We have confirmed the bug.

Basically, the code that implements DBINFO 205 will return correct information for 
the first 208 datasets, and then have no chance of being correct for datasets 209..240.

Instead, for datasets 209..240 it will be retrieving data from the stack frame, which
is fairly likely to be different from time-to-time, and always likely to be unrelated to
the correct values that should be returned. 

> ...  but the expansion flag returns 0 (meaning expansion is off) for
> datasets numbered 208-240.

Depending upon luck, you may get 0 when a 1 is appropriate, or a 1 when a 0 is appropriate.
I.e., it won't always be the case that 0 is being returned for all datasets above 208 (expandable
or not).

We're investigating the possibility of producing some kind of patch for this.

Stan
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *