NW6 SP4, BM 37SP3, Johnson tuning ncf applied.

IBM Netfinity server, 2 Ghz P4, 2 GB RAM. 73GB HD, with 5 volumes created.

HTTP Proxy is being used.

3 Cache volmes (CACHE1, CACHE2, CACHE3), 4GB volume size each - 256
directories spread across these volumes. Traditional volumes were created.

When checking server health (Netware Remote Manager), shows "Bad" on
Directory Entries status. Even though each of these volumes show about 40%
space is available, the available directory entries show either 1% or 2%
across these volumes.

Was getting reports of inconsistent web access from users as well. We have
about 200 users going though this Proxy.

Am assuming memory is being exhausted from continous cache object creation.

Is there a away to tune this, so there are always available directory

If so, what is an adequate amount (%) to have available for each volume?

Matt Mello
Director of Technology
Holly Area Schools
email: mmello@holly.k12.mi.us