Folks,
This has been a source of confusion for me so please help clarify. I have learned tonight that when modifying the OSPF pacing timers, there are three timers to consider: group, flood and retransmission. The DOCCD has explained that the three of these timers can be modified if OSPF convergence needs to be modified. The group pacing and the flood pacing is obvious to be but the retransmission seems to be vague.
Reading http://www.cisco.com/en/US/docs/ios-xml/ios/iproute_ospf/configuration/12-4t/iro-update.html explains that the flood and retransmission timers modify interpacket time spacing or gaps between transmissions for LSAs sitting in "OSPF queues". Two different queues are used for LSAs based on their first or second attempt. All the pacing times are configured in router configuration mode and are considered global.
I haven't found the configuration document that explains the 'ip ospf retransmit-interval' command yet. The only document I found was the reference guide on the command: http://www.cisco.com/en/US/docs/ios-xml/ios/iproute_ospf/command/ospf-i1.html#wp1685875890. This document explains a simple timer that waits to retransmit the an LSA if no Ack is received.
My first question is....when an LSA is transmitted, does it go by the tranmission rate of the flood pacing timer? ...and if the LSA is retransmitted, it goes by the retransmission pacing timer? Therefore, the pacing timers are governing all LSA transmissions for the OSPF process and the 'ip ospf retransmit-interval' value is local to the interface. Which would mean that the 'ip ospf transmit-interval' needs to use the retranmission pacing timer in order to do its job. Please provide some clarification around this.
Does the 'ip ospf retransmit-interval' time value account for the individual LSAs? ...or does it group them together?
Thanks,
Mikee