Errors in 8.0.4.0.0 EE on NT

From: Steven Cardinal <scardinal_at_yahoo.com>
Date: 22 Sep 1999 07:12:18 PDT
Message-ID: <7sao42$r9m@journal.concentric.net>

We keep getting the following error on our development server (we haven't moved to production yet) :

KCF: write/open error block=0x12c07 online=1

     file=4 E:\ORANT\DATA\DATA01.DBF
     error=27072 txt: 'OSD-04008: WriteFile() failure, unable to write to

file
O/S-Error: (OS 33) The process cannot access the file because another process has locked a portion of the file.' Instance terminating due to error 1242

I have deleted and rebuilt the database and, during the same process got that error, which appears to corrupt the database. The process that runs is:

DELETE FROM DeDupImportTable A WHERE ROWID >  (
  SELECT MIN(rowid) FROM DeDupImportTable B   WHERE A.vchPatientFName = B.vchPatientFName

  AND A.vchPatientLName = B.vchPatientLName
  AND A.vchAddress1 = B.vchAddress1
  AND A.vchCity = B.vchCity
  AND A.vchState = B.vchState

);

 The statement runs fine with small numbers of rows but, when we have 1,000,000 records, it craps out (about 5 hours into the process). The server has the following settings:

  processes                = 30
  timed_statistics         = TRUE
  shared_pool_size         = 9000000
  control_files            = E:\orant\control\control01.ctl,
E:\orant\control\control02.ctl
  db_block_buffers         = 40000
  db_block_size            = 8192
  log_buffer               = 102400000
  log_checkpoint_interval  = 10000
  log_simultaneous_copies  = 8
  db_files                 = 80
  db_file_multiblock_read_count= 32
  rollback_segments        = USERRBS

  sequence_cache_entries = 30
  sequence_cache_hash_buckets= 23

  global_names             = TRUE
  sort_area_size           = 25000000
  text_enable              = TRUE
  parallel_max_servers     = 20

I have 2 questions - what could be causing it (I've tried to download 8.0.5 with no success - is there a relevant bug in 8.0.4?)? second - there's got to be a better way to remove duplicates than the statement we are using above - we don't have control over the source data. Any help?

Thanks Steve Received on Wed Sep 22 1999 - 09:12:18 CDT