RE: FYI: Issue with Oracle 8.1.6 and corruption after 248 days o

Is this fix included in the 8.1.6.2 patch?

Ian MacGregor
Stanford Linear Accelerator Center
ian_at_slac.stanford.edu
-----Original Message-----
From: bmcgraw_at_mindspring.com [mailto:bmcgraw_at_mindspring.com] Sent: Tuesday, August 22, 2000 1:12 PM
To: oracledba_at_quickdoc.co.uk
Subject: Re: FYI: Issue with Oracle 8.1.6 and corruption after 248 days ofuptime

I was curious enough to call Oracle about this, after I didn't find anything on Metalink about it. FYI:

It's a legitimate bug (1084273/1227119) that is Solaris-specific. There is a patch available for it outside of 8.1.7, however. The patch is available on oracle-ftp.oracle.com under /server/outgoing/SUN_SOLARIS2/BUG1227119

The analyst said that the patch is for the 32-bit version, and is checking to see if the patch needs to be applied to the 64-bit version as well.

Brian

Jay Weinshenker wrote:

  I'm still waiting for someone to ask if this issue pertains to NT databases
  as well.. just so I can laugh and point out the server itself won't make it
  long enough to be an issue...

  J
  At 01:50 PM 8/22/00 -0500, you wrote:
>Same here. I'm currently setting up Oracle at a new client site, so I
>know it's
>not an issue with my databases.
>
>Brian
>
>Jay Weinshenker wrote:
>
> > Didn't bother. My clients never manage to have a DB up for 248 days,
so it
> > isn't that big of an issue for me...
> >
> > Feel free to do so though
> >
> > J
> > At 02:36 PM 8/22/00 -0400, you wrote:
> > >Jay -
> > >
> > >I didn't find anything on Metalink pertaining to this. Do you know
if
> > >your source verified this with Oracle?
> > >
> > >Brian
> > >--
> > >SHORT DESCRIPTION: Oracle database corruption on Solaris when
  uptime
> 248 days
> > >
> > >KEYWORDS: oracle solaris
> > >
> > >QUESTION:
> > >This was forwarded to me today at my client site from a Sun rep.
Seems
> > >legit; I don't propose to test out whether they're correct about the
> > >corruption! FYI.
> > >
> > >SUMMARY:
> > >Oracle database corrupts when Solaris is running more than 248 days
> without
> > >reboot.
> > >
> > >Description
> > >Customer production system running Oracle version of 8.1.6 or less
will
> > >have the database control file corrupted if the machine has been
running
> > >248 days without a reboot. This will require a rebuild and recovery
of the
> > >database.
> > >
> > >Solution
> > >The issue is how Oracle interprets the number of "clock ticks" that
the
> > >kernel uses to keep track of time since the machine was booted. When
the
> > >number of clock ticks reaches (2 ^ 32) + 1, Oracle reads it as a
negative
> > >number and the corruption occurs.
> > >The permanent solution is to install Oracle 8.1.7 when it releases.
> > >
> > >The temporary workaround is to REBOOT THE SERVER BEFORE IT
  REACHES 248
> DAYS
> > >OF UPTIME.
> > >
> > >Oracle has confirmed the problem.
> > >
> > >Many rebooted their systems at Y2K. If these production systems have
run
> > >continuously since then, their machines will begin to reach 248 days
of
> > >uptime starting September 4, 2000 just after noon.
> > >
> > >----- End forwarded message -----

--
------------------------------------
|Brian McGraw    --      Oracle DBA|
|----------------------------------|
|mailto:BMcGraw_at_mindspring.com     |
|http://bmcgraw.home.mindspring.com|
------------------------------------


--------
If you're bored, then visit the list's website: http://www.lazydba.com
(updated daily)
to unsubscribe, send a blank email to oracledba-unsubscribe_at_quickdoc.co.uk