Thanks Kris this is what I was looking for. I am setting this all up and
did have some problems with RACFSMF which gave me data on both 301 and
302 because of a SWITCH. So I wanted to clear the alternate one to test
out my changes to RACFSMF. I certainly LINK to other users all of the
time but was not sure about if I would hurt anything by LINKING to the
302 disk in write mode and erasing from another user that was the real
question/worry. 

 

Thank You,

 

Terry Martin

Lockheed Martin - Citic

z/OS and z/VM Performance Tuning and Operating Systems Support

Office - 443 348-2102

Cell - 443 632-4191

 

From: The IBM z/VM Operating System [mailto:[email protected]] On
Behalf Of Kris Buelens
Sent: Wednesday, March 10, 2010 2:56 AM
To: [email protected]
Subject: Re: ERASE SMF DATA on ALTERNATE DISK

 

If RACFSMF failed, then you can indeed have data on 301 and 302.  When
the active SMF disk gets full, RACF will not be able to switch and SMF
recording stops.
So, we created REXX code for our system automation to check for this
situation.

You could try a simple XAUTOLOG RACFSMF and see if that cleans the
situation.  Otherwise: from any user with RACF permission to LINK to
RACFVM 301 or 302, you can issue
   CP LINK RACFVM 30x 1111 M
   ACCESS 1111 Z
  and do what you like, such as ERASE...
<q>
Q1. Why do some many VM users continuously log ON and OFF do switch
userids, where a simple LINK from their own user could be enough?
Q2 And, when a LOGON is required (or the best after all) why don't they
use LOGON BY, but try they to remember the passwords of all these
service machines?

Since I'm back in the field for many customers I have to explain this
over and over again.  Even without RACF, there is a LOGON BY capability.
</q>

2010/3/10 Alan Altmark <[email protected]>

On Wednesday, 03/10/2010 at 12:11 EST, "Martin, Terry R. (CMS/CTR)
(CTR)"

<[email protected]> wrote:

> I want to ERASE the SMF DATA file from the ALTERNATE disk(302 in my
case). How
> can I do this without logging on to the RACFVM Server. Can I do this
from the
> RACFSMF machine?

RACFSMF already does that after it archives the file to the 192 disk.
RACFSMF is designed to be autologged at the same time every day (e.g.
midnight).  Its behavior is governed by the settings in the PROFILE EXEC
(SMFPROF EXEC).

But, in general, an SMSG RACFVM SMF SWITCH causes RACF to close the
active
SMF log and detach the disk it is on.  You have to look at the SMF
CONTROL
file on RACFVM's 191 in order to know which disk is CURRENT so you know
which disk to go after.

Alan Altmark
z/VM Development
IBM Endicott




-- 
Kris Buelens,
IBM Belgium, VM customer support

Reply via email to