Based on the E37, CNMLINK already has multiple extents.  When you reallocate it 
on your target volume, give it a larger secondary quantity.  You may also need 
to adjust the primary quantity.

NUCLEUS cannot tolerate extents.  It needs to be large enough to accept any 
possible single SYSMOD.  There is an SMPE option to automatically compress a 
PDS that experiences a D37 and retry the action.  Using this option, when a 
subsequent SYSMOD runs out of space, the PDS is compressed and hopefully the 
recovered space (gas created by previous SYSMODs) is sufficient to accommodate 
the new SYSMOD.

It is not a popular approach but LNKLST PDSes work perfectly well if they 
occupy multiple extents ***at the time you IPL***.  There is no problem unless 
a PDS grows into another extent while it is use.  But this should not even be 
necessary.  When it is time to copy your target PDSes to production, you can 
see how much space each occupies and reallocate any production PDS that is too 
small before you perform the copy.

> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]] On
> Behalf Of venkat kulkarni
> Sent: Wednesday, July 22, 2015 2:11 AM
> To: [email protected]
> Subject: Library out of space issue while APPLY RSU
> 
> Hello Group,
>                   I am applying RSU on z/OS 1.13 system and encountered
> D37-04 and E37-04 for various datasets like
> 
> D37-04  SYS1.SASMMOD1  ---  is used by XCFAS , LLA,
> D37-04  SYS1.SISPLOAD   ---  is used by XCFAS , LLA,
> D37-04  SYS1.SGIMLMD0  ---  is used by XCFAS , LLA,
> D37-04  SYS1.NUCLEUS   ---
> D37-04  SYS1.LINKLIB       ---  is used by XCFAS , LLA,
> E37-04  SYS1.CNMLINK   ---  is used by XCFAS , LLA, NETVIEW
> 
> 
> Now, renaming the dataset to .old and then  create new dataset big big size
> and then using ISPF option 3.3 for copy,   we have only option left is
> 
> 1) Unallocate linkist (SETPROG LNKLST,UNALLOCATE) and then stop LLA ( P LLA)
> 
> 2) For CNMLINK dataset, I will have to stop NETVIEW as well along with
> Unallocate linkist (SETPROG LNKLST,UNALLOCATE) and then stop LLA ( P LLA)
> 
> 
> 3) But for NUCLEUS dataset, I will have to bring down alll system in
> sysplex and then do remaining and creating bigger dataet and then copy
> content using ISP 3.3
> 
> 
> 
> Do we have any other way to overcome this issue rather then copping LLA and
> XCFAS etc.
> 
> 
> Regards
> Venkat
> 
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN

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

Reply via email to