That's why my sed scripts ignore router-to-switch connections, I use a 3750 breakout switch and a set of 4 lab switches. It's the only way to get full functionality with a Dynamips lab.
In early testing of the setup I used all 3550s and it works, you just won't get CDP both ways if you're using the QinQ/breakout switch method, and you lose all the 3560-specific features. 3550s usually run US$75-85 on eBay, which is pretty reasonable in my opinion. Sent from my iPad On Aug 5, 2012, at 8:41 PM, Adam Booth <[email protected]> wrote: Hi John, I don't know if you're aware of it but another option is to look at using real switches with dynamips/GNS3. Tentatively you could get by with 4 x 3550s which support a reasonable amount of feature parity with the 3560 (but missing out on QoS, private VLAN and I think IPv6) which are relatively inexpensive to acquire on ebay these days. You will either need to have a lot of physical interfaces representing each router interface connecting to a switch (using multiple 4 Port Ethernet NICs or a bunch of USB NICs), or you can look at using another switch using dot1q tunnelling to act as a breakout switch. There are a number of blogs that work document how to do this, and it's what I used for my home lab (though I did have some 3750s which are pretty much stackable 3560s) Tentatively while you need 4 switches (plus possibly another as a breakout switch) only two need to be a 3560/3750 since most of the workbook labs that rely on 3560 specific functionality occur on Cat1/Cat2 meaning Cat3/4 can be lower cost 3550s. Cheers, Adam On Mon, Aug 6, 2012 at 10:00 AM, John Olsen <[email protected]> wrote: > Hi Guys > > Ok, Thanks to Bob for sending me the text file, however this does not seem > to do all that I need. > > What started me thinking about this was that so far I have been working > through the labs from book 1 by building my own config in GNS3. Apart from > the first few, where the switches are a vital part of the lab, this works > pretty well, with a bit of care over the inevitable different interface > designations. However I have come to lab 13, where the CAT switches need to > take part in the BGP peering. So what I have been trying to do is make the > IPexpert GNS3 config work to do this. First thing I found that I needed > that I did not have was a cross reference between the interfaces used on > the actual Cat switches and the interfaces that replace them in the GNS3 > config, so I now have a spreadsheet that does that. Next I need to get > them configured to do as much as possible, eg set up the appropriate > trunking. At the moment I am getting kind of bogged down in that..does > anyone have any configs for that, eg the equivalent of the initial configs > as provided in the workbooks, only for the GNS topology instead of the Cat > switches. Or is this not actually feasable? Which labs is the GNS3 topology > really useful for and which ones would I be better to do using rack time? I > do have lab time available on the Proctor labs, but find an eight hour > stretch a bit hard to schedule. > > regards > John > > > On 1/08/2012 12:35 p.m., Bob McCouch wrote: > >> bob@lab:~/scripts$ cat ipx-config-fixups.txt >> ##############################**######################### >> ##### THIS IS NOT AN EXECUTABLE SCRIPT ##### >> ##############################**######################### >> >> >> > ______________________________**_________________ > For more information regarding industry leading CCIE Lab training, please > visit www.ipexpert.com > > Are you a CCNP or CCIE and looking for a job? Check out > www.PlatinumPlacement.com > > http://onlinestudylist.com/**mailman/listinfo/ccie_rs<http://onlinestudylist.com/mailman/listinfo/ccie_rs> > _______________________________________________ For more information regarding industry leading CCIE Lab training, please visit www.ipexpert.com Are you a CCNP or CCIE and looking for a job? Check out www.PlatinumPlacement.com http://onlinestudylist.com/mailman/listinfo/ccie_rs
