Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> Re: DB Freeze

Re: DB Freeze

From: Danisment Gazi Unal <dunal_at_ubTools.com>
Date: Tue, 21 May 2002 15:07:25 -0800
Message-ID: <F001.00466D66.20020521150725@fatcity.com>


Hello,

Please be careful while reading system state dumps. Because, Oracle's system state dumping function doesn't make consistent read. I mean you can see a process which waiting for a lock while there is no holding process.

regards...

"Gogala, Mladen" wrote:

> Next time do sqlplus "/ as sysdba", then "oradebug setmypid", then
> "oradebug dump systemstate 10" and then "oradebug tracefile_name" to find
> out the name of the generated trace file. That is good enough for oracle.
> If your platform is 64 bit HP-UX 11.0, then you need an OS patch. Included
> is the note about it from the mighty and all-knowing oracle:
> Article-ID: <Note:73835.1>
> Circulation: PUBLISHED (EXTERNAL)
> Folder: server.DBA.Admin
> Topic: Alerts
> Title: ALERT:HP-UX: ORA-7445 and Core Dump Generating Oracle
> Stack
> Trace
> Document-Type: ALERT
> Impact: MEDIUM
> Skill-Level: NOVICE
> Updated-Date: 03-MAY-2001 06:19:51
>
> ORA-7445 and Core Dump Generating Oracle Stack Trace
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> Versions Affected
> ~~~~~~~~~~~~~~~~~
> Oracle 8.X (64 bit only)
>
> Platforms Affected
> ~~~~~~~~~~~~~~~~~~
> HP-UX 11.00 (64 bit only)
>
> Description
> ~~~~~~~~~~~
> Whenever an Oracle stack trace is required, the shadow process will core
> dump.
> The stack trace in the corresponding trace file will be incomplete.
>
> Likelihood of Occurrence
> ~~~~~~~~~~~~~~~~~~~~~~~~
> Every time an Oracle stack trace is required. Some examples are:
> 1. an event has been set, and a stack trace is required in the trace file
>
> 2. some Oracle error requires a stack trace to be generated
>
> Possible Symptoms
> ~~~~~~~~~~~~~~~~~
> The alert.log will show something like:
> ORA-07445: exception encountered: core dump [11] [4286564440] [240] [0]
> [] []
>
> The corresponding trace file will include the following in the stack:
> _doprnt
> _fprintf
> U_get_previous_frame
> skdstgframe
>
> Error from U_get_previous_frame_x is 1
> Stack is not Windable
> Stack has no Unwind_descriptor
>
>
> Workaround
> ~~~~~~~~~~
> No workarounds available.
>
> Patches
> ~~~~~~~
> This problem is fixed by a HP patch. The first patch to include the
> fix is:
> PHSS_16849: s700_800 11.0 LIBCL patch
>
> This patch will be superceeded; please contact HP Support for detailed
> patch
> information.
>
> Oracle must be relinked after applying this patch, as follows:
> %cd $ORACLE_HOME/rdbms/lib
> %make -f ins_rdbms.mk ioracle
> The instance needs to be shutdown when relinking.
>
> References
> ~~~~~~~~~~
> bug 939745 : ERRORSTACK EVENT CRASHES FOR 64 BIT ORACLE ON HPUX 11.0
>
> Note 43507.1 : ALERT: HP-UX Patch Levels Advised
>
> ____________________________________________________________________________
> __
> Oracle WorldWide Customer
> Support
>
>
> > -----Original Message-----
> > From: Jamadagni, Rajendra [mailto:Rajendra.Jamadagni_at_espn.com]
> > Sent: Tuesday, May 21, 2002 4:00 PM
> > To: Multiple recipients of list ORACLE-L
> > Subject: DB Freeze
> >
> >
> > Last Friday, our 9012 database froze, I mean the only way we
> > could get in
> > was through sqlplus, no other tool would connect. We needed
> > to bring the
> > other side up as soon as possible so didn't spend any time running any
> > queries to see what was the problem. The other side (of the cluster)
> > wouldn't come up because it reported that some resource it needed was
> > locked. So we had to shot the pmon process.
> >
> > When analyzing the situation later, we found that Oracle didn't dump a
> > single useful trace file, so contacting OWS was pretty much
> > useless. OS logs
> > were clean, no alarms raised there.
> >
> > Has anyone encountered this situation that Oracle freezes and crashes
> > without any trace files? How does one collect useful
> > information in such
> > cases? Are there any tools that we could use to gather at-lease some
> > information before the DB crashes?
> >
> > Thanks in advance for any ideas and tips.
> > Raj
> > ______________________________________________________
> > Rajendra Jamadagni MIS, ESPN Inc.
> > Rajendra dot Jamadagni at ESPN dot com
> > Any opinion expressed here is personal and doesn't reflect
> > that of ESPN Inc.
> >
> > QOTD: Any clod can have facts, but having an opinion is an art!
> >
> >
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author: Gogala, Mladen
> INET: MGogala_at_oxhp.com
>
> Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> San Diego, California -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from). You may
> also send the HELP command for other information (like subscribing).

--
Danisment Gazi Unal
http://www.ubTools.com


-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Danisment Gazi Unal
  INET: dunal_at_ubTools.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).
Received on Tue May 21 2002 - 18:07:25 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US