[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference orarep::nomahs::rdb_60

Title:Oracle Rdb - Still a strategic database for DEC on Alpha AXP!
Notice:RDB_60 is archived, please use RDB_70..
Moderator:NOVA::SMITHISON
Created:Fri Mar 18 1994
Last Modified:Fri May 30 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:5118
Total number of notes:28246

4962.0. "Exception at S0 address" by M5::LWILCOX (Chocolate in January!!) Sat Jan 25 1997 14:46

From time to time we've seen bugchecks with exception at 

nnnnnnnn : S0 Address + nnnnnnnn

Does the S0 address itself have any significance?  i.e., this will happen
under condition xyz?  Or is it somewhat meaningless?

Thanks.

Liz
T.RTitleUserPersonal
Name		DateLines
4962.1system spaceHOTRDB::LASTOVICAIs it possible to be totally partial?Sat Jan 25 1997 17:114
    an S0 indication means that the address is in VMS system space (those
    addresses at 80000000 and above).  You can generally use SDA on the
    system where the failure occured to determine what code in VMS contains
    the offending address.
4962.2M5::LWILCOXChocolate in January!!Mon Jan 27 1997 10:1013
<<< Note 4962.1 by HOTRDB::LASTOVICA "Is it possible to be totally partial?" >>>
                               -< system space >-

>>    an S0 indication means that the address is in VMS system space (those
>>    addresses at 80000000 and above).  You can generally use SDA on the
>>    system where the failure occured to determine what code in VMS contains
>>    the offending address.

Thanks, Norm.  To go a little farther, would it be accurate or inaccurate
to say then that the exception was not the result of an Rdb failure, but
that VMS had some problem and Rdb's reaction to that VMS problem was a bugcheck?

Liz
4962.3HOTRDB::LASTOVICAIs it possible to be totally partial?Mon Jan 27 1997 10:556
    I don't think that explaination would be correct in all cases.  You
    need to determine where and why the accvio happened to have any idea if
    it is an Rdb problem or a VMS problem.  For example, if the accvio
    happened in DECCRTL (perhaps installed /RESIDENT on an Alpha system)
    and Rdb had passed a bogus parameter to some C function, then it would
    likely be an Rdb bug.
4962.4Thanks!M5::LWILCOXChocolate in January!!Mon Jan 27 1997 11:235
<<< Note 4962.3 by HOTRDB::LASTOVICA "Is it possible to be totally partial?" >>>

Thanks, Norm, that's what I needed to know.

Liz