DexaflowResilienceBackstopDisabled
What DexaflowResilienceBackstopDisabled means and what to do when it fires.
Symptom. dexaflow_resilience_backstop_disabled{config_key} == 1: the
control plane booted with that backstop off. The boot log has a matching WARN.
Impact. A failure the backstop would catch (a lost agent, a pod that never
started, an orphaned run) now leaves work stuck running or queued.
First commands.
kubectl -n <namespace> logs deploy/<release> | grep -i 'backstop'
helm -n <namespace> get values <release> -a | grep -n '<config_key>'
Decide. Turn the setting back on unless the change was deliberate and
temporary. If it was, silence this alert for that config_key with an
expiry, not forever. See
Scheduler resilience.