[Linuxha-users] Re: Netgear 1Gbps network card and mii-tool

[Michael Mansour]

Hi Simon,

I also asked the same query on the SL list (RHEL people) and the following is 
what they had to say:

"
Well the first thing I'd say is that the realtek chipset's are terrible
if you really want performance. They don't do any packet header
processing or other tcp hardware offloading, causing your CPU to take
the brunt of the packet load (with gigabit this is bad).

I would recommend Broadcom or Intel based gigabit cards.

Regardless of this fact, mii-tool doesn't support reading out gigabit
link status. It'll give you a link up status with 100FD for 1000FD cards
linked at 1000FD with flow control enabled.

I know it at least reports on the broadcom and intel based cards with a
link up status.

Otherwise, I think you might need to find another tool to get real link
status out of a gigabit nic.

P.S. the tool you're looking for is ethtool. 

mii-tool is obsolete.
"

So I'm going to try and get Intel cards and use those. The link UP status is 
enough for linuxha.net to do its monitoring yes?

Michael.

> Hello Michael,
> 	Running Linuxha.net with cards that do not support the relevant 
> hooks for the mii-tool is not a problem... However, if you wish to 
> have IP failover under 0.8.3 and above (when it comes out - soon!) 
> the cards on the network in question are best being supported by mii-
> tool.
> 
> 	In your situation you would get fail-over when physical link 
> problems occurred on the 100Mbit/sec cards, but not on the 1Gbit/sec 
> cards.
> 
> Regards,
> Simon.
> 
> On Mon, 2005-02-14 at 10:09 +1000, Michael Mansour wrote:
> > Hi Simon,
> > 
> > As discussed earlier, I have purchased 2x Netgear Gigabit cards:
> > 
> > http://www.netgear.com/products/details/GA311.php
> > 
> > On boot these show up as:
> > 
> > r8169 Gigabit Ethernet driver 1.2 loaded
> > r8169: PCI device 00:0f.0: unknown chip version, assuming RTL-8169
> > r8169: PCI device 00:0f.0: TxConfig = 0x4000000
> > divert: allocating divert_blk for eth2
> > eth2: Identified chip type is 'RTL-8169'.
> > eth2: RealTek RTL8169 Gigabit Ethernet at 0xxxxxxxxx, xx:xx:xx:xx:xx:xx, 
IRQ 4
> > eth2: Auto-negotiation Enabled.
> > eth2: 1000Mbps Full-duplex operation.
> > 
> > and on the standby node:
> > 
> > r8169 Gigabit Ethernet driver 1.2 loaded
> > PCI: Found IRQ 4 for device 00:09.0
> > r8169: PCI device 00:09.0: unknown chip version, assuming RTL-8169
> > r8169: PCI device 00:09.0: TxConfig = 0x4000000
> > divert: allocating divert_blk for eth1
> > eth1: Identified chip type is 'RTL-8169'.
> > eth1: RealTek RTL8169 Gigabit Ethernet at 0xxxxxxxxx, xx:xx:xx:xx:xx:xx, 
IRQ 4
> > eth1: Auto-negotiation Enabled.
> > eth1: 1000Mbps Full-duplex operation.
> > 
> > Each card has a RealTek chip on it, so the assumption the kernel makes 
above 
> > is correct.
> > 
> > I've given these cards 10.x.x.x subnets, and am able to ping between the 
two 
> > (I've used a cross-over cable to hook each of them up to one another)
> > 
> > However, mii-tools shows me the following:
> > 
> > # mii-tool
> > eth0: negotiated 100baseTx-HD, link ok
> > eth1: negotiated 100baseTx-HD, link ok
> > SIOCGMIIPHY on 'eth2' failed: Operation not supported
> > 
> > Is this an issue when running with linuxha.net?
> > 
> > Thanks.
> > 
> > Michael.
> > 
> > 
> >
------- End of Original Message -------