IIRC, you may need to sit down with the affected machines afterwards to make sure that their hostnames have not gotten stuck with the "(2)" designation.
I am getting my host names changed to indicate a duplicate like this "(2)" is placed after the name. There are no duplicate names. I have given most of the computers (about 120) very unique names such as adding a random letter after the names to insure uniqueness because of this issue and I still am getting duplicate names {with a (2) after the name}. Any ideas? We are using OD with xserves holding the home directories. Running Snow leopard with latest updates on servers and desktops. Mostly ethernet with a couple airports out there.
6 replies
The last time I saw something like this, there was a switch on our network that had, for want of a better phrase, gone feral. After power-cycling the switch, everything was fine. I would imagine that what is going on in your situation is something like this, or, as a previous poster suggested, an odd network configuration.
IIRC, you may need to sit down with the affected machines afterwards to make sure that their hostnames have not gotten stuck with the "(2)" designation.
IIRC, you may need to sit down with the affected machines afterwards to make sure that their hostnames have not gotten stuck with the "(2)" designation.
Your switch error probably created a network loop.
A loop can also arise when you have multiple switches and cables snaking around, and where a switch configuration that isn't a tree is created. Non-tree switch configurations with unmanaged switches are bad.
Switches with spanning-tree checks can detect and avoid creating loops (and you need that when you want some network redundancy), but most (all?) unmanaged switches lack these checks. And yes, occasionally managed switches can fail to correctly detect a loop; to botch the spanning-tree check.
A loop can also arise when you have multiple switches and cables snaking around, and where a switch configuration that isn't a tree is created. Non-tree switch configurations with unmanaged switches are bad.
Switches with spanning-tree checks can detect and avoid creating loops (and you need that when you want some network redundancy), but most (all?) unmanaged switches lack these checks. And yes, occasionally managed switches can fail to correctly detect a loop; to botch the spanning-tree check.
The auto-renaming by the OS helps identify Macs on local-area networks. You can control what the name of an individual Mac is by going to Apple menu --> System Preferences --> Sharing and then making sure that the "Computer Name" is what you want it to be, and then pressing the "Edit..." button and making sure the Local Hostname and Hostname fields are both identically-named as well (spaces are not allowed and are converted to dashes.) I also check the boxes that say "Use dynamic global hostname" and "Advertise services in this domain using Bonjour", and I leave the "User" and "Password" fields empty. After a reboot, everything is as I wish. Hope this helps everyone!
Only occurs if the network sees other devices having the same name. You'll have to clean whatever databases you've built that store names.
Is there a loop in your network somewhere?
I'm not the OP, so can't answer. 😉
duplicate host names