Hi all....OK, I've read through what feels like hundreds of threads about mozilla having a resolving host problem. To date, I have not found a solution that works on my redhat 9 system. Here's the basics on my system and the problem as I see it.

My system: (forgive me..I know my system is WAY out of date but here it goes....
Celeron 300MHz with just under half a gig of memory.
Running duel boot RedHat9 (updated regularly with yum) and Win2k
3c905B 100BaseTX[Cyclone] ethernet card
Running off direct ethernet link to University network via DHCP
DNS settings are correct as specified by university sys admins

My problem on the RedHat side:
I can surf to google, yahoo, cnn, countless adult sites , etc.....
but when I try to get to my homepage or my wifes page, mozilla just sits there and tries to resolve the host and never gets there. We pay yahoo to serve out our websites and using mozilla on the Win2k side works fine. I've tried using firefox and Opera and both do the same thing. This implies that there is a network problem (DNS, etc) but I've confirmed with the powers that be that my DHCP and DNS settings are correct. I can ping yahoo, google, as well as my work system on campus with no problems.

I've tried the fixes described on different threads but with no luck. I'm not quite a linux newbie but I'm not a pro either so detailed howto's would be greatly appreciated! Oh great one's, please share the knowledge and help me figure this out before I go completely mad!

THANKS!

please feel free to directly email me if you wish...

If the issue is specific to your system and you dont mind doing it the hard way, add some:

your.page's.ip.here yourserver.dnsname.com

to your /etc/hosts file.
also, even if you dont want to do that, check the file out anyway to see if theres some funky entry for one of your pages thats messing it up.

If you have bind-utils installed, then you can try using dig to query the campus nameserver, and other nameservers.

Check out 'man dig'

Enjoy!
--- Cerbere

depending on the type of connection you have you might try this


ifconfig eth0 mtu 1410

DavidPhillips....

Much thanks!! the ifconfig etho mtu 1410 command worked like a charm. I'm not having anymore problems on any site I've gone to so far. Are there any negative side affects that I should know about resulting from modifying the ifconfig eth0 like that? Is this a permanent thing or will I have to run that command every time I reboot? If it's not permanent, is there a way to make it so?
Thanks again.....

That's great Steve,


The downside is smaller packets mean there will be a slight increase in the required number of packets transmitted with less transmission time per packet. The added overhead of the headers will mean more data transmitted. It is generally considered to be best to use the maximum usable packet size. You could figure only the sites that were a problem will be affected by the change. You could try to increase the number to see where it fails, although I consider 1410 to be a safe number to use in a dsl or other pppoe, or double router connections. You could test other numbers. You might even get by with 1492.


1410 is more or less a catch all, where you can start and go up from there.


Maybe ping would be a good way to test it like this...


david@zeus:~> ping -s 1500 -M do linuxquestions.org
PING linuxquestions.org (64.179.4.149) 1500(1528) bytes of data.
ping: local error: Message too long, mtu=1500
ping: local error: Message too long, mtu=1500

--- linuxquestions.org ping statistics ---
2 packets transmitted, 0 received, +2 errors, 100% packet loss, time 999ms


david@zeus:~> ping -s 1472 -M do linuxquestions.org
PING linuxquestions.org (64.179.4.149) 1472(1500) bytes of data.
1480 bytes from web1.linuxquestions.org (64.179.4.149): icmp_seq=1 ttl=48 time=524 ms
1480 bytes from web1.linuxquestions.org (64.179.4.149): icmp_seq=2 ttl=48 time=512 ms
1480 bytes from web1.linuxquestions.org (64.179.4.149): icmp_seq=3 ttl=48 time=539 ms

--- linuxquestions.org ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2000ms
rtt min/avg/max/mdev = 512.997/525.542/539.028/10.680 ms



david@zeus:~> ping -s 1473 -M do linuxquestions.org
PING linuxquestions.org (64.179.4.149) 1473(1501) bytes of data.
ping: local error: Message too long, mtu=1500
ping: local error: Message too long, mtu=1500

--- linuxquestions.org ping statistics ---
2 packets transmitted, 0 received, +2 errors, 100% packet loss, time 999ms



You should be aware that by this method you have 28 bytes added, so in my case here though 1500 is the correct setting for my connection I use a size of 1472. If you look at the other number shown in the local response it's 1500..

1472(1500)

PING linuxquestions.org (64.179.4.149) 1472(1500) bytes of data.

To prevent it from showing a local error set the mtu to a number high enough to test the outside network. If you see local errors as shown above it's telling you the packet size your using exceeds the interfaces mtu setting.


You could also try this way...

david@zeus:~> ping -s 1500 -M want linuxquestions.org
PING linuxquestions.org (64.179.4.149) 1500(1528) bytes of data.
1508 bytes from web1.linuxquestions.org (64.179.4.149): icmp_seq=1 ttl=48 time=1151 ms
1508 bytes from web1.linuxquestions.org (64.179.4.149): icmp_seq=2 ttl=48 time=730 ms
1508 bytes from web1.linuxquestions.org (64.179.4.149): icmp_seq=3 ttl=48 time=610 ms

--- linuxquestions.org ping statistics ---
4 packets transmitted, 3 received, 25% packet loss, time 3093ms
rtt min/avg/max/mdev = 610.842/831.104/1151.481/231.791 ms, pipe 2



The config file for the eth0 interface is /etc/sysconfig/network-scripts/ifcfg-eth0


Add this entry so ifup will bring up the interface with your required settings...


MTU=1410

That's it.