If the library is in extents at the time of the IPL, there is no problem.  When 
the linklist DEB is constructed, the extents will be handled properly.

A problem can occur if linklist library goes into extents while the linklist is 
active.  The additional extent is not added to the DEB.  If an attempt is made 
to access any data in the new extent, an extent violation occurs and the load 
module will not be loaded.

The issue is not unique to linklist libraries.  Any active library, such as the 
STEPLIB of a long running task, can have the same problem if the task attempts 
to access a member from an extent that is not part of the existing DEB.  

Things to keep in mind:

     The DEB is attached to a DCB.  So two programs accessing the same dataset 
via their own DCBs can have different DEBs which means one program could access 
the new extent will a different one could fail.

     The access methods automatically update the DEB for output DCBs that grow 
into a new extent.

     If you replace a member of a PDS with a new version of the same size, the 
new data goes at the end of any existing data (and may be in a new extent).  If 
you then compress the PDS, everything will move back into the original extents 
and the new extent will be empty.  If the PDS is in the linklist, you can 
refresh LLA so it has the correct TTRs for all the members (the compress 
changed any that followed the replaced member).  THIS PROCESS CAN CAUSE OTHER 
PROBLEMS THAT HAVE BEEN DISCUSSED HERE PREVIOUSLY.

> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]]
> On Behalf Of Jake anderson
> Sent: Sunday, August 10, 2014 4:08 AM
> To: [email protected]
> Subject: Re: Extents more than One for load modules library
> 
> Bob
> 
> Yes they are part of linklist. Before and after IPL there extents were 2
> and 4.

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

Reply via email to