Disk or Network Error

S

Slickdock

Intermittent "Disk or Network Error" message. Shutting down Access
and restarting mde at client resolves it.
Running fe.mde with be.mdb Access 2002. BE.mdb is on MS2003 server.
FE.mde's on each client's WinXP SP3 machine. BE.mdb is compacted
regularly, file size 20mb. FE is compacted before being distributed,
and is 11mb. There are anywhere from 20-40 simultaneous users, each
running their own FE.mde.

Message can even appear when the workstation is at rest (user went to
lunch and returns to find the message). There is an onTimer event
that checks for inactivity and exits the database if inactive for 4
hours. Other than that, can't think what it could even be checking at
the server if no one is at the machine.

Not everyone gets the error. It is intermittent, and someone might get
the message when all other users are fine.
Appreciate your input.
 
J

John Spencer

I would suspect network problems of some type.

A "bad" NIC card in the relevant computer(s) or a bad network component.

Is it always the same computers? If so, what do they have in common on the
network - same router, same set of ports, etc.?


John Spencer
Access MVP 2002-2005, 2007-2010
The Hilltop Institute
University of Maryland Baltimore County
 
S

Slickdock

That was my first suspicion. But it does not seem to be the case.
These are computers with nothing in common across those network
components. When googling the error, I am surprisingly finding
software type replies. Like making sure the temp files are cleared
out, or checking the version of msjet40.dll. Since these are also not
factors, I thought I'd post my own discussion and see what else anyone
might be able to offer.
 
J

John W. Vinson

That was my first suspicion. But it does not seem to be the case.
These are computers with nothing in common across those network
components. When googling the error, I am surprisingly finding
software type replies. Like making sure the temp files are cleared
out, or checking the version of msjet40.dll. Since these are also not
factors, I thought I'd post my own discussion and see what else anyone
might be able to offer.

This seems to be a "catchall" error message that Access sometimes issues when
there's a problem and the program cannot (correctly) identify the source of
the problem. I've had it happen a number of times, and in every case the root
of the problem was either a corrupt table record or a problem with the VBA
project. Correcting the error (which sometimes required deleting and
reentering a corrupt record, or rebuilding an index) solved the problem; in
none of the cases I've seen was there actually a hardware error.
--

John W. Vinson [MVP]
Microsoft's replacements for these newsgroups:
http://social.msdn.microsoft.com/Forums/en-US/accessdev/
http://social.answers.microsoft.com/Forums/en-US/addbuz/
and see also http://www.utteraccess.com
 
S

Slickdock

Oh boy. A "Catchall" error message. I hate to hear that.
Thank you for your input...As for a bad record or table, the back end
is compacted nightly. I know compacting may not correct a "corrupt"
record, but often times the error appears when the user is not even
attempting to access a record...they are not even at the computer.
They see the message when they return from lunch or from an outside
meeting. As for a problem with the VBA project, the app is debugged
and compiled before distributing.

Any other thoughts, anyone?
 
P

Phil Hunt

Open Control Panel and look under System and see if there is indeed disk
error ?


Oh boy. A "Catchall" error message. I hate to hear that.
Thank you for your input...As for a bad record or table, the back end
is compacted nightly. I know compacting may not correct a "corrupt"
record, but often times the error appears when the user is not even
attempting to access a record...they are not even at the computer.
They see the message when they return from lunch or from an outside
meeting. As for a problem with the VBA project, the app is debugged
and compiled before distributing.

Any other thoughts, anyone?
 
S

Slickdock

Open Control Panel and look under System and see if there is indeed disk
error ?


Oh boy. A "Catchall" error message. I hate to hear that.
Thank you for your input...As for a bad record or table, the back end
is compacted nightly.  I know compacting may not correct a "corrupt"
record, but often times the error appears when the user is not even
attempting to access a record...they are not even at the computer.
They see the message when they return from lunch or from an outside
meeting.  As for a problem with the VBA project, the app is debugged
and compiled before distributing.

Any other thoughts, anyone?




- Show quoted text -

Is the disk error referring to the client or the server?
 
