Hi,

I hope someone here can help.

We have an IBM server with 4 CPU's, 4gb of RAM and running Pervasive 8.6.
Periodically ALL clients lose connection, and retry to connect simply
returns a variety of errors, but most prevailant is 3014. I then go to the
server and the monitor is showing everything running ok, and my application
runs on the server with no issues. A reboot of the server allows all users
to reconnect.

The server is also running Mcafee VirusScan Enterprise 8.0.0.

Has anyone come across this sort of failure before and what was the
solution?

--
Neil Brownlee.

Without any pattern of occurrence, you may receive a status 85
when the file is closed because Anti-Virus software opens and
locks the file to scan causing the next database operation to fail.
To resolve, set the Anti-Virus software to not scan Pervasive SQL
data files. Consult your Anti-Virus software manual for
instructions on how to exclude files.

For our network we excluded the directory where our data files are
stored.

Gil

Neil Brownlee wrote:

nelsonsoft wrote:
I've also seen this with backup programs, if not configured properly.

One of my customers had an automatic backup running every night, and it
would corrupt our data files, and cause a number of weird errors. You'd
see error messages about "error xx.xxx.x in file, file rolled back" in
PVSW.LOG.

We resolved that by having our customer use the Pervasive SqlUtil
-startbu and -endbu commands to put our database in "backup mode", and
having them exclude the "delta files" (Pervasive writes all changes to
what they call delta files, with an extension of ".^^^" files in this mode).

Paul

More news on this :-

When the whole lots goes down, the following appears in the PVSW.LOG

W3CACHE 000012DC W3DBMGR.EXE SERVER W Resources allocated

Why does this happen?

--
Neil