J
Jason McClellan
Hello.. I have a database here which is used for purchasing/stock etc. It
is an Access front end, a very simple MDE with a few queries that people
run. Backend is SQL, access utilizes linked tables. The queries return all
records by default.
Recently, two events occurred; the server which SQL lives on was
re-installed with Windows 2003 server (SQL is version 2000 standard) and the
database exceeded 32,768 records.
Previously, everything worked fine. But now, I am seeing alot of lock
contention, PID's will have 'runnable' status, with command 'SELECT', Wait
Type 'NETWORKIO' and an ever-increasing Wait Time. This occurs only in a
certain condition; if a query is run, this lock is exhibited. The lock is
released as soon as they do something with the query, filter, edit
something, or even just go to the last record (with the > button). But
until something is done with it, this lock is held on the entire table, and
nobody else can edit/add etc anything. This occurs even if the person
running the query has read-only access.
My only/best guess is that for whatever reason, there is some kind of pause,
or timeout, or something occuring when access gets to the 32k records point,
and this puts a kaibosh on the query until the user forces it to do
something. I can't be 100% sure, but there was a time after the server was
reinstalled during which it did not do this, so I am more inclined to lean
towards the 32k theory, but I can't find any info on any possible causes..
in fact I can find extremely little info even on the NETWORKIO wait type.
I would appreciate any help/ideas/info anyone might have.
Thanks
Jason
is an Access front end, a very simple MDE with a few queries that people
run. Backend is SQL, access utilizes linked tables. The queries return all
records by default.
Recently, two events occurred; the server which SQL lives on was
re-installed with Windows 2003 server (SQL is version 2000 standard) and the
database exceeded 32,768 records.
Previously, everything worked fine. But now, I am seeing alot of lock
contention, PID's will have 'runnable' status, with command 'SELECT', Wait
Type 'NETWORKIO' and an ever-increasing Wait Time. This occurs only in a
certain condition; if a query is run, this lock is exhibited. The lock is
released as soon as they do something with the query, filter, edit
something, or even just go to the last record (with the > button). But
until something is done with it, this lock is held on the entire table, and
nobody else can edit/add etc anything. This occurs even if the person
running the query has read-only access.
My only/best guess is that for whatever reason, there is some kind of pause,
or timeout, or something occuring when access gets to the 32k records point,
and this puts a kaibosh on the query until the user forces it to do
something. I can't be 100% sure, but there was a time after the server was
reinstalled during which it did not do this, so I am more inclined to lean
towards the 32k theory, but I can't find any info on any possible causes..
in fact I can find extremely little info even on the NETWORKIO wait type.
I would appreciate any help/ideas/info anyone might have.
Thanks
Jason