Helm 'Error: UPGRADE FAILED: another operation (install/upgrade/rollback) is in progress'. How do I clear it?

Cause: Helm stores each release revision as a record in the cluster. If a previous helm install, upgrade or rollback was interrupted (CI job killed, timeout, lost connection) the latest revision can stay in a pending status (pending-install, pending-upgrade or pending-rollback). Helm then refuses new operations on that release. It can also be a genuine operation still running in another pipeline, so check that first. Fix (best first): 1. Make sure no other helm process or CI job is still working on this release. Wait for it to finish if so. 2. Inspect the state: helm history RELEASE -n NAMESPACE, and helm list -n NAMESPACE --pending. Look for a latest revision with a pending status. 3. Roll back to the last good revision: helm rollback RELEASE REVISION -n NAMESPACE, using the number of the last revision with status deployed (or superseded and known good). Without a revision number Helm rolls back to the previous release. Then rerun the upgrade. 4. If the release has no good revision (a first install that never completed), uninstall it with helm uninstall RELEASE -n NAMESPACE and install again. Review what the half-finished install created first so you do not delete data. 5. Prevent a repeat: give upgrades a realistic --timeout and use --wait, consider --rollback-on-failure (older Helm releases call this --atomic) so failures roll back cleanly, serialize deploys per release in CI, and do not kill running helm processes. 6. Avoid hand-editing or deleting the release Secrets that Helm uses for storage unless you understand the consequences and have a backup; that loses revision history. Verify: helm history RELEASE shows the newest revision as deployed after the retried upgrade, and helm list --pending shows nothing for the release. Sources: https://helm.sh/docs/helm/helm_rollback/ https://helm.sh/docs/helm/helm_history/ https://helm.sh/docs/helm/helm_upgrade/ https://helm.sh/docs/helm/helm_list/