Barbara, > I am always willing to shameface incompetence, though.
I always like your posts but the above made my day - well said. Rob Scott Lead Developer Rocket Software 275 Grove Street * Newton, MA 02466-2272 * USA Tel: +1.617.614.2305 Email: [email protected] Web: www.rocketsoftware.com -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Barbara Nitz Sent: 14 July 2010 08:54 To: [email protected] Subject: Re: Automatic SVCDUMP Dataset space allocation -Query The actual space required when the dump is allocated is determined dynamically by DUMPSRV address space. The customer doesn't have influence over it (other than telling the system which dataclass to use which in turn will lead to right volumes that hopefully have enough free space). Whenever you see a *partial* dump, there is a reason code *why* that dump was partial. And there are tons of reasons why a dump might be partial other than dump data set too small. So first of all, get a copy of that dump and use IPCS to see the reason codes (yes, they're in there.) Or take a look at the syslog when the dump was taken and check the iea611i message. I would not put it past *any* vendor (including IBM) to NOT having properly analyzed the reason codes themselves and hence coming back with total bullshit. The most common reason for a partial dump these days is the very old, very inappropriate MAXSPACE default which is still 500M. This governs the size of the dump dataspace, and if that is filled before the dump is fully captured, then you'll get a partial dump. If OMVS is involved, 500M if definitely not large enough, depending on what you need to dump. So if you haven't changed that default to something else, this might be your problem. Which is speculation without the actual code for the partial dump. Regards, Barbara Nitz ps: My "favourite" bullshit is having a vendor tell me that 'storage is not in dump' because they are too stupid to realize that the address is in another address space. (And then request another, "complete" dump.) So they look very incompetent when *I* can see the storage but they cannot. I am always willing to shameface incompetence, though. The next one is 'storage not in dump' when we have a standalone dump and I can prove that the actual address isn't getmained, so *cannot* be in dump. I hate having to play these games! ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
