge k8s cluster upgrade | ||
| Resolved | taavi | T308205 Re-enable CronJobControllerV2 |
Event Timeline
Comment Actions
toolsbeta has the v2 controller enabled, tried creating a cronjob there and it worked (though there were no previously existing ones).
Comment Actions
Mentioned in SAL (#wikimedia-cloud) [2022-05-12T12:36:57Z] <taavi> re-enable CronJobControllerV2 T308205