Re: [Wg-test-framework] Lenovo issue

[Don Koch <dkoch@xxxxxxxxxxx>]

On Wed, 22 Apr 2015 08:14:34 -0400
Ian Campbell <ian.campbell@xxxxxxxxxx> wrote:

> On Tue, 2015-04-21 at 14:03 -0400, Don Koch wrote:
> > On Tue, 21 Apr 2015 13:33:52 -0400
> > Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx> wrote:
> > 
> > > Don Koch writes ("Re: [Wg-test-framework] Lenovo issue"):
> > > > On Tue, 21 Apr 2015 12:13:50 -0400
> > > > > Can you please confirm:
> > > > > 
> > > > >  * Which physical ports on the machine AMT has been set up on ?
> > > > >    AIUI the machines have three ports: eth0, eth1 and the management
> > > > >    port.  I think we should be using the management port.
> > > > 
> > > > AIUI: the management port is shared with the machine currently showing 
> > > > up as eth2.
> > > 
> > > You mean the back of the machine only has two physical ethernet
> > > ports ?
> > 
> > Yes, there are three: two with sequential MAC addresses (which 
> > coincidentally show
> > up as eth0 and eth1) and one with an out-of-sequence MAC addr (eth2 and the
> > AMT port).
> > 
> > Sorry for the confusion.
> > 
> > > Is it possible to dedicate eth2 to AMT ?  I am quite unhappy with port
> > > sharing which seems likely to cause all sorts of unwanted excitement.
> > 
> > The only reason that the AMT landed on the same IP addr as the machine
> > is that eth1 (the "normal" port) and eth2 (the AMT port) both have the
> > same IP address. You should be able to fix this in your DHCP server
> > settings. (I did this for the huxelrebe1 AMT port to put it on a
> > different address than the normal linux port.)
> > 
> > > > For whatever reason, the normal ports are on eth1. Eth0 is unassigned.
> > 
> > Actually, this is because the ethernet cable is plugged into port 1 instead 
> > of
> > port 0 (not that they're labelled - in fact, *because* they aren't 
> > labelled).
> 
> This is all very confusing naming-wise, and I at least have no idea what
> the configuration is right now. Please lets avoid names such as "eth<N>"
> for the purposes of this conversation.
> 
> Also, since the chassis is apparently not labelled I don't think it
> helps to talk about what things might be labelled if they were. Perhaps
> you could take a photo and annotate it up with some names for clarity.
> 
> Please can you confirm:
> 
>      1. That the physical port associated with the "management
>         interface" Ethernet port is wired into the switch.

Yes.

>      2. That the first physical non-management Ethernet port is also
>         wired into the switch.

No, the second one is (based on MAC addresses).

>      3. The precise MAC address associated with both of those, in those
>         terms (not in terms of the names Linux might have chosen, nor
>         whatever might or might not be stencilled on the case)

For huxelrebe0:
  port 0: 44:39:c4:6d:20:4d
  port 1: 44:39:c4:6d:20:4e
  port 2 (mgmt): 44:39:c4:39:a1:26

For huxelrebe1:
  port 0: 44:39:c4:6d:20:5f
  port 1: 44:39:c4:6d:20:60
  port 2 (mgmt): 44:39:c4:39:a1:19

>      4. That the BIOS is configured such that:
>              A. all AMT/lightsout/etc management traffic uses the
>                 management interface Ethernet port;

Yes.

>              B. the management port is not visible to the Operating
>                 System.

No. It shows up as a third ethernet device (coincidentally as eth2 on
both). I don't think it can be turned off. I would hope that this
shouldn't matter.

>              C. the non-management port is configured solely for the use
>                 of the Operating System and will never be interfered
>                 with by any BIOS activity;

Yes. (s/port is/ports are/)

> Ian.

-d

_______________________________________________
Wg-test-framework mailing list
Wg-test-framework@xxxxxxxxxxxxxxxxxxxx
http://lists.xenproject.org/cgi-bin/mailman/listinfo/wg-test-framework