Peter Relson wrote:

>Shmuel has this fully correct

Indeed. I'm not surprised. ;-D

>This thread was to some extent confusing LLA managing of the directory entry 
>with LLA managing of the module. They are often the same. Not always.

Agreed, this is why locating that errant module is a journey a la 'Lord of the 
Rings'. ;-)

>When LLA is active, BLDL will always ask LLA if it has the directory entry 
>cached. If it does, LLA will return it to BLDL, BLDL will return it to the 
>invoker. Otherwise, BLDL will do find the directory entry itself. There is 
>nothing in the directory entry that could / would / should care whether the 
>data came from LLA or from I/O.

This is what I suspected after lots of RTFM.

>When a fetch is done, the directory entry is used. LLA is asked if, given the 
>directory entry, is LLA managing the data set and member and LLA may indicate 
>"yes, and here is the retrieved module". The concatenation number (K byte) in 
>the directory entry indicates which data set within the concatenation has the 
>module. The K byte is a reason that concatenations cannot have more than 256 
>data sets.

Peter! You have stated it better than those books. I sometimes wish you're the 
author of all those books so we dummies can understand those z/Os mysteries. ;-D

Thanks. I (and I believe perhaps others too) am very grateful for you kind 
explanatory posts.

Will you keep my post under your manager's nose so you will get a good 
promotion? ;-D

Groete / Greetings
Elardus Engelbrecht

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

Reply via email to