Hi,

Congratulations on a great release with Harbour 2.0!

If this is the wrong place to ask, let me know, and I'll post elsewhere, but
I'm having the continuation of a long term problem with large data file
(>2gb) read errors.  Mostly I get 1010's from standard errorblock trapping,
but sometimes get

------------------------------------------------------------------------
Application Internal Error -app.exe
Terminated at: 2010.01.15 12:39:02
Unrecoverable error 9201: hb_cdxPageSeekKey: wrong parent key.
Called from DBSEEK(0)

In hb_out.log - no trapping by errorsys, the app just dies.  I've tried many
many iterations of lock types, dirty read's on/off etc.  The file in
question is normally getting extended via a Flock()/Append From/UNLOCK code
segment (the write process never crashes), and the read process (the one
that crashes with 1010's and the above occasional hb_out.log's) usually dies
on DBSeek, DBSkip, or a good old fashioned field read (x := FIELD->x).
Additionally, the indexes are never actually corrupted, but the reading
application exe sure thinks so...

FWIW, these errors only start to occur when the data file size breaches 2gb
or so.

My rddsys is as follows.


  REQUEST _DBF
  REQUEST DBFFPT
  REQUEST DBFCDX
  rddSetDefault( "DBFCDX" ) 

and my (current - I've tried _many_ iterations) RddInfo settings are:

  // default lock scheme...
  RddInfo( RDDI_LOCKSCHEME, DB_DBFLOCK_VFP )
  
  // enable index dirty reads...
  //RddInfo( RDDI_DIRTYREAD, True )
  
Server environments include Win 2000/2003/2008 server, all applications that
share these files run locally, no network sharing of any kind.

I've been wrestling with this for months now, and would appreciate some
steering?

TIA,

FS


_______________________________________________
Harbour mailing list (attachment size limit: 40KB)
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour

Reply via email to