Re: RESIZE datafile bug???

> 2. I found a work-around myself ... if the data-files can be resized to the
> same size, or larger, than their size high-water, then the bug
> doesn't occur. The gotcha here is to determine that larger size, and for
> that matter, identify data files needing this sort of fix.
>
> 3. Oracle support believes that the critical size number is probably encoded
> in the data file header block. If this is so, and if Oracle Support
> can both determine and publish the byte locations in the data file of
> this information, then it would be a simple matter to match the extracted
> bytes from the data file, along with an "ls -l" output of the same
> file to both determine if a data-file resize is needed, and what actual
> size specification is required.
> Thanks Bill, For sharing this information and potential serious bug.
What about a
alter session set events 'immediate trace name blockdump level <dba>" where dba is the datablock address.

Venkat Jayaram
Principal Consultant
Enterprise Systems Consulting, Inc.

std disclaimer applies Received on Sun Nov 17 1996 - 00:00:00 CST