mtrp:policies
This is an old revision of the document!
Table of Contents
Routing Policies
Outbound Path Selection
Outbound path selection applies to direct-neighbor candidates.
Supported modes:
FIRST_VALID: keep the first usable candidate foundLAST_IN: prefer the most recently seen candidateBEST_METRIC: prefer the lowest metric candidateMOST_RELIABLE: prefer the candidate with the best reliability score
Learned Route Selection
Learned route selection applies to forwarding-table candidates.
Supported modes:
FIRST_VALID: keep the first usable forwarding candidateLAST_UPDATED: prefer the most recently updated candidateBEST_METRIC: prefer the lowest metric candidateSTICKY_BEST_METRIC: keep the current preferred route until a meaningfully better candidate appears
Sticky Route Behavior
STICKY_BEST_METRIC uses:
- preferred-route state
- improvement delta
- hold-down timing
This reduces route flapping when several similar candidates exist.
Reliability Note
Current reliability scoring is hop-local.
It reflects local send success and failure behavior toward a next hop, not full end-to-end delivery confirmation.
/var/www/wiki.moshnetworks.com/data/attic/mtrp/policies.1781095924.txt.gz · Last modified: by mosh
