I upgraded a NetWare 6.5 sp2 to server to sp3, os3a and nss3a. Things
appear to be well, no more volume dismounts. And the portal makes it
easier to track the memory fragmentation.

I was looking at the TID on memory fragmentation and the only question I
have is how long do I watch it to determine if its an issue. I did the
udpates on Sunday and at that time it was 1.8 percent. Then Monday with
a load on the server it went to 2.1. During the backup that night it
went up but after the backup went down. I think it was at 5 something.
After a few more days its at 6.4.

I am not sure if that is significant enough to do any of the steps on
the tid. I did see some messages generated by the server auto tuning
Monday when the backup kicked, which did not surprise me. But nothing
since then.

The messages were. Logical address space running low restart the server
with a specific u switch, which I did not do. Then it changed the max
file cache. And then it repeated the max file cache message