Thanks to everyone for their suggestions.
I can't try most of them out until after the long UK bank holiday weekend as I only have ssh access to the server from home and don't want to disable networking!
There are three network cards in the server, but I've only got the RealTek connected. Looking at /var/log/messages suggests the new kernel is handling all three NIC's differently from the old kernel.
Does anyone know what no GSI means with NICs and the new kernel?
Thanks in advance
Clive
old kernel:-
Aug 25 08:14:53 cc kernel: 8139too Fast Ethernet driver 0.9.27 Aug 25 08:14:53 cc kernel: ACPI: PCI interrupt 0000:00:0e.0[A] -> GSI 5 (level, low) -> IRQ 5 Aug 25 08:14:53 cc kernel: eth0: RealTek RTL8139 at 0xa800, 00:10:dc:59:bd:dc, IRQ 5 Aug 25 08:14:53 cc kernel: ACPI: PCI interrupt 0000:00:06.0[A] -> GSI 5 (level, low) -> IRQ 5 Aug 25 08:14:54 cc kernel: 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html Aug 25 08:14:54 cc kernel: 0000:00:06.0: 3Com PCI 3c905B Cyclone 100baseTx at 0xa400. Vers LK1.1.19 Aug 25 08:14:54 cc kernel: via-rhine.c:v1.10-LK1.2.0-2.6 June-10-2004 Written by Donald Becker Aug 25 08:14:54 cc kernel: ACPI: PCI interrupt 0000:00: 05.0[A] -> GSI 11 (level, low) -> IRQ 11 Aug 25 08:14:54 cc rpcidmapd: rpc.idmapd startup succeeded Aug 25 08:14:54 cc kernel: eth2: VIA Rhine at 0xe3000000, 00:50:ba:a1:86:44, IRQ 11. Aug 25 08:14:54 cc kernel: eth2: MII PHY found at address 8, status 0x7809 advertising 05e1 Link 0000. Aug 25 08:14:54 cc kernel: ACPI: PCI interrupt 0000:00:11.5[C] -> GSI 10 (level, low) -> IRQ 10 Aug 25 08:14:54 cc kernel: pciehp: acpi_pciehprm:_SB_.PCI0 evaluate _BBN fail=0x5 Aug 25 08:14:54 cc kernel: pciehp: acpi_pciehprm:get_device PCI ROOT HID fail=0x5 Aug 25 08:14:54 cc kernel: USB Universal Host Controller Interface driver v2.2 Aug 25 08:14:54 cc kernel: ACPI: PCI interrupt 0000:00:11.2[D] -> GSI 11 (level, low) -> IRQ 11 Aug 25 08:14:54 cc kernel: uhci_hcd 0000:00: 11.2: UHCI Host Controller Aug 25 08:14:54 cc kernel: uhci_hcd 0000:00:11.2: irq 11, io base 0000b000
new kernel
Aug 25 08:07:53 cc kernel: 8139too Fast Ethernet driver 0.9.27 Aug 25 08:07:53 cc kernel: ACPI: PCI interrupt 0000:00:0e.0[A]: no GSI - using IRQ 5 Aug 25 08:07:53 cc kernel: eth0: RealTek RTL8139 at 0xa800, 00:10:dc:59:bd:dc, IRQ 5 Aug 25 08:07:53 cc kernel: ACPI: PCI interrupt 0000:00:06.0[A]: no GSI - using IRQ 5 Aug 25 08:07:53 cc kernel: 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html Aug 25 08:07:53 cc kernel: 0000:00:06.0: 3Com PCI 3c905B Cyclone 100baseTx at 0xa400. Vers LK1.1.19 Aug 25 08:07:53 cc kernel: via-rhine.c:v1.10-LK1.2.0-2.6 June-10-2004 Written by Donald Becker Aug 25 08:07:53 cc kernel: ACPI: PCI interrupt 0000:00: 05.0[A]: no GSI - using IRQ 11 Aug 25 08:07:53 cc kernel: eth2: VIA Rhine at 0xe3000000, 00:50:ba:a1:86:44, IRQ 11. Aug 25 08:07:53 cc kernel: eth2: MII PHY found at address 8, status 0x7809 advertising 05e1 Link 0000. Aug 25 08:07:53 cc kernel: ACPI: PCI interrupt 0000:00:11.5[C]: no GSI - using IRQ 9 Aug 25 08:07:53 cc kernel: pciehp: acpi_pciehprm:_SB_.PCI0 evaluate _BBN fail=0x5 Aug 25 08:07:53 cc kernel: pciehp: acpi_pciehprm:get_device PCI ROOT HID fail=0x5 Aug 25 08:07:53 cc kernel: USB Universal Host Controller Interface driver v2.2 Aug 25 08:07:53 cc kernel: ACPI: PCI interrupt 0000:00:11.2[D]: no GSI - using IRQ 11 Aug 25 08:07:53 cc kernel: uhci_hcd 0000:00:11.2 : UHCI Host Controller Aug 25 08:07:53 cc kernel: uhci_hcd 0000:00:11.2: irq 11, io base 0000b000
Hi
Yum has just updated 5 of my Linux boxes running CentOS 4.3 to the 2.6.9-42 kernel.
All is well on four of the boxes, but on the fifth the network card (eth0) will not work. I've downgraded the kernel to the previous version and the network card works fine again.
I've tried removing and re-installing the new kernel, but the network card won't work with it.
The details of the card are as follows:
lspci:- 00:0e.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)
/etc/modprobe.conf:- alias eth0 8139too
Any suggestions welcomed..