SUMMARY: name resolution, but rather sshd problem
Meier Adrian <[email protected]>
| Newsgroups | gmane.linux.admin.managers |
|---|---|
| Message-ID | <[email protected]> |
Hi list, First I must give a big thanks to all responses. But my problem is not resolved :-( Jeremy Rodriguez @cmsmechanical, Stephen Reppucci @logsoft, Chris Ricker @gatech, Martin Schmitt @scsy and Bob Vickers @cs.rhul think my problem is based on the reverse lookup (domain name service) made by the ssh server. I tried to add the ssh client machine in /etc/hosts on the remote machine. In addition I disabled "VerifyReverseMapping" in sshd_config chupacabra @linuxlink Has the same problem and waits for my summary. Please tell me whether my described answers helps you? Steve Sandau @bath.tmac Gives me the following hint: "On any of my machines, 'dig' always needs a FQDN to work. Just a hostname will time out" I think this is a task for a Red Hat consultant - we have support on these machines (will keep you uptodate) kind regards Adrian > Hi all, > > We use Red Hat Advanced Server 2.1 (Pensacola) without any > errata or quartal updates. To administrate the servers > remotly we use the OpenSSH suite supplied by official RPMs > and did not change the default sshd_config. > > A measurement with the shell keyword "time" shows me that > "ssh hostname" take 21 seconds till I can enter the password. > By adding the prefix domain.ch (ssh hostname.domain.ch) the > same not even take a half second. > > The stub resolver (resolv.conf) ist properly configurated > with valid named servers and "search domain.ch" plus "domain > domain.ch". In /etc/hosts there is a correct entry below > localhost with IP address and hostname.domain.ch. The > nsswitch.conf (and host.conf) points "hosts" to files then > dns. The error is definitely not our named configuration, > because with SUN and Windows servers it works fine. > > The really freaky is: the output of "nslookup hostname", > "traceroute i37390" and "ping hostname" appears immediately. > The answer given by "dig hostname" is "connection timed out; > no servers could be reached" although a trace show me that he > take the right named server. > > Because i think the problem is the ssh client, here a debug: > # ssh -v -v hostname > OpenSSH_3.1p1, SSH protocols 1.5/2.0, OpenSSL 0x0090602f > debug1: Reading configuration data /etc/ssh/ssh_config > debug1: Applying options for * > debug1: Rhosts Authentication disabled, originating port will > not be trusted. > debug1: restore_uid > debug1: ssh_connect: getuid 0 geteuid 0 anon 1 > [waiting, waiting] > debug1: Connecting to hostname [192.168.x.x] port 22. > > The IP settings on eth0 are correct > (/etc/sysconfig/network-scripts/ifcfg-eth0). > > Hope anyone can help me. Any suggestions are welcome. > > best regards > Adrian Meier > > UNIX systems administrator > T-Systems Switzerland _______________________________________________ > LinuxManagers mailing list - http://www.linuxmanagers.org > submissions: LinuxManagers-35TzE1X9F6582KRnZfj+bdi2O/[email protected] > subscribe/unsubscribe: > http://www.linuxmanagers.org/mailman/listinfo/> linuxmanagers _______________________________________________ LinuxManagers mailing list - http://www.linuxmanagers.org submissions: LinuxManagers-35TzE1X9F6582KRnZfj+bdi2O/[email protected] subscribe/unsubscribe: http://www.linuxmanagers.org/mailman/listinfo/linuxmanagers