Upon upgrading some client workstations, there has been a huge
slow-down in db access. These users were on 98 and are now in 2k,
accessing the DB over a share. Some of the 2k users are experiencing
extreme slow downs, particularly when first logging into the front end
and unlocking the records. Opportunistic file locking is not enabled
(the reg key isnt there, I assume it defaults to off?).

It seems to be just be sitting there, locking, reading and unlocking
DBF, CDX and FPT files. It doesnt seem to be an issue with the temp
dir/caching. Hardware/network appears fine. Most 2k clients are
fine.

Any advice would be highly appreciated, most of my db experience is
with Unix.

To respond via email: mikes AT ccs REMOVE dot neu dot edu

thanks!

An update:

Utilizing filemon I've observed this behavior:

On "bad clients"

33037 9:16:31 AM NTVDM.EXE:1760 OPEN Y:\MY-DATA\USERPREF.DBF SUCCESS Options:
Open Access: All
33038 9:16:31 AM NTVDM.EXE:1760 OPEN \\SERVER\PROGRAM\MY-DATA\USERPREF.DBF SUCCESS Options:
Open Access: All
33039 9:16:31 AM NTVDM.EXE:1760 CLOSE \\SERVER\PROGRAM\MY-DATA\USERPREF.DBF SUCCESS
33043 9:16:32 AM NTVDM.EXE:1760 READ
C:\WINNT\CSC\00000001 SUCCESS Offset: 0 Length: 4096

On "good clients"
1663 9:11:01 AM NTVDM.EXE:1784 OPEN Y:\MY-DATA\USERPREF.DBF SUCCESS Options:
Open Access: All
1664 9:11:01 AM NTVDM.EXE:1784 READ
C:\WINNT\CSC\00000001 SUCCESS Offset: 0 Length: 4096

It seems to be hitting the UNC and opening the exact same file twice
on the bad clients. Anyone have any theories?

please respond via newsgroup or email provided in first post. thanks.