(Resolved): ORA-00600 errors after upgrade to 11g

My favorite reply from someone was "Good Luck!"....

In case anyone is interested in the solution, here's what happened. We upgraded our production database to 11g in March, after having upgraded 5 other test databases starting last November. About a week or 2 after the upgrade, we started getting ORA-00600 and ORA-07445 messages...20-60 over the course of about an hour, about once or twice a week. We never had complaints from users and we never found any data corruption. We weren't able to replicate the problem in any of our test databases. We opened about 10 SRs with Oracle and the most common suggestion was to get rid of our invalid objects.

At the same time, we found that we were hitting a different Oracle bug, 10213073...creating private synonyms for objects that already had public synonyms resulted in invalid objects. In our case, the result was thousands (9000+) of invalid objects, not just a few. Since we create private synonyms on a regular basis, our database seemed to always have tons of invalid objects.

We were able to apply the patch for the bug about a month ago and haven't had an ORA-00600 or ORA-07445 error since! I love it when the fix for one problem resolves another one, too!

• Maureen

Andre van Winssen wrote:
> Hi Maureen, if your ORA-00600 [pevm_INSTC2] happens in plsql code that
> passes REF cursor parameters up multiple levels
> of the PL/SQL stack then what you see might be a revival of bug 6068126.
>
> Andre
>
> -----Oorspronkelijk bericht-----
> Van: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org]
> Namens Maureen English
> Verzonden: zondag 20 maart 2011 18:50
> Aan: Oracle-L_at_freelists.org
> Onderwerp: ORA-00600 errors after upgrade to 11g
>
> Hi,
>
> Our 11g upgrade seemed to go pretty well, until we started getting an
> overwhelming number of trace files with ORA-00600 and ORA-07445 messages.
>
> These trace files all seem to be related to our self service web
> application,
> but we're unable to find any helpful information and so far, Oracle Support
> is
> still at the information gathering stage.
>
> Here's an example of the messages...any ideas/suggestions regarding
> what's going on would be most appreciated!
>
> DDE: Problem Key 'ORA 600 [pevm_INSTC2]' was flood controlled (0x6)
> (incident: 229161)
> ORA-00600: internal error code, arguments: [pevm_INSTC2], [], [], [],
> [], [], [], [], [], [], [], []
> *********START PLSQL RUNTIME DUMP************
> ***Got internal error Exception caught in pl/sql run-time while running
> PLSQL***
> ***Got ORA-600 while running PLSQL***
> PACKAGE BODY BANINST1.BWSKALOG:
> library unit=c00000021e2fc2f8 line=132 opcode=0 static
> link=9fffffffbeb4ac20 scope=1
> FP=9fffffffbeb4da60 PC=c00000022aebd5a0 Page=0 AP=9fffffffbeb4ac20
> ST=9fffffffbeb4da70
> DL0=9fffffffbf362238 GF=9fffffffbf362ce8 DL1=9fffffffbf362780
> DPF=9fffffffbf362cd0 DS=c00000022ae9f5e8
> DON library unit variable list instantiation
>
>
> and
>
> *** 2011-03-17 17:49:49.563
> *** SESSION ID:(2308.62935) 2011-03-17 17:49:49.563
> *** CLIENT ID:() 2011-03-17 17:49:49.563
> *** SERVICE NAME:(mydb.alaska.edu <http://PROD.alaska.edu>) 2011-03-17
> 17:49:49.563
> *** MODULE NAME:(httpd_at_ssbserver.alaska.edu
> <mailto:httpd_at_veleska.alaska.edu> (TNS V1-V3)) 2011-03-17 17:49:49.563
> *** ACTION NAME:() 2011-03-17 17:49:49.563
>
> Exception [type: SIGSEGV, Address not mapped to object]
> [ADDR:0x6100000000000001] [PC:0x400000000623F9F1, opipls()+9521] [flags: 0x0
> , count: 1]
> Incident 229601 created, dump file:
> /ORACLE/home/admin/diag/rdbms/mydb/MYDB/incident/incdir_229601/MYDB_ora_8996
> _i229601.trc
> ORA-07445: exception encountered: core dump [opipls()+9521] [SIGSEGV]
> [ADDR:0x6100000000000001] [PC:0x400000000623F9F1] [Address not
> mapped to object] []
>
>
> *** 2011-03-17 17:50:19.129
> ssexhd: crashing the process...
> Shadow_Core_Dump = PARTIAL
>
>
>
> - Maureen
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>

--
http://www.freelists.org/webpage/oracle-l