We have a two node cluster (Hp Proliant DL380 Cluster package) with 6 Gig
of Ram. The Server is running only a pervasive Database (v8.6) and service
pack 4a and the nss update. Initally the server indicated that it needed
more logical memory and the health monitor in NRM was also showing this.
We restarted the server with the following -u671252480 which was
recomended by the server. Now the system has been running for more than a
month. Yesterday it indicated on the system console that it needed more
logical memory and recomended -u664567808 but the health monitor does not
indicate that it needs more memory with a warning and the Statistic Trend
Graphs indicates no sudden drop in the logical memory. The server seems to
going great everything is working. The Workload yesterday was very high,
but this not any different than a week ago or a month.

The only parameter we changed was the NSS Cache Balance Percent. Initally
the server default was 80%, it was changed to 60%.
I do not understand why the system console is saying logical memory needs
more memory??????
Any help is appreciated.

Dev Daljeet