Route Summarization
daveba123
Users Awaiting Email Confirmation Posts: 7 ■□□□□□□□□□
in CCNA & CCENT
Hello folks,
Before reading details about route summarization, my preconceived idea of autosummarization was that it was what advanced routing protocols used because it was automatic and automatic is better and more sophisticated. Obviously after more reading it is the complete opposite.
I was surprised to learn that not only was the implementation of autosummarization retired in newer protocols, but I couldn't find any reference to an alternative automatic summary feature in newer protocols like EIGRP and OSPF. My reasoning was that while auto summary was originally designed to aggregate classful routes in routing protocol updates that didn't include masks, a revised version that did include masks in updates was developed. So that's my question.
Is there anything that exists as a autosummary 2.0? I understand the importance of summarization and I'm just surprised that I can't find anything that's designed to do it automatically--albeit more precisely that autosummarization--to include masks in the routing updates. Is manual summarization the only option for today's relevant networks?
Thanks!
David
Before reading details about route summarization, my preconceived idea of autosummarization was that it was what advanced routing protocols used because it was automatic and automatic is better and more sophisticated. Obviously after more reading it is the complete opposite.
I was surprised to learn that not only was the implementation of autosummarization retired in newer protocols, but I couldn't find any reference to an alternative automatic summary feature in newer protocols like EIGRP and OSPF. My reasoning was that while auto summary was originally designed to aggregate classful routes in routing protocol updates that didn't include masks, a revised version that did include masks in updates was developed. So that's my question.
Is there anything that exists as a autosummary 2.0? I understand the importance of summarization and I'm just surprised that I can't find anything that's designed to do it automatically--albeit more precisely that autosummarization--to include masks in the routing updates. Is manual summarization the only option for today's relevant networks?
Thanks!
David
Comments
-
TechGuy215
Member Posts: 404 ■■■■□□□□□□
I'm pretty sure that RIP, IGRP, EIGRP and BGP auto-summarize by default, and that OSPF and IS-IS do not automatically summarize.
Also, I found this older post on the TE Forums, might that might help you:
http://www.techexams.net/forums/ccna-ccent/66463-command-no-auto-summary.html* Currently pursuing: PhD: Information Security and Information Assurance
* Certifications: CISSP, CEH, CHFI, CCNA:Sec, CCNA:R&S, CWNA, ITILv3, VCA-DCV, LPIC-1, A+, Network+, Security+, Linux+, Project+, and many more...
* Degrees: MSc: Cybersecurity and Information Assurance; BSc: Information Technology - Security; AAS: IT Network Systems Administration -
theodoxa
Member Posts: 1,340 ■■■■□□□□□□
I remember reading somewhere that the real reason for auto-summarization was backwards compatibility. By auto-summarizing to the classful boundary, classless protocols like RIPv2 and EIGRP could be backwards compatible with their classful predecessors RIPv1 and IGRP. OSPF does not auto-summarize. For BGP auto-summary might be a good thing as it would help reduce the size of internet routing tables which are WAY TOO LARGE. Then again, I would hope most ISPs were already summarizing their networks before advertising them.R&S: CCENT → CCNA → CCNP → CCIE [ ]
Security: CCNA [ ]
Virtualization: VCA-DCV [ ] -
NetworkVeteran
Member Posts: 2,338 ■■■■■■■■□□
automatic is better and more sophisticated. Obviously after more reading it is the complete opposite.
EIGRP indeed supports auto-summary, like other distance vector protocols.I was surprised to learn that not only was the implementation of autosummarization retired in newer protocols, but I couldn't find any reference to an alternative automatic summary feature in newer protocols like EIGRP and OSPF.
"Auto-summary is bad and should be turned off." is a common misconception. I once remember someone who spent time disabling auto-summary thinking it must be causing "mysterious problems", only to then go ahead and implement precisely the same manual summarization scheme! Auto-summary is like an automatic transmission, cruise control, or autoqos. It's important to know its limitations, but when used properly, it can save time and lead to reasonably optimized routing updates with minimal effort.
Nope! In fact, if your routing protocol doesn't include subnet masks in its updates, "auto-summary" and "no auto-summary" have no effect. The command and its opposite are actually only useful for protocols that include subnet masks in their updates!My reasoning was that while auto summary was originally designed to aggregate classful routes in routing protocol updates that didn't include masks
That is auto-summary! You can also manually summarize., a revised version that did include masks in updates was developed. -
theodoxa
Member Posts: 1,340 ■■■■□□□□□□
NetworkVeteran wrote: »"Auto-summary is bad and should be turned off." is a common misconception. I once remember someone who spent time disabling auto-summary thinking it must be causing "mysterious problems", only to then go ahead and implement precisely the same manual summarization scheme! Auto-summary is like an automatic transmission, cruise control, or autoqos. It's important to know its limitations, but when used properly, it can save time and lead to reasonably optimized routing updates with minimal effort.
That depends on the subnetting scheme. I try to always turn it off and only turn it back on where it would be useful.R&S: CCENT → CCNA → CCNP → CCIE [ ]
Security: CCNA [ ]
Virtualization: VCA-DCV [ ]