Network & Infrastructure / OSPF & Underlay Convergence
OSPF E1 versus E2 external metrics and unexpected exits
A router selects a distant redistribution point even though a nearer point advertises the same external destination. Comparing interface cost alone does not explain external-path selection.
Environment and inputs: OSPFv2 IPv4 underlays, with OSPFv3 explicitly identified where used; timers and diagnostics vary by platform. Diagnostic/design example; not tested on production equipment.
Collect the evidence
Record external route type, external metric, internal cost to each ASBR or forwarding address and route preference. Check competing advertisements for the exact destination and prefix length.
Explain the result
Type-1 external cost includes the internal path contribution, while type-2 comparison emphasizes the external metric and uses internal distance in relevant tie-breaking cases. Different types are not interchangeable.
Plan the fix
Define whether internal distance should influence the external exit choice. Correct the redistribution metric type and values for the intended prefix class rather than changing unrelated underlay costs.
Check the result and rollback
Validate selection from several source areas and after loss of either exit. Revert if the change attracts traffic to a less capable ASBR or alters unrelated external services.