Network & Infrastructure / Linux Networking & Host Performance
Destination-specific Linux routing evidence instead of cache folklore
A host takes an unexpected path after a policy change, and a generic route-cache flush is proposed. The real issue may be policy precedence, source selection or a different namespace.
Environment and inputs: Linux network namespaces, host routing and network drivers; availability and semantics depend on kernel and driver versions. Diagnostic/design example; not tested on production equipment.
Collect the evidence
Record the exact kernel version, routing rules, table entries and destination-specific lookup context. Compare the application's chosen source address and interface with the expected path.
Explain the result
Linux routing internals and cache behavior have changed over time. Historical commands and explanations may not describe the active kernel, so a flush is not an evidence-based universal repair.
Plan the fix
Correct the demonstrated route or policy mismatch using current documentation. Preserve before/after lookup results and make the configuration persistent through the actual network management system.
Check the result and rollback
Test the same application flow and management access after renewal or restart. Revert if the persistent manager recreates conflicting rules or the fix only changes an unrelated default lookup.