Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!agate!usenet.ins.cwru.edu!eagle!galaxy!hubler From: hubler@galaxy.lerc.nasa.gov (Dale Hubler) Newsgroups: comp.unix.aix Subject: Re: FAQ about AIX Keywords: AIX dump bind rlmd Message-ID: <1991Apr9.114656.22105@eagle.lerc.nasa.gov> Date: 9 Apr 91 11:46:56 GMT References: <2897@speedy.mcnc.org> Sender: news@eagle.lerc.nasa.gov Reply-To: hubler@galaxy.lerc.nasa.gov (Dale Hubler) Distribution: na Organization: NASA Lewis Research Center, Cleveland, Ohio Lines: 26 In article <2897@speedy.mcnc.org> kerm@mcnc.org (Cary E. Burnette) writes: >2. If you are running name server can you make the network services > check the /etc/hosts file first. My RS/6000 LAN has an unreliable > link to its nameserver. When this link is down I still want to be able > to communicate within the RS/6000 LAN. Know when the link is down I > have to wait a long time for telnet to time-out and try the /etc/hosts > file. Most of the traffic is among the RS/6000's and since it is small > I do not want to set up one as a nameserver. But on the few occasions > that I do want to go out on the internet I want to have the name server > available. We are in the same situation. The IBM's seem to have a hard time recovering from the loss of the nameserver. We are planning to set up a cacheing name server for our group. That will cache the information for routing that we are using most the time. The few times it needs more info it can go to the primary nameserver. The rest of the machines point to the local caching nameserver, keeping the traffic local. When the nameserver or the communications to it are down, our group can still operate without delays. I wish they would look at their own host table first prior to checking the nameserver. -- Dale A. Hubler -- Sverdrup Technology -- (216) 977-7014 hubler@galaxy.lerc.nasa.gov Try Eggo brand Common Sense Oat Bran frozen waffles, it's the health food of the 90's