By default IGRP and EIGRP don't use reliability and load, as you probably know. But when IGRP and EIGRP do use them, reliability and load are dynamically learned. Reliability is the worst reliability for any interface in the path. It's based on the ability to send and receive keepalives, CRCs, etc. You can see the reliability for a particular interface with the "show int" command. Load is cumulative. You can see load for each interface with the "show int" command.
The best document I have found actually covers IGRP, but EIGRP behaves like IGRP with regards to this question. The IGRP doc is here: http://www.cisco.com/warp/public/103/5.html TAC did some terrific documentation on EIGRP here: http://www.cisco.com/warp/public/103/eigrp-toc.html The reason you won't hear much about reliability and load is because they really don't work with EIGRP. I think it was just a Cisco marketing ploy to make a big deal of them. In his book, "EIGRP Network Design Solutions," Ivan Pepelnjak points out that routers can't dynamically keep track of reliability and load for an entire path if the routers in the path don't update each other on a regular basis. IGRP routers update each other every 90 seconds. EIGRP routers don't. Regarding manually assigning them, you only do that when redistributing. You have to make up something because you're redistributing from a routing protocol that doesn't track reliability and load. Priscilla At 12:10 AM 4/24/02, Kevin Jones wrote: >Hello everyone, > >I have a question that I have been struggling with for quite some time. Are >the reliability and load metrics in EIGRP (or IGRP for that matter) >dynamically learned? If so, why do we manually assign values like we do for >bandwidth and delay. I have searched numerous Cisco white papers and have >found only one article where it mentions the two as being dynamically >learned. Since I have not found any others that mention it, I am starting >to feel that the one article is a typo (or I am just not understanding it >the way it is worded). I would think that if they were dynamically learned, >then there would be more information about the process. No other routing >protocol is able to detect such statistics on the fly (to my knowledge). I >understand that dynamic detection might not be a good thing, esp. if the >reliability and load were constantly changing, but never-the-less there >should be more info somewhere. > >If you can find more than one specific white paper and lead me to them, I >would appreciate it. > >Thank you, > >Kevin Jones >CCNA, CCDA, CCNP, CCDP >A+, Net+, I-Net+ ________________________ Priscilla Oppenheimer http://www.priscilla.com Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=42455&t=42406 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
