You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit a234a61
Browse filesBrowse the repository at this point in the historyBrowse files
test(e2e): make crust-gather failure snapshots reliable and complete
The per-test crust-gather snapshot silently dropped the tenant-cluster
dump: the collect's output went to /dev/null, the tenant kubeconfig's
localhost port-forward is usually dead by the time the EXIT trap runs,
and Helm release storage (kind Secret) was excluded along with every
other Secret. The dump that mattered most for a tenant HelmRelease stuck
mid-install was therefore missing, with no clue why.
- capture each collect's stdout/stderr/rc to a sibling .collect.log and
print a per-target status, so a failed collect leaves evidence in the
artifact instead of vanishing
- probe the tenant API and re-establish the port-forward from the
kubeconfig (name + server port) before collecting, so the tenant dump
is actually reachable
- bound every collect with `timeout 5m` so a wedged forward can't stall
the job
- add a dedicated include-only collect for Helm release secrets
(sh.helm.release.*) for host and each tenant, so the artifact carries
each release's revision history and status
crust-gather v0.15.2 puts all CLI filters in one FilterList evaluated
with .all(), so `--include-kind Secret --include-name <re>` captures
only secrets matching both — verified from source.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Signed-off-by: Myasnikov Daniil <[email protected]>
0 commit comments