On 09/07/2014 06:17 AM, Shmuel Metz (Seymour J.) wrote:
> In <[email protected]>, on 09/05/2014
>    at 02:45 PM, Ed Gould <[email protected]> said:
>
>> I guess we will have to disagree on what holds can be bypassed.
> What he wrote was "You should never (unless told by IBM who probably
> never will)   BYPASS error holds." Dropping the word "error" changes
> its meaning drastically.
>
>> In summary I think its OK to bypass *SOME* hold errors
> What do you mean by "hold errors"? The OP is referring to error holds,
> and you have not given a case where it is advisable to bypass an error
> hold. That has nothing to do with bypassing, e.g., DOC.
>  
There were rare occasions where we bypassed a specific ERROR hold:

* where we had a known exposure to a problem which had actually affected
our production system adversely and the only available resolving PTF
also had an ERROR hold, but its potential error was less of an exposure
than the other problem fixed by the PTF in error.

* where the ERROR hold sounded like an issue that didn't apply to us and
prevented ACCEPT of maintenance prior to a new round of maintenance,
where the PTF now in error had been APPLYed and running in production
without problems for many months and some error hold  had been issued
after the previous round of maintenance.  In this case we had empirical
evidence the ERROR was not a serious exposure for us and we preferred to
expend our resources to moving and testing a new maintenance level
rather than resolving a non-issue.  It made sense to get the
distribution libraries to a level that would allow a potential RESTORE
to return to a level that had successfully run without failure in our
environment, even if IBM now regarded one of those modules in error.

But, bypassing an arbitrary ERROR HOLD without understanding the APAR
involved or its implications for your installation is always a bad idea.

-- 
Joel C. Ewing,    Bentonville, AR       [email protected] 

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

Reply via email to