Can anybody please shed some light on our most annoying problem. On our system, workstations seem to hold locks on Index files. They are operating as if nothing is wrong, working throught he menus and reading other tables, unyet in NLM_STATS it shows that they have a lock on an index file. This causes other workstations on our network to 'lock up' waiting for the lock to be released (obviously). Now 'Locked Up' has become a system bashing catchword, as it happens several times a day. It is happening on Dos machines as well as Win95 (I have read all the discussion notes on Win95 locks), on a Novell 3.12 Network. ARev is at version 3.11 and the NLM at version 1.10. Please help! I am currently collecting data from the PC's that are holding the locks, with a view to changing their network cards.

You did not mention whether you have a dedicated indexer or not. Setting this up permits setting the "Update indexes before filter" to "No" for other users (for most applications). (Obviously, for non-indexer workstations, the time between update settings would be normally set to zero. (Disabled))

Obviously, this is a bit more hardware intensive but sounds like it would provide a quick payback. We found that this also added to the overall stability of the system.

We have users connected to our servers from all over the state (via T1) thus this also helps keep the network traffic accross the T1 minimized since the INDEXER is local to the server.

Hope this helps.

Maybe an NLM update would help?

There was also a TSR (REVLOCK.EXE) that used to float around that would display the current number of locks held by a session.

[email protected]

Sprezzatura, Inc.

View this thread on the forum...