Thanks. Yes, i did turn off the oplocks, as per recommendations of some samba sites. problem persist.
Actually, the application is running normal operations. However, .dbf is known for index corruption (oplocking something to do with this also), and so the application have a re-indexing process. Here, the application is deleting all indexes, and start an update process. Here many small .dbf files are being read. If the databases are in the workstation, and issue the update command, the whole process takes only 5-10mins. But when it is in the ibay (smb), it took almost 30-45 minutes to finish.
I checked with ext2, ext3 and reiserfs filesystem, if there are any issues. Reiserfs is known to be speedy in small files. Could this be a filesystem issue? Or Samba still have solve .dbf bugs?
Any help would be appreciated.