Greetings from Sydney. I'm a network noob, just upgraded my internet from ppp to cable. I use usb to connect the motorola cable modem as eth0, works great. I also use eth1 to allow a second box to share the internet connection. Eth1 is "via rhine" builtin on the gigabyte m/b. Also works great.
However - I find that when the network is up many gnome and kde gui apps take forever to appear on the screen after loading. I.e. program loads but takes maybe 5-10 minutes to appear on the screen. If I stop the network the apps appear instantly on loading on the screen. Looks to me like the network is grabbing some resource and holding on to it.
Details :-
kernel 2.6.12
mandrake 10
x.org 6.8.2
I used the Mandrake tools to connect the lan, I have no idea what it does under the hood.
thanks for any hints.

I have had issues with Mandrake, not setting up default gateway properly for instance.
Start with some checking:
# ifconfig
# route
# cat /etc/resolv.conf

Do this both with network up and down, report results here.

------------------- ooopppsss??? Message submitted wile writing?

I have had issues with Mandrake, not setting up default gateway properly for instance.
Start with some checking:
# ifconfig
# route
# cat /etc/resolv.conf
# cat /etc/hosts

Do this both with network up and down, report results here.

thanks for replying so quick ! details follow :-

[root@c211-30-153-231 root]# ifconfig
eth0 Link encap:Ethernet HWaddr 00:12:C9:4C:17:65
inet addr:211.30.153.231 Bcast:211.30.153.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:513810 errors:0 dropped:0 overruns:0 frame:0
TX packets:73232 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:155681849 (148.4 Mb) TX bytes:8402078 (8.0 Mb)

eth2 Link encap:Ethernet HWaddr 00:C0F:14:31:7C
inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:28 errors:0 dropped:0 overruns:0 frame:0
TX packets:1589 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:3174 (3.0 Kb) TX bytes:232880 (227.4 Kb)
Interrupt:19 Base address:0x6000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:1248 errors:0 dropped:0 overruns:0 frame:0
TX packets:1248 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:81283 (79.3 Kb) TX bytes:81283 (79.3 Kb)

[root@c211-30-153-231 root]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
211.30.153.0 * 255.255.255.0 U 0 0 0 eth0
192.168.0.0 * 255.255.255.0 U 0 0 0 eth2
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default 211.30.153.1.op 0.0.0.0 UG 0 0 0 eth0

[root@c211-30-153-231 root]# cat /etc/resolv.conf
search nsw.optushome.com.au
nameserver 203.2.75.132
nameserver 198.142.0.51

[root@c211-30-153-231 root]# cat /etc/hosts
127.0.0.1 localhost.localdomain host.hyenainternet.com host localhost

==========================================================================================

[root@c211-30-153-231 root]# /etc/init.d/network stop
Shutting down interface eth0: [ OK ]
Shutting down interface eth2: [ OK ]
Shutting down loopback interface: [ OK ]

[root@c211-30-153-231 root]# ifconfig

[root@c211-30-153-231 root]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface

[root@c211-30-153-231 root]# cat /etc/resolv.conf
search nsw.optushome.com.au
nameserver 203.2.75.132
nameserver 198.142.0.51

[root@c211-30-153-231 root]# cat /etc/hosts
127.0.0.1 localhost.localdomain host.hyenainternet.com host localhost

I think I've solved this.. I had the kernel pre-emption settings for a server, so I recompiled with appropriate settings for a workstation - presumbly necessary to allow guis to pre-empt whatever the lan is making the kernel do. Odd because the kernel originally seemed to decide on server setting for itself, I would ghave thought the configure utility would have forced me to select an option.

Uh.. the problem came back. I recompiled the 2.6.12 kernel to allow pre-emption, and gui apps loaded and displayed correctly for a few hours.. then the old problem came back, guis take 5-10 minutes to display when network is started. But why did it work OK for maybe 10 hours before the problem came back ? Something getting clogged up ?

Hmmm...
Strange this... Just one suggestion for now:
Add your nic's to /etc/hosts, sometimes it takes a long time if name/IP can't be found.

There seems to be a pattern.. starting an rtl8139 fast ethernet card seems to trigger the problem. But unloading its driver (8139too) does not clear the problem, nether does restarting the network without it. ??!! No problems using the builtin via rhine ethernet.

rtl8139 turned out not to be the problem.
Using gdb, programs always seem to hangup trying to create a new thread. How can the network affect thisd ?

I finally discovered that the problem is with the via rhine ethernet... terrible things happen if I compile the kernel with VIA_RHINE_MMIO. It's supposed to improve performance. When I turned this off, and booted with noapic everything plays nicely with the network.