Re: RSCS Problem - return code=- 1 error number=54 (Connection reset by peer)

2011-03-27 Thread Alan Altmark
On Saturday, 03/26/2011 at 01:34 EDT, Undetermined origin c/o LISTSERV 
administrator   wrote:

> Thanks for the suggestions but I think there is a fundamental  problem 
with our 
> TCP/IP set up. I tried sending files via FTP and get exactly  the same 
problem, 
> small ones get through but larger transfers hang. Time to get  my 
networking 
> hat on :-)

Look at your MTUs.

Alan Altmark

z/VM and Linux on System z Consultant
IBM System Lab Services and Training 
ibm.com/systems/services/labservices 
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott


Re: Support Element (SE) vs HMC and IP Addresses

2011-03-27 Thread Alan Altmark
On Saturday, 03/26/2011 at 05:06 EDT, George Henke/NYLIC 
 wrote:
> Also, would someone kindly explain the functional differences between 
the 
> Support Element (SE) and the Hardware Management Console (HMC) other 
than the 
> HMC can: 
> 
> do everything an SE can do 
> do it for more than one CEC 
> do it as a portal for the intra/internet   

You've pretty much got it, though "everything an SE can do" is possible 
only via Single Object Operations.  I would also say that the SE's primary 
job is the run the CPC (CEC).  The HMC's primary job is to talk to you.

Alan Altmark

z/VM and Linux on System z Consultant
IBM System Lab Services and Training 
ibm.com/systems/services/labservices 
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott


Re: Support Element (SE) vs HMC and IP Addresses

2011-03-27 Thread Alan Altmark
On Saturday, 03/26/2011 at 05:06 EDT, George Henke/NYLIC 
 wrote:
> Alan Altmark wrote: 
> 
>  
> 
> Our networking people say the SE's are reachable from our corporate 
intranet so 
> that we can provide redundance for  two mainframes, 1000 miles apart. 
> 
> But I suspect they are reachable only through the HMC. 
> 
> If the SE's are connected directly to the corporate intranet,, what 
problems 
> might we expect to have? 

I meant that your SEs are on a private network, wherever they are.  It is 
a Best Practice not to have other network devices on that network.  To 
quote from the z10 Installation Guide:
  If connection to the enterprise LAN is desired, it is recommended that 
an
  Ethernet bridge or router be installed to isolate the Hardware 
Management
  Console and Support Element from other systems.

The SEs are secure, but keeping on an isolated LAN segment is simply 
another layer of protection.

Alan Altmark

z/VM and Linux on System z Consultant
IBM System Lab Services and Training 
ibm.com/systems/services/labservices 
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott


Re: Supporting Dot.1q trunk in z/Linux

2011-03-27 Thread Alan Altmark
On Friday, 03/25/2011 at 11:35 EDT, "Martin, Terry R. (CMS/CTR) (CTR)" 
 wrote:
> Thanks Alan, this is what I was looking for. So the Vswitch and the NIC 
def for 
> the guest does not change correct? It appears that the only thing that 
changes 
> is at the switch level where it becomes a Trunk connection correct?

1.  The switch port will be reconfigured as a trunk
2.  The port will be authorized to one or more VLANs
3.  There is a "default VLAN id" in the switch.
4.  You will add a VLAN to every SET VSWITCH GRANT (or authorize in your 
ESM)

So, you will
DEFINE VSWITCH  VLAN 666 NATIVE nnn 
SET VSWITCH ... GRANT userid VLAN vvv

Where: 666 is some VLAN that is NOT authorized on the trunk.
   nnn is the default VLAN id of the switch
   vvv is the VLAN you want the specified guest to use


Alan Altmark

z/VM and Linux on System z Consultant
IBM System Lab Services and Training 
ibm.com/systems/services/labservices 
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott


Re: updating IPLPARMS to new console address

2011-03-27 Thread Phillip Gramly
I have the address in the SYSTEM CONFIG in Operator_Consoles
but it doesn't use B00 which is the first address.
somehow it is overridden.
I don't get a SAPL screen, it waits until my 3174 terminal is available and use 
it.

ok, so shutdown restart is not a permanent change, but only as a temp change 
followed by
the shutdown restart command?

