[ Date Index ][
Thread Index ]
[ <= Previous by date / thread ] [ Next by date / thread => ]
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Pete Walker wrote: > Neil Williams wrote: > >> > Yes. AFAIK, the cable from hub1 to hub2 should go into hub 2's uplink > (if it has one). Not sure what you mean by uplink button Most older ethernet hubs have an uplink button, I think mostly took the role of a crossover cable. Quite a lot also used 10B2 as an option for uplink - if the uplink is in a cabinet or secured computer room this was a good choice as you were less likely to pull the wrong cable, and some gave you an extra port this way. > as the hubs I > have used just have 2 sockets for port 8, one with a crossover for > uplink. New hubs have some sort of autodetect which means they can work > out if it is uplink or a normal connection. My worse ethernet hub experience was a network in France which didn't have Ethernet keepalive - it took ages to get a SUN networking guru who could tell me what the mysterious flashing diagnostics meant, and where to set the ignore keepalive flag. The sins of others we've seen with Ethernet hubs, spanning trees and the like would make your hair curl. There was a report of a hospital discussed before that deployed Ethernet with too many hubs between end points, broke the spanning tree algorithmn, and caused all sorts of weird problems, can't remember if it was here or NANOG. It was portrayed as a brilliant technical success for those that fixed it, but looked to me like basic ethernet misconfiguration, we fixed worse all the time without comment or press releases, and definitely without spending 6 figure sums with large equipment vendors. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQE+Pt1oGFXfHI9FVgYRAv9BAJ0cCrSorLKBQ9AuHjRSLHAOphwhJwCdEiN1 K2leQdihRwLFnYyqqUDcCGk= =z9mj -----END PGP SIGNATURE----- -- The Mailing List for the Devon & Cornwall LUG Mail majordomo@xxxxxxxxxxxx with "unsubscribe list" in the message body to unsubscribe.