[7.180] Realtek 8169 driver not updated - GBLAN not working

[pbrdodk101]

The Jetway 3xGBLAN daughterboard (J2AD3RTLANG) is still not working.
ASG recognises the additional NICs but there is no traffic going through them - as before. Given that 8169 is supposed to be supported I hoped that this would have been fixed in this release.

[HolgerE]

Please provide output of 'lspci -vvx' and 'lspci -vvxn'. Thanks.

[pbrdodk101]

sorry for the delay ... busy week
both outputs attached. let me know if you need anything else.

[pbrdodk101]

does anyone know if this problem going to be looked at in this release? next? ever?

[grgcombs]

This issue is why I opted to go back to 100BT with Jetway. Extremely irritating ... It looked like it was going to work, with all the appropriate link lights and such, but no amount of driver coercion was effective.

[pbrdodk101]

tell me about it. very annoying, esp since HCL states that it should be supported.

[jzkkn5]

This looks like the same issue I have. As ou can see by this repley the driver in Astaro seems to not include the updates

00:09.0 0200: 10ec:8167 (Rev 10)

OK, so it looks like the r8169 driver needs to be updated. The number of device ids supported by the kernel on my Fedora 8 desktop is much higher than the one on Astaro v7.101 and includes your NIC.

Fedora 8:

Code:
# modinfo r8169
filename: /lib/modules/2.6.23.9-85.fc8/kernel/drivers/net/r8169.ko
version: 2.2LK-NAPI
license: GPL
description: RealTek RTL-8169 Gigabit Ethernet driver
author: Realtek and the Linux r8169 crew <netdev@vger.kernel.org>
srcversion: 5BFB3849E45989130EA697D
alias: pci:v00001737d00001032sv*sd00000024bc*sc*i*
alias: pci:v000016ECd00000116sv*sd*bc*sc*i*
alias: pci:v00001259d0000C107sv*sd*bc*sc*i*
alias: pci:v00001186d00004300sv*sd*bc*sc*i*
alias: pci:v000010ECd00008169sv*sd*bc*sc*i*
alias: pci:v000010ECd00008168sv*sd*bc*sc*i*
alias: pci:v000010ECd00008167sv*sd*bc*sc*i*
alias: pci:v000010ECd00008136sv*sd*bc*sc*i*
alias: pci:v000010ECd00008129sv*sd*bc*sc*i*
depends:
vermagic: 2.6.23.9-85.fc8 SMP mod_unload
parm: rx_copybreak:Copy breakpoint for copy-only-tiny-frames (int)
parm: use_dac:Enable PCI DAC. Unsafe on 32 bit PCI slot. (int)
parm: debugebug verbosity level (0=none, ..., 16=all) (int)Astaro 7.101:

Code:
# modinfo r8169
filename: /lib/modules/2.6.16.43-54.5-default/kernel/drivers/net/r8169.ko
version: 2.2LK-NAPI
license: GPL
description: RealTek RTL-8169 Gigabit Ethernet driver
author: Realtek and the Linux r8169 crew <netdev@vger.kernel.org>
srcversion: 32FF39AE2A46DED2640AEAA
alias: pci:v00001737d00001032sv*sd00000024bc*sc*i*
alias: pci:v000016ECd00000116sv*sd*bc*sc*i*
alias: pci:v00001186d00004300sv*sd*bc*sc*i*
alias: pci:v000010ECd00008169sv*sd*bc*sc*i*
depends:
vermagic: 2.6.16.43-54.5-default 686 REGPARM gcc-4.1
parm: media:force phy operation. Deprecated by ethtool (8). (array of int)
parm: rx_copybreak:Copy breakpoint for copy-only-tiny-frames (int)
parm: use_dac:Enable PCI DAC. Unsafe on 32 bit PCI slot. (int)
parm: debugebug verbosity level (0=none, ..., 16=all) (int)
parm: ignore_parity_err:Ignore PCI parity error as target. Default: false (bool)You can see how the r8169 driver for Fedora 8 matches the device ID of your NIC while the r8169 driver for Astaro does not.

Edit: BTW, you can see that support for your NIC was added to Linus' tree in July 2006, here's a link to the patch that added it:
r8169: sync with vendor's driver

Unfortunately, there is no support for this NIC in the 2.6.16 stable branch which could make backporting the changes a bit more difficult.

[Flancer]

I also support the idea of having Astaro support this NIC.

What would happen if Astaro uses a newer kernel?

[Gert Hansen]

Hi there all,

this is, unfortunatly, not as easy as it looks.

We would love to use a newer kernel version, but in terms of stability for our exisiting customer base, we can only offer this with a new major version.

THe problem is, that the driver support can change in both ways, means newer cards get supported, but also other cards not anymore or in a different way.
This way we must du all the compatibility testing again and we can not just upgrade as there might be customers out there where their currently running system would not work anymore.

Once we work on ASG V8, than we will use a newer kernel with improved hardware support.
Until than we do some backports from time to time, if they are feasable.

I hope that is understandable
best regards

Gert

[pbrdodk101]

of course. can you please make this limitation clear in the HCL? HCL states that 8169 is supported and this is not exactly correct.