From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf 
Of Scott Rohling
Sent: Sunday, March 27, 2011 2:37 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: updating IPLPARMS to new console address

You can set the console addresses in SYSTEM CONFIG that z/VM should use... see 
the 'Operator_Consoles' variable.   SET IPLPARMS is for a subsequent SHUTDOWN 
RESTART ..  which it doesn't sound like you did from the way you worded it?

Scott Rohling
On Sun, Mar 27, 2011 at 12:50 PM, Phillip Gramly 
mailto:phil...@cdg.ws>> wrote:
I want to change the IPLPARMS to use an ICC console address instead of a real 
address on our old 3174.
I used the CP command SET IPLPARMS and replied CONS=B00 and got the message 
back:
IPL parameters have been replaced

If I check it with Q IPLPARMS it shows B00
indeed, when I issue SHUTDOWN, the messages start displaying on the OSA-ICC B00 
address TN3270
But, when I restart VM, it IPLs on the old 3174 address.

How do I make the SET IPLPARMS override permanent?

Phillip Gramly
Systems Programmer
Communications Data Group
Champaign, IL



Re: updating IPLPARMS to new console address

2011-03-27 Thread Scott Rohling
You can set the console addresses in SYSTEM CONFIG that z/VM should use...
see the 'Operator_Consoles' variable.   SET IPLPARMS is for a subsequent
SHUTDOWN RESTART ..  which it doesn't sound like you did from the way you
worded it?

Scott Rohling

On Sun, Mar 27, 2011 at 12:50 PM, Phillip Gramly  wrote:

> I want to change the IPLPARMS to use an ICC console address instead of a
> real address on our old 3174.
> I used the CP command SET IPLPARMS and replied CONS=B00 and got the message
> back:
> IPL parameters have been replaced
>
> If I check it with Q IPLPARMS it shows B00
> indeed, when I issue SHUTDOWN, the messages start displaying on the OSA-ICC
> B00 address TN3270
> But, when I restart VM, it IPLs on the old 3174 address.
>
> How do I make the SET IPLPARMS override permanent?
>
> Phillip Gramly
> Systems Programmer
> Communications Data Group
> Champaign, IL
>


Re: updating IPLPARMS to new console address

2011-03-27 Thread Rich Smrcina
Does the load profile for the VM system contain a loadparm that would point to the 
console address that z/VM should use for startup messages?


On 03/27/2011 01:50 PM, Phillip Gramly wrote:

I want to change the IPLPARMS to use an ICC console address instead of a real 
address on our old 3174.
I used the CP command SET IPLPARMS and replied CONS=B00 and got the message 
back:
IPL parameters have been replaced

If I check it with Q IPLPARMS it shows B00
indeed, when I issue SHUTDOWN, the messages start displaying on the OSA-ICC B00 
address TN3270
But, when I restart VM, it IPLs on the old 3174 address.

How do I make the SET IPLPARMS override permanent?

Phillip Gramly
Systems Programmer
Communications Data Group
Champaign, IL





--
Rich Smrcina
Velocity Software, Inc.
http://www.velocitysoftware.com

Catch the WAVV! http://www.wavv.org
WAVV 2011 - April 15-19, 2011 Colorado Springs, CO


updating IPLPARMS to new console address

2011-03-27 Thread Phillip Gramly
I want to change the IPLPARMS to use an ICC console address instead of a real 
address on our old 3174.
I used the CP command SET IPLPARMS and replied CONS=B00 and got the message 
back:
IPL parameters have been replaced

If I check it with Q IPLPARMS it shows B00
indeed, when I issue SHUTDOWN, the messages start displaying on the OSA-ICC B00 
address TN3270
But, when I restart VM, it IPLs on the old 3174 address.

How do I make the SET IPLPARMS override permanent?

