I seem to have a problem

Server abends and hangs on reboot with Tree block error indicating corruption of NSS pool.
OK I've done the usual pool rebuild which resolves the problem, until the next time !
Checking the logs I've discovered the cause !

There are a couple of CRON routines for backing up GroupWise as follows:
I created a volume in the pool NSS1 called MAILBAK
and I load CRON & TBX via the AUTOEXEC.NCF
DBCOPY is then used to copy GW from MAIL: to MAILBAK:
Then MAILBAK: is later copied off the server

Every night CRON triggers COPY-GW.NCF which reads:
dbcopy mail:gwdom mailbak:gwdom
dbcopy mail:gwpo mailbak:gwpo

This COPY-GW.NCF routine appears to work just fine.

However, a couple of hours before COPY-GW.NCF triggers, another routine DEL-MAILBAK.NCF is initiated, which clears the MAILBAK volume of the previous day's backup, ready for the next backup.
del /s /q mailbak:gwdom
del /s /q mailbak:gwpo
This deletes all the files on MAILBAK: but not the folders, but that's OK

In a nutshell every so often, but not everyday, at exactly the same time as DEL-MAILBAK.NCF triggers, the server abends with the Tree block NNS pool corruption

I have attempted to solve the problem by changing the properties of the volume MAILBAK: to no salvage and flush files immediately, but I don't really know if this will work

Suggestion to resolve this problem would be very much appreciated