A security namespace is relabeled and later FQDN-based egress policy behaves inconsistently for only some pods.
A Cilium network policy blocks unexpected egress only in one security...
Policy looks right, yet the egress decision is made on stale name-to-identity context for one namespace after relabeling.
Scenario
What to check first
- Identify the primary failure signal in the The FQDN Policy Evaluated Yesterday's Identity scenario.
- Separate visible symptoms from the underlying technical dependency.
- Describe the safest recovery path and the follow-up prevention work.
Checking checklist
- Summarize the current impact and the last known change.
- Collect direct evidence from logs, runtime state, and configuration before changing anything.
- Separate immediate recovery from permanent prevention work.
Recovery and prevention
Inspect DNS proxy and endpoint identity cache state before broadening the egress policy.
Questions worth viewing together
Community-field security problem inspired by Cilium community cases where DNS-proxy-backed policy decisions used stale endpoint identity after relabeling. Name-based policy decisions can stale independently from pod labels and policy objects.
Teams often blame DNS when the decision cache still maps the name to an old endpoint identity.
Identity-aware network policy engines need convergence checks after relabeling workloads that use FQDN rules.
Similar cases seen in the field