I'd infer that, based on all the conversation so far, you have a real CTCA
at 2030. For this exercise, just totally forget about the real device; it
has no bearing on the outcome.
What you're trying to do is connect your first-level TCPIP machine with your
first-level zVM guest machine. Both virtual machines need to have a virtual
CTCA defined; it really doesn't matter what the addresses are, but the TCPIP
virtual CTCA will need to be defined in the TCPIP configuration. You do this
in both cases with the DEFINE command, or with the CP Directory SPECIAL
statement.
Once the two virtual machines are logged in, you need to do a COUPLE command
on one of them, to "connect the ends" of the two virtual CTCA devices.
When the second-level zVM is up, the first-level virtual CTCA will look and
act like a real CTCA. You'll attach the device to the TCPIP running in the
second-level zVM. You can plumb the second-level Linux in any way you like.
The real fun comes when you convert to vSwitch (and you really should).
Instead of CTCAs connecting specific virtual machines, you add virtual NICs
to each, and connect them all to the vSwitch as a "backbone". To the
second-level zVM, the "physical" connection will look like an OSA card. Once
you get your mind around it, the plumbing is infinitely simpler than CTCA or
IUCV connections to and between the Linux images and TCPIP.
The best advice is to draw a picture (or convert the one Kris is sending you
to your configuration). Label each connection as physical or virtual, and
with the commands needed to utilize it (ATTACH and DETACH for real CTCAs,
DEFINE, COUPLE and DETACH for virtual CTCAs, giving the destination for
each). Each line should have text at both ends, because the commands are
possibly different at each end; virtual addresses may be different (but not
necessarily so), and the name of the userid(s) involved will be different.
Also add to this diagram the TCP/IP addresses you expect each interface to
be seen as, and the mask to be used.
Having a good picture, both in your mind and on paper, of what you're doing
will go a long way to understanding how to plumb it all up.
--
.~. Robert P. Nix Mayo Foundation
/V\ RO-OE-5-55 200 First Street SW
/( )\ 507-284-0844 Rochester, MN 55905
^^-^^ -----
"In theory, theory and practice are the same, but
in practice, theory and practice are different."
On 5/23/07 7:37 AM, "Anne Crabtree" <[EMAIL PROTECTED]> wrote:
> OK, I apologize for being so dumb about this but it's really confusing!
>
> I am trying to set up a connection from 2nd level VM to first level VM
> and was using the virtual CTC to do so. The reason I am doing this is
> so that I can try to start TCPIP on 2nd level which I have not done
> yet.
> Once TCPIP on 2nd level is running, I wanted to start a Linux guest.
> On 2nd level TCPIP, I was going to define another vctc to connect to the
> linux on second level. All this is to verify that maintenance I put on
> 2nd Level VM will not affect Linux before moving that maintenance to
> first level where there are productional linuxes running...
>
> So now, I'm not exactly sure where I am in that process even after all
> this advice.
> I detached the vctc from an linux guest on first level that is not
> being used.
> I think it has been defined on 1st level to my second level VM
> (VMTEST).
> I just IPL'd VMTEST and issued the CP SEND CP VMTEST DEF CTCA 2030
> (and same for 2031)
> If I query 2030 it says: CTCA 2030 FREE
> TCPIP is not running on 2nd level. I've never even tried to start it
> because I'm still trying to get verification from someone who knows
> TCP/IP (although he doesn't know VM) that I coded the IP addresses for
> 2nd level correctly. When you said :
> CP SEND CP VMTEST COUPLE 2030 guest2 vaddr2
> I have no idea what to put for guest2 or even if I need it! And I
> can't attach to TCPIP there since it's not even running yet. I'm
> assuming what I've done so far means 2nd level is attached to 1st level
> though that virtual ctc but I wouldn't swear on it.
>
> I'm sure you can see my confusion.
