On Fri, Feb 12, 2016 at 9:33 AM, Ed Gould <[email protected]> wrote:

> On Feb 12, 2016, at 2:36 AM, Vince Getgood wrote:
>
>>
>> Ed,
>> That’s more or less what John said.  I thought the WAIT parameter would
>> resolve that issue though.
>>
>> On more investigation, there is nothing in the joblog for the HSMLIST
>> step regarding the final disposition of 'MAINT.TEMP.LIST', regardless of if
>> I run it on it's own, or as part of the two step job coded in my OP.
>>
>> This lends credence to what Ed and John are saying.  Even though I've
>> coded WAIT on the DFHSM LIST command, the HSMLIST step is finishing and the
>> SORTIT step starts before 'MAINT.TEMP.LIST' is allocated, hence the
>> ARC0141I message and the JCL error.
>>
>> Looks like I may have to do two jobs, or put some sort of "filler" step
>> between the HSMLIST & SORTIT steps.
>> Vince.
>>
>> It looks (to me) that you may have to get a scheduling package involved
> and let it figure out when the DSN is created and then run your second part
> of the job.
>

​We do that a lot, especially for "outside" files create by things like ftp
uploads. Unfortunately, that "ability" is a part of OEM software and not
generic. I wish that z/OS had an API (similar to Linux' inotify) to which
you could register a "handler" to be notified when a DSN or DSN "pattern"
name was CREATED, OPENed (& whether INPUT, OUTPUT, or UPDATE), or DELETED.
And no, I don't mean using the IEFUxx SMF exits to examine the SMF records
and DIY up something.​ That's basically what we have with CA-7's "external
dataset" ability.



>
> Ed
>
>

-- 
The man has the intellect of a lobotomized turtle.

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to