On Fri, 28 Dec 2007 13:50:54 -0600, Tom Marchant wrote:
>On Fri, 28 Dec 2007 13:12:24 -0600, Rick Fochtman wrote:
>
>>IIRC, SYS1.NULLFILE was recognized by the system as a special case of
>>DUMMY. But it filled in the DSNAME field of the JFCB.
>
>You are thinking of DSN=NULLFILE. Now we also have PATH=/dev/null.
>
Which was a grievous and irresponsible and unnecessary design
blunder. Who benefits from this special treatment of PATH=/dev/null?
It only adds confusion by making the behavior of PATH=/dev/null
different from the behavior of PATH=/./dev/null and from the
behavior of PATH=[any symbolic link to /dev/null].
Will RACF deal with single-level DSNAMEs? The system ought by
default be configured to give UACC=ALL to NULLFILE.
Now I must try:
DSN=NULLFILE,DISP=(,CATLG) and
DSN=NULLFILE,DISP=(OLD,DELETE)
An old-timer colleague was surprisingly recently recently unwilling
to use DSN=NULLFILE because in his experience the system issued an
ENQ for NULLFILE. Even if it was a bug that was rapidly fixed,
he was once burned and forever shy.
-- gil
----------------------------------------------------------------------
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
