Many thanks, guys.....your suggestions have gotten me on the right track now, I think. There does appear to be a problem (error) with a DCSS file that this particular vendor product uses. When I go to back up the DCSS in question I get this (from MAINT):
Ready; T=0.01/0.01 10:Ready; T=0.01/0.01 10:49:16 dcssbkup xxxxxx HCPABN9300E A system soft abend NTF001 has occurred on CPU 0000 and the virtual CPU has been put into stopped state. (I've changed the DCSS name to protect the vendor, which might not be a fault here.) This is exactly the same soft abend I see when I try to run the module. I'm going to see if I can't simply purge this DCSS and rebuild it from scratch. Thanks again for the help, appreciate it. DJ ----- Original Message ----- From: Alan Altmark <[email protected]> To: [email protected] Subject: Re: CP soft abend code NTF001 Date: Fri, 23 Jul 2010 11:30:48 -0400 > On Friday, 07/23/2010 at 10:29 EDT, dave > <[email protected]> wrote: > > > I have a CMS module, coded in assembler, that, when run, > > causes a NTF001 soft abend when run from one virtual > > machine, but when run from another virtual machine it > > runs normally. > > > > Both virtual machines are set to XA, both have the same > > amount of virtual storage defined, both have the same > > set of options set, and NUCXMAP shows the same nucleus > > contents in both. What have I overlooked? > > It doesn't have anything to do with CMS, per se. Would it > help to know that HCPNTF is involved in locating and > reading a DCSS or NSS? That would seem to lead me to the > conclusion that something is changing a spool file or > queue without proper serialization. > > Alan Altmark > z/VM Development > IBM Endicott
