There is an index that is constantly being corrupted - about once a week. I have included in my program a call to update_index that will rebuild the index if bad data is detected but this shouldn't be necessary. Why is the index being corrupted so often? A dedicated indexing machine is attached to the network and for some reason it is missing out this index. I don't think this is connected with the index corruption.

Can someone help?

How do you know the data is bad? If you can programatically know what should be comming back, why use the index in the first place? Can't you just use the list you know should be there? But, what do I know? I'm just a dumn gnu. All I know is that power corrupts and absolute power corrupts absolutely. Perhaps you should try running the computer on a 110 volt main.

Gary Gnu

Jacob

Have you tried removing all indexes from that table, and re-creating them. Remember to get all users (including index server) logged off when doing this. Also, check that the bang file (!filename) has been deleted after removing the indexes. If it remains behind, you can be sure the index was damaged. Make sure all your tables are flagged shareable (if on netware). Do you have any update_index calls in your programs? (apart from the one you mentioned). If so - do you really need them? For best results, leave indexing solely to the index server(s).

View this thread on the forum...