greetz
first of all, as a small hint try to keep your case written in a simple
form, ive tried to read it
several times and still i am not sure whether i got the whole picture! i had
to take out a pen
and paper to draw a bit ;)
anyhow before i go on this goose chase so to speak, there are several
questions that would be
good to present first.
as you have placed the suse box as an in between server to separate
networks,
how did you configure the itnerfaces on the box?
did you give the primary interface (im presuming its eth0) the following
parameters
ip 192.168.1.2
mask 255.255.255.0
gw 192.168.1.1 (or 192.168.1.254 varies on implementation ^^ )
dns 127.0.0.1, 172.16.4.2 (what ever your operator dns is )
for the remaining interfaces did you give a gateway or just gave an ip /mask
eth1
ip 192.168.2.1
mask 255.255.255.0
eth2
ip 192.168.3.1
mask 255.255.255.0
did you attempt to configure any routing?
there is no need if you just kept the network configuration simple as
mentioned above :)
personally about 4 years ago i tried to use ddns and found it to cause
enough headaches so
i just gave up on it back then and stuck to a static dns setup dunno whether
it works better nowadays
so if you would use a static dns and drop ddns that should help a lot.....i
guess :)
the samba can mess up a bit things in a sence that when windows attempt to
resolve
depending on how you have configured your windows clients, they might
resolve using
first the lmhost file which might contain static entries or then again it
would resolve from the network
so double check your lmhost for static entries on the windows systems
and if you decide to remove the ddns from your network, unselect the
register dns from the windows
network settings cause that would spare your suse box from log records :)
now back to your dns configuration,
what interfaces have you got the dns to listen on?
i think to ease the goose chase here is for you to send a more elaborative
description
of your environment setup and your dns named.conf file
(personally i rewrote the named.conf on my suse 9.3, its now cleaner and
simpler than what yast generates :) )
regards
RutePoint
PS: Happy New Year
On 1/1/06, Eric Hines
Folks,
(Retransmitted because my last got labelled SPAM; hopefully, I gotten the offending parts out this time.)
I'm having trouble getting name resolution to work, and I'd like your help. I'm running a SUSE Pro 9.3 server that also runs a Samba server, a dhcp server, and a dns server. This server box has three NICs (all addresses are private--192.168--so I'll just use the last two octets): a .1.0 that faces the Net (with sserver on .1.2, via YaST and in the hosts file (a router/switch is on .1.1 that serves as firewall for the whole network, including some devices not in my part)), a .2.0 subnet (lserver0 and lserver01 on .2.2) with a Win2k PC, and a .3.0 subnet (lserver02 on .3.1) with a laptop that's dual bootable between WinXP and SUSE Pro 9.3. Both subnets must go through the .1.0 interface to get to the Internet. The name resolution problems seem limited to my LAN. I get good addressing from the dhcp server, and both subnet devices get onto the Internet just fine. I mention the Samba server only for completeness; I don't (yet) think it's involved in these problems.
What I'm seeing is this: I can get hostname resolution from both the PC and the laptop as XP with one exception: a name a few months ago, lserver0 had a .1.103 address. When the PC pings lserver0 (which now has the .2.2 address), it gets back, still, .1.103 on one of Window's standard 4 replies; the other 3 replies are timed out errors. Flushing the resolver cache (ipconfig -flushdns) has no effect on this. The laptop as XP has no such problem. However, the laptop as SUSE cannot ping by hostname--it only gets unknown host. It can, however, ping by FQDN. All "three" devices, though, have always been able to ping sserver (the interface facing the Internet) successfully.
An ethereal trace has the PC sometimes failing to get a right answer via DNS but then getting the right answer via (Windows?) NETBIOS name service lookup, and sometimes getting the right answer via dns, and so in both cases getting successful pings. It's like dynamic dns isn't updating the reverse lookup tables; although, frankly, I'm uncertain that the forward tables are getting updated, either--vis., the problem with pinging lserver0 described above. However, I have "ddns-updates on;" and "ddns-update-style interim;" in my dhcpd.conf file (without the quotes).
My questions are these: what should I look for to get this running correctly? How can I check to see whether ddns really is functioning as it should? How do I read the tables where the dns functionality caches its data on the various devices (i.e., what command presents those tables)?
What other data do you all need in order to help me with this? I have 9 DNS files (although I suspect the root.hint file is not a problem); my dhcpd and named config files; I can generate ethereal traces, if these are useful; etc. They'll make for a large email, though. I hope, though, that this is just something I'm doing that's incredibly brain dead and easy to fix.
Thanks for your help.
Eric Hines
There is no nonsense so errant that it cannot be made the creed of the vast majority by adequate governmental action. --Bertrand Russell
-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
-- ___________________________ If computers were made in heaven, would they be perfect? ___________________________