Our parent "company" had the idea of letting a TelCo do the same for us. The TelCo was/is contracted to set up a "gateway" between the old network and the new network. Lucky for us we're smarter than they are (at least we think we are...) and are having things our way. :-)
It's difficult to try keeping a "hands-off" approach to letting a vendor install a major piece of equipment in the core of MY network. Maybe you know how it is... watching an engineer (CCNP with some experience) who is supposed to know what he's doing but brings down the network instead. Add a CCIE with a few years experience to the mix, three more tries with two of them resulting in network outages. The suggested configuration for our first 6500 (a 6513 with dual Sup2/MSFC2/PFC2) was to run in HSRP between the two MSFCs. (For those who don't know, an MSFC - Multiservice Switch Feature Card or something like that, is a routing module on the switch supervisor). We are running IP, IPX and AppleTalk. What they didn't tell us was we had to keep the configurations sync'ed between the two MSFCs otherwise there would be problems. Config Sync doesn't support AppleTalk. Guess what... we had problems. We're now running SRM. I guess I should get to the point here. My experience says "forklift" upgrades are bad. Way too much room for things to go bad. Considering the problems introducing a single new core box, I'd take it slow. make sure everything runs as planned while you upgrade. Make sure Spamming-Tree (did I say that??) is working properly, VTP is communicating and the hardware has burned in. That brings up another thing: after the first few weeks, one of the lights on the switch fabric module stopped showing green (looked burned out). While tempted to say it's just a light, you have no idea what the root problem causing the light not to function is. We had the vendor replace the card. It seems our vendor keeps shooting itself in the foot. Their engineers keep doing things that prevents them from earning any respect from us. On a positive note, I know one engineer with that company who I do respect. Too bad they keep him in design. :-) Ken >>> "Chuck's Long Road" 10/07/02 09:45PM >>> interesting. The following may or may not be feasible, depending upon space in closets, and cost of implementation. It is something my employer is supposed to be doing for the various "branch offices" of a major customer we have during a major network forklift upgrade. It has tended not to happen this way for a lot of reasons, political and practical. 1) Place all new switches into the various closets. Connect them up 2) set up a gateway ( single link ) between the new core switch(s) and the old core switch(s) 3) test connectivity by taking a laptop with as many user applications as practical, and go from closet to closet testing connectivity to the various servers, services, etc. 4) assuming point three results in connectivity everywhere, do a closet by closet migration of users from the old switches to the new switches. 5) migrate all devices ( servers, internet, etc ) onto the new core switches. 6) assuming all remains well, unplug the old stuff and if Cisco is not offering you a generous trade in, sell it on one of the auction sites. Like I said, sometimes time, space, and cost does not permit this. Chuck -- www.chuckslongroad.info like my web site? take the survey! ""Azhar Teza"" wrote in message [snip] Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=55150&t=55064 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
