Skip to content

Fix(ci): prevent delete-env failure due to busy persistent disk on te…#3800

Open
prkalle wants to merge 1 commit into
cloudfoundry:mainfrom
prkalle:fix/delete-env
Open

Fix(ci): prevent delete-env failure due to busy persistent disk on te…#3800
prkalle wants to merge 1 commit into
cloudfoundry:mainfrom
prkalle:fix/delete-env

Conversation

@prkalle

@prkalle prkalle commented Jun 18, 2026

Copy link
Copy Markdown
Contributor

Description of the Change

This PR fixes the transient delete-env CI failure caused by busy persistent disk during BOSH Lite teardown

Why Is This PR Valuable?

This PR ensures BOSH Lite environments are reliably torn down after integration tests complete. Previously, the delete-env job intermittently failed, leaving orphaned GCP resources (director VM, persistent disk) and dangling GCS state that required manual cleanup. It also caused the CI run to report a job failure even when all integration tests had passed.

Applicable Issues

List any applicable GitHub Issues here

How Urgent Is The Change?

Fairly urgent

Other Relevant Parties

Who else is affected by the change?

…ardown

Add bosh delete-deployment before bbl down in delete-bosh-lite.yml so
that CF warden containers release their bind mounts on the director's
persistent disk prior to bosh delete-env attempting to unmount it.
Without this, umount fails with "target is busy" (exit status 32).

Also align tool versions with repo variables (BBL_CLI_VERSION,
BOSH_CLI_VERSION) instead of hardcoded values, and remove the redundant
--gcp-service-account-key flag since BBL reads BBL_GCP_SERVICE_ACCOUNT_KEY
from the environment automatically.

Signed-off-by: Prem Kumar Kalle <prem.kalle@broadcom.com>
@prkalle prkalle marked this pull request as ready for review June 18, 2026 19:09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant