Hide Forgot
Description of problem: So I've just upgraded a number of our machines from Red Hat 7.3 to Fedora Core 1. I've noticed that a number of them are having trouble with their network cards now. If I try and do anything network related then the machines will just hang or time out. If I look at the output of the dmesg command I see errors like so: eth0: ns83820_tx_watch: 0 3 3 eth0: tx_timeout: tx_done_idx=0 free_idx=3 cmdsts=2800002a eth0: after: tx_done_idx=3 free_idx=3 cmdsts=00000000 eth0: ns83820_tx_watch: 3 4 1 eth0: tx_timeout: tx_done_idx=3 free_idx=4 cmdsts=0800002a eth0: after: tx_done_idx=4 free_idx=4 cmdsts=00000000 eth0: ns83820_tx_watch: 4 7 3 eth0: tx_timeout: tx_done_idx=4 free_idx=7 cmdsts=0800002a eth0: after: tx_done_idx=7 free_idx=7 cmdsts=00000000 If I shut down the network service, unload the ns83820 module, and then restart the network service it usually fixes the problem. None of this ever happened on Red Hat 7.3. Version-Release number of selected component (if applicable): kernel-smp-2.4.22-1.2179.nptl How reproducible: Happens half the time after I boot the machines. Steps to Reproduce: 1. Boot a machine. 2. 3. Actual results: Sometimes the network won't work and I'll see those errors in the output of dmesg. Expected results: I expect things to work fine like they did under Red Hat 7.3 Additional info: The network cards in question are Netgear GA311 Gigabit PCI cards.
I've just noticed that this will happen ocassionally long after the machine has booted and/or I've fixed things by unloading and loading the module again.
My boss said he found a solution: turn off kudzu at boot time. However, I haven't tested this theory yet.
I've got an SMC9452TX card that IDs as a National Semiconductor 83820. This error cropped up during the install of Fedora Core 2, so I had to fall back to the onboard NIC. Post install, same problem remains. Update to 2.6.6-1.435 -- problem remains. Shut down networking, removed the ns83820 module, started up networking, problem remains; NIC can't even pull a DHCP address. Shut down networking, removed the module, manually inserted the module, started up networking, problem remains. Other facts: athlon 1800XP system, nforce2 motherboard, nforce ethernet NIC works great....
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/