[Return to contents]

"Record is locked- retry -cancel" error message

Last updated: 9 Jan 1998

Sometimes a shared Approach database repeatedly returns a "record is locked- retry -cancel" error message, even though the database has been set to optimistic record locking.

This can occur when the link to the network drive where the data and view files are held is down. In this situation you need to exit Approach, reestablish the link to the network drive (usually by rebooting your workstation), and reopening the database in Approach.

It will also occur if a user does a search that causes a new index to be created. Everyone else will receive a "locked" message until the indexing is completed. And, of course, if you have a lot of users and they often attempt to work on the same records then they are going to frequently get record locking errors.

It can also occur for a variety of other more obscure reasons. Some things that have been suggested to keep these at bay are:

• delete excessive apr*.tmp files created by Approach. If this is a regular problem you may want to include the following statement in your autoexec.bat file: "del c:\windows\temp\apr*.tmp"
• keep the users system clocks within a minute or so of each other. Locking can only work properly if all PCs are synchronized. For example:

        10:00:00 User A locks a record
        10:00:01 User B locks another record
        9:59:59 User C (whose time is not synchronized) tries to access User A's record, which is not yet locked because it isn't 10:00:00 yet. Ooops.

Get the network administrator to set up the network to synchronize the PCs clocks.

• try compressing the database, and if all else has failed try deleting and recreating the .adx files. See articles 'Deleting and recreating index files (.adx)' and 'Repairing corrupt index (.adx) files'.
• put a blank record as the first record in the database. This record must be devoid of any join relations. ( don't know what this does, but apparently it can help!)