Re: [Wg-test-framework] Project Update part 2 part 1 was submitted over the holidays.

[Don Koch <dkoch@xxxxxxxxxxx>]

On Fri, 9 Jan 2015 18:05:38 +0000
Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx> wrote:

> Don Koch writes ("Re: [Wg-test-framework] Project Update part 2 part 1 was 
> submitted over the holidays."):
> > On Fri, 9 Jan 2015 00:01:35 -0500
> > "Paul L. George" <pgeorge@xxxxxxxxxx> wrote:
> > > We are getting to the point where we are going to be done with the 
> > > hardware
> > > setup.  Chris has configured the switches and the PDU's.  We cannot get 
> > > the
> > > VLANS to talk to each other the way Ian wants without some routing on the
> > > main servers.
> > 
> > Which is my understanding. See below.
> 
> Yes, I'm expecting the main servers to do routing.
> 
> > > Newcastle's Internal LAN  is plugged in to the test portion VLAN along 
> > > with
> > > all the test servers we have installed so far.
> > > 
> > > Gateshead's internal LAN is plugged into the infrastructure VLAN, Chris 
> > > did
> > > this so we can see the PDU's it is temporary until we figure out the
> > > interVLAN communications.
> > 
> > Maybe I'm confused, but I thought both were to be connected to the test
> > VLAN. The infrastructure machines are VMs running on the main servers.
> 
> Both servers' internal ethernet ports should be on all three vlans;
> the switch should be configured to provide tagged packets for those
> ports.
> 
> > >  Chris configured the PDU's and the switches according to Ian's plumbing
> > > diagram.  With a level 2 switch you cannot do inter LAN routing.  On the 
> > > HP
> > > switch TAGGED PORTS are not visible from both VLANS. This is the way IAN
> > > envisioned them working.  I can get a different switch but I would need to
> > > know exactly what switch he wanted that had the correct capabilities
> 
> I don't know what you mean by "TAGGED PORTS are not visible from both
> VLANS".  I would expect a tagged port to trunk all the vlans.
> 
> The servers' internal ports need to be configured so that the
> appropriate vlan is enslaved onto each relevant bridge.
> 
> > > Don, did you figure out what we had to do with software to fix the
> > > Comtrol issues.  I did not swap the cables because I did not want
> > > to take the chance to screwing up the only Comtrol components that
> > > were working.
> > 
> > I sent you a reply earlier on this. Last I looked, neither console was
> > available from the other machine. I don't know if this is a Comtrol
> > issue, a wiring issue or something else. Nothing seems to have changed
> > as far as the configuration is concerned. I note that both machines
> > claimed to have found their respective port sets.
> > 
> > Sounds like I'm going to need to get access to the colo.
> 
> Oh dear, it sounds like we are going backwards.
> 
> > >  The DHCP LAN on the TEST VLAN has to be setup.
> 
> That will have to be done after the switch and host config is working,
> I think.
> 
> 
> > I still need to allocate the disks and create said VMs. I was waiting
> > for Ian to confirm/update the disk/memory/CPU sizes/counts for each of
> > these. (If he did, I missed the mail. I recall seeing notes about a couple
> > of them.)
> 
> I'll do this.  Which of the various files in the git repo should I be
> looking at ?

setup-info.ods (or .xls).

> How are the PDUs ?  Are they set up ?  I was expecting to set them up
> via their serial interface but if the serial cards aren't working then
> we're blocked.

Looks like they are set up (right, Paul?). I don't know if they hooked up the
serial interfaces or not, but the Comtrol issue is blocking in any case.

> Thanks,
> Ian.

Thanks,
-d


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