On 6/26/2007 3:13 AM, Binyamin Dissen wrote:
On Mon, 25 Jun 2007 16:07:16 +0800 Charokee Sun <[EMAIL PROTECTED]>
wrote:
:>Our company recently had been given a task by our client to test the
:>responsiveness and capabilities of their systems programmer in their test
:>environment. Our tasks assigned include to hack their system to
:>cause/simulate system and application outage, of course not to the extent of
:>hanging/re-IPLing the whole system.
:>Just want to know whether anyone out there have done any similar test before
:>and willing to share what they have tested. Btw, we will be provided with
:>powerful TSO user IDs but won't be allowed to touch any system module and
:>probably just to change some control blocks in memory. Somehow, it seems
:>easy to think what you can do in CICS and DB/2 but it's easier said than
:>done in MVS.
:>The systems programmers will be given at least 1-1/2 hour limit to resolve
:>the problem.
Altering some CICS options in memory seems easy enough.
Probably true, but I don't think I would consider that a meaningful
test. What should we expect the system programmer to do other than
(a) recognize that CICS had a problem; and
(b) tell operations to restart it.
CICS does allow changing of some options via commands, and I suppose one
might expect the system programmer to recognize that and correct the
problem by issuing the appropriate commands. But if the "hacker" needed
to change bits in control blocks in storage to cause the problem, I
would not expect the system programmer to figure out which bits and
write a program to fix them. In that situation I would expect a simple
restart of the region.
And I would not expect it to take 1.5 hours to figure this out and get
it restarted.
Walt
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
