[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

5012.0. "Rdb 7.0-00: RMU/Unload bugchecks at BLI$CALLG + BC" by AVMSV1::EKREISLE (Erich Kreisler) Mon Feb 10 1997 12:44

Hello,

A customer has problems unloading a view under Rdb 7.0. Whenever he unloads 
a special view a bugcheckdump is written to the screen or disk, depending 
on the circumstances described below. 

First here are the exceptions:
***** Exception at 001B7ABC : BLI$CALLG + 000000BC                              
%COSI-F-UNEXPERR, unexpected system error                                       
-COSI-F-UNEXPERR, unexpected system error
Saved PC =3D 005E0DAC : STD$DUMP_ALPHA_VMS_STACK + 000000DC                     
Saved PC =3D 005BE4C4 : KOD$BUGCHECK_DUMP + 00000D24                            
Saved PC =3D 002EF24C : RMUUNL$NAME_LEN + 000028D4                              
Saved PC =3D 80895D44 : symbol not found                                        
Saved PC =3D E53122BC : symbol not found                                        
Saved PC =3D 001E65EC : COSI_CHF_SIGNAL + 00000164                              
Saved PC =3D 001B7ABC : BLI$CALLG + 000000BC                                    
Saved PC =3D 001E6464 : COSI$CHF_SIGNAL + 0000013C                              
Saved PC =3D 001B7ABC : BLI$CALLG + 000000BC                                    
Saved PC =3D 002F1328 : RMUUNL$NAME_LEN + 000049B0                              
Saved PC =3D 002EBC18 : RMU$UNLOAD + 00000648                                   
Saved PC =3D 002EB1CC : RMUCLI$UNLOAD + 00000E84                                
Saved PC =3D 00207F14 : RMU_DISPATCH + 00000F34
Saved PC =3D 00206FB0 : RMU_STARTUP + 000004A0                                  
Saved PC =3D 0014002C : RMU$MAIN + 0000002C                                     
Saved PC =3D 7EE6C160 : symbol not found 

Versions at customer site: 
	Rdb 7.0-00 and 7.0-01 (prerelease)
	OpenVMS 6.2 with patches (as suggested in 9.166)
Versions here in the office:
	Rdb 7.0-01 (prerelease)
	OpenVMS 6.1


These are the commands the customer issues:
1. small number of buffers
	fails for the customer if he has no indices defined on the biggest
	tables, works for me, no matter if I have indices defined on the 
	table or not. The dump is written to the screen.

   $ rmu/unload/buffer=100/exte=10000/virtual/rms=(file=x.x) dwhs -
	V_SUMWSSTAG disk$data2:[lotto]V_SUMWSSTAG      

2. big number of buffers
	I always get exceeded quota, but I don't have an AXP with 2GB of
	memory. 
	The executes for a while at the customer site, and then bugchecks,
	indices are defined on the main tables. Dump written to disk.

   $ rmu/unload/buffer=10000/exte=10000/virtual/rms=(file=x.x) dwhs -
	V_SUMWSSTAG disk$data2:[lotto]V_SUMWSSTAG

I can't reproduce any of those problems under Rdb 7.0.01 (prerelease) and
OpenVMS 6.1 and the customer database. The dumps will be delivered to me
tomorrow afternoon.
Can this be related to OpenVMS 6.2 ? Any hints are welcome.

Thank you in advance,
	erich
T.RTitleUserPersonal
Name
DateLines
5012.1Another one......M5::BLEHLBACRDB: 34% better than real lifeTue Feb 18 1997 15:1811
Hi,

  Any ideas on this??

  I have a customer with a similar problem.

  Customer is on Rdb7, axp, vms 6.2, trying to unload a table.

  He is going to try the unload on a vax.

                    Thanks,   Barry
5012.2DUCATI::LASTOVICAIs it possible to be totally partial?Tue Feb 18 1997 15:311
If you can reproduce it, please BUG it.
5012.3stay tuned ....M5::BLEHLBACRDB: 34% better than real lifeTue Feb 18 1997 16:054
>>If you can reproduce it, please BUG it.

OK ..... I'll see what I can do.   Thanks!!
5012.4AVMSV1::EKREISLEErich KreislerWed Feb 19 1997 09:049
We tried to reproduce it with the customers systemdisk and database on our small
machine and failed. The customer has an 8400 with 2GB of memory.
The other error that gets also written out to the screen is RMS-F-DME. The
customer checked this with the articles in DSIN, but could not find a solution
yet. On the other hand I filed a bug 452852 for this problem.

re .-1: I hope you can reproduce, we can't do it in every environment.

erich