Careful on going MST with Cisco. I wanted to do this too but Cisco said not to do it. If you are moving away from Cisco at your core anyway I'd hold off on MST. For us going to PVST RST worked well. Just can't do lags.
Reinhard: Thanks for the tips. We may be able to get the LAGs going after all. John On Fri, Dec 28, 2012 at 12:06 PM, Strebler, Reinhard (SCC) < [email protected]> wrote: > Hello Adam, > > With some luck I can give you some advice. > > > Am 28.12.2012 16:00, Baack, Adam wrote: > >> Disclaimer: We are mostly server guys here and know enough switching to >> be dangerous. We know very little about spanning tree, etc. >> >> We're implementing our phase 1 of our Enterasys replacement and have been >> installing B5 stacks in all of our headquarter IDFs. They are connecting >> back to a Cisco 6509 (Sup 2). >> >> We just started having problems where the 6509 would err-disable the LACP >> lag between the 6509 and the IDFs. The lag is set for trunking. In the >> terminal of the 6509 it shows channel mismatch error. We spoke with GTAC >> and they said something about the B5 stacks not participating in spanning >> tree and the 6509 doesn't know what to do so it disables it. >> > > We are familiar with this situation. We are having Cat6500 as well and > Enterasys E1/C2/C3/C5/G3 and DFE as layer 2 switches. On the other have we > are having different types of Catalyst switches, too. > > On the Cat6500 you should configure: > no spanning-tree etherchannel guard misconfig > > This helps against some internal checks disabling port-channels etc. > > You have configured LACP on Enterasys? What is the proper config on > Cat6500? > > On Enterasys you should use "set port protected" against loops during > setup phase of LACP/Port-channel. On Cat6500 there should be a similiar > command "port-channel standalone-disable", which have not tested so long. > > > Also, some of our IDFs have another Cisco switch farther down... example: >> 6509 --> B5 Stack --> Cisco 3750. The 3750 is also trunking. These >> setups seem to be the ones with problems. >> > > This is a crazy situation. We do not have any concept for this right now. > My suggestion: Go to mst on *all* Cisco switches (including Cat6500). > > > GTAC suggested possibly moving from PVST to RSTP so all of our >> spanning-tree is the same. Unfortunately I'm not exactly sure how we would >> do that or if that's the solution for us. >> > > Well, that's Cisco and spanning tree :-( > > On Cat6500 you have three choices: > (config)#spanning-tree mode ? > mst Multiple spanning tree mode > pvst Per-Vlan spanning tree mode > rapid-pvst Per-Vlan rapid spanning tree mode > > mst is IEEE 802.1s (Multiple Instance STP) > The other two ate Cisco proprietary. Enterasys will not be able to > understand the proprietary protocol and will forward the frames as regular > multicasts. This results in a break of spanning tree in an Cisco cloud and > a Enterasys cloud, which partially overlaps. Be award of this and of > possible results. > > With Cisco in proprietary spanning tree mode Cisco will ignore > IEEE-STP-BPDUs - which is crazy, too. > > I hope I could give you some advice. In case of any questions please don't > hesitate in contacting me > > Kind regards and a Happy New Year > Reinhard > -- > Karlsruhe Institute of Technology (KIT) > Steinbuch Centre for Computing (SCC) > Dipl.-Ing. Reinhard Strebler > Head of Department Networking und Telecommunikation (NET) > Kaiserstraße 12 > 76131 Karlsruhe > Phone: +49 721 608-42068/29046 > Fax: +49 721 32550 > E-Mail: [email protected] > www.scc.kit.edu > KIT - University of the State of Baden-Württemberg and > National Research Center of the Helmholtz Association > > --- > To unsubscribe from enterasys, send email to [email protected] with the > body: unsubscribe enterasys [email protected] > -- John Kaftan IT Infrastructure Manager Utica College --- To unsubscribe from enterasys, send email to [email protected] with the body: unsubscribe enterasys [email protected]