J

John W. Vinson

Oh boy. A "Catchall" error message. I hate to hear that.
Thank you for your input...As for a bad record or table, the back end
is compacted nightly. I know compacting may not correct a "corrupt"
record, but often times the error appears when the user is not even
attempting to access a record...they are not even at the computer.
They see the message when they return from lunch or from an outside
meeting. As for a problem with the VBA project, the app is debugged
and compiled before distributing.

You're not connecting over a wireless link or a WAN are you? It might actually
BE a network error if so.
--

John W. Vinson [MVP]
Microsoft's replacements for these newsgroups:
http://social.msdn.microsoft.com/Forums/en-US/accessdev/
http://social.answers.microsoft.com/Forums/en-US/addbuz/
and see also http://www.utteraccess.com
 
D

David-W-Fenton

I would suspect network problems of some type.

I disagree with this diagnosis. I have much more often seen this
error because of software or networking issues. The most recent was
a case where the server had NETBIOS over TCP/IP turned ON. We didn't
diagnose that as the cause until we moved the back end to a
different server that had it turned OFF.

A lot of domains get incorrectly configured for both DNS and for
domain name, the former because many people don't really understand
DNS, and the latter because many people don't get that an NT domain
is not the same as an Internet domain. You don't want to append the
domain suffix of your company's website, mycompany.com, to your
connection. Instead, you should have a domain name like
mydomain.local, which won't get mixed up with Internet domains.

I have seen both of those problems cause this kind of error. I've
also seen Exchange Server hotfixes screw up the SMB redirector on a
machine that was also a file server and lead to dropped network
connections and corruption.

All of those are software issues, and not hardware at all, unless
you define "hardware" as including "correctly installed and
configured drivers for hardware devices".
 
D

David-W-Fenton

This seems to be a "catchall" error message that Access sometimes
issues when there's a problem and the program cannot (correctly)
identify the source of the problem.

No, I don't think that's true at all. It is entirely due to the
local instance of Jet being unable to ping the workgroup file on the
other end of the network connection before some timeout kicks in (I
don't know what the timeout is). This can be causes by any number of
issues, including hardware failure or software
misconfigurations/bugs. The key is that something interferes with
packets getting through in a timely manner and Jet gives up rather
than be optimistic and assume things will clear up.

But I have never seen it happen for any other reason except
something interfering with the networking.

It can even happen with a file stored on the same machine where it's
being opened, if it's been opened through a UNC path to itself.

It's not that there are a whole host of causes that trigger this
message, it's that there are myriad things that can interfere with
the networking and cause Jet to be unable to connect to its
workgroup file on the other end of the wire.
 
D

David-W-Fenton

You're not connecting over a wireless link or a WAN are you? It
might actually BE a network error if so.

It's ALWAYS a network error in my experience. But there are so many
causes of things that interfere with networking that it can be
incredibly difficult to diagnose, since almost any hardware or
software component in the environment can cause the problem.
 
J

John Spencer

Could it be a timeout error? Are the users actively using the database when
this occurs or do they have the database open, but are not using the database
for a period of time?

John Spencer
Access MVP 2002-2005, 2007-2010
The Hilltop Institute
University of Maryland Baltimore County
 
K

Karen Hart

Could it be a timeout error?  Are the users actively using the databasewhen
this occurs or do they have the database open, but are not using the database
for a period of time?

John Spencer
Access MVP 2002-2005, 2007-2010
The Hilltop Institute
University of Maryland Baltimore County




- Show quoted text -

Both.
I did mention that I have an OnTimer event that after 4 hours of
inactivity, the front end quits. But I don't think its that, because
the mde actually closes at that point. Plus, it happens when they are
actively working also.
 
Joined
Jun 19, 2012
Messages
1
Reaction score
0
There is a good chance you have Unicast Flood Storms going on. On your LAN switch interfaces, turn on Storm-Control for Unicast and then check the log. Use config as follows:

conf t
interface Gigabitethernet0/4
storm-control unicast level 80:00 30:00
storm-control action trap
exit
logging buffered 9048 debug
end
wr
show log

If you have unicast flooding occurring on your switches then that will cause your database clients to disconnect. One cause of unicast flooding is asymmetric routing.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top