At 10:08 PM 01-10-02 -0400, you wrote:

>At 10/1/2002 07:16 PM, you wrote:
><details snipped>
>
> >gbak: writing data for table XRDS511
> >gbak: 3190 records written
> >gbak: writing index XRDS501_IDX
> >gbak: writing index XRDS501_LOC_IDX
> >gbak: writing data for table XRDS501
> >gbak: ERROR: database file appears corrupt ()
> >gbak: ERROR: wrong page type
> >gbak: ERROR: page 292250 is of wrong type (expected 7, found 5)
> >gbak: ERROR: gds_$start_request failed
> >gbak: Exiting before completion due to errors
>
>running gfix -mend -ignore on this database gave the following output:
>
>Summary of validation errors
> Number of record level errors : 1
> Number of index page errors : 377
> Number of database page errors : 18415
>
>Running a query looking for the missing records does still shows that they
>are missing. Does that mean gfix was unable to fix the problem?

Yes. Despite its name, gfix doesn't "fix" damaged data. It has a limited
capability to identify and/or mend corrupted structures and it can discern
that data are corrupt; but it has no "data repair" capability.

This problem seems to have the signs of data overwriting - either from
beginning to overwrite from the start of the database (because of
overflowing the OS file size maximum) or because the server is seeing the
database as two databases (the bad connection string bug).

heLen