Phillip Gramly
Systems Programmer
Communications Data Group
Champaign, IL


Re: Supporting Dot.1q trunk in z/Linux

2011-03-27 Thread Martin, Terry R. (CMS/CTR) (CTR)
Thanks Mary!

Thank You,

Terry Martin
Lockheed Martin
CMS - CITIC
3300 Lord Baltimore Drive, Suite 200, 21244  
Engineering Computing
Mainframe Support
Cell - 443 632-4191



-Original Message-
From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf 
Of Marcy Cortes
Sent: Sunday, March 27, 2011 1:22 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: Supporting Dot.1q trunk in z/Linux

Marcy :)

Your using a trunk port requires the vswitch to "vlan tag" the stuff sent over 
it.

So say you have default vlan of 100 on the vswitch going out over it and your 
vswitch has that defined as the default.
Then if the IP x.x.x.x is on vlan 100 then you don't need a vlan on the grant.
If the IP x.x.y.x is on vlan 200 then you will need a VLAN parm on the vswitch 
grant.
Like set vswitch vswitch1 grant Linux8 vlan 200.
I kind of feel it's safer to always put a vlan number on any grant on a vswitch 
using a trunk port.

Symptoms of getting it wrong will be no packets going anywhere on a 'q vswitch 
details"




Marcy 



-Original Message-
From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf 
Of Martin, Terry R. (CMS/CTR) (CTR)
Sent: Saturday, March 26, 2011 10:06 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: [IBMVM] Supporting Dot.1q trunk in z/Linux

Mary,

What would be the LAN name on the GRANT? And is the GRANT the place I would 
need to specify LAN parameter? 

Thank You,

Terry Martin
Lockheed Martin
CMS - CITIC
3300 Lord Baltimore Drive, Suite 200, 21244  
Engineering Computing
Mainframe Support
Cell - 443 632-4191



-Original Message-
From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf 
Of Marcy Cortes
Sent: Saturday, March 26, 2011 10:34 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: Supporting Dot.1q trunk in z/Linux

Right, your vswitch and nicdef statements don't change.You may have to put 
a vlan on your SET VSWITCH GRANT statements.

Marcy 
-Original Message-
From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf 
Of Martin, Terry R. (CMS/CTR) (CTR)
Sent: Thursday, March 24, 2011 8:05 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: [IBMVM] Supporting Dot.1q trunk in z/Linux

Thanks Alan, this is what I was looking for. So the Vswitch and the NIC def for 
the guest does not change correct? It appears that the only thing that changes 
is at the switch level where it becomes a Trunk connection correct?

Thank You,

Terry Martin
Lockheed Martin
CMS - CITIC
3300 Lord Baltimore Drive, Suite 200, 21244  
Engineering Computing
Mainframe Support
Cell - 443 632-4191


-Original Message-
From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf 
Of Alan Altmark
Sent: Thursday, March 24, 2011 3:29 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: Supporting Dot.1q trunk in z/Linux

On Thursday, 03/24/2011 at 01:01 EDT, "Martin, Terry R. (CMS/CTR) (CTR)" 
 wrote:

> So my question is from my z/Linux side what changes to I need to make 
and where 
> are these changes reflected In my setup on the z/Linux. Just so you know 
my 
> z/VM TCP/IP stack is not really being used for the z/Linux guests that 
is being 
> done within z/Linux in my case RHEL 5.2  

You will make NO changes to your Linuxes.  Those on existing subnet A will 
remain there and be on VLAN "A".  New ones on Subnet B will be assigned to 
VLAN "B".  When a Linux on A and one on B want to talk to each other, the 
traffic will flow into the VSWITCH, down the OSA, out of the box, into the 
router, and back.

This may generate a requirement to reassign IP addrs so that those two 
servers are placed in the same subnet, with no need for traffic to flow 
out of the box.

Alan Altmark

z/VM and Linux on System z Consultant
IBM System Lab Services and Training 
ibm.com/systems/services/labservices 
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott