I've been reviewing the article from about a month ago, at
http://3000newswire.blogs.com/3000_newswire/2006/02/index.html, and I am not
quite sure....

I don't *THINK* we have the logging-related excessive IO the article
describes, but as the article says, it might be easy to overlook on a fast
machine.

We have an A500-200 with one dtc 72 mx, connected via thinlan -- old 10-only
hub -- newer 10/100 switch.  I don't see any transcievers, or any
readily-accessible white jumpers to flip.  So I don't think we qualify for the
problem, but if we do have it I'm not sure how to fix it.

Listfile of the H......A & H......B files says file 'A' has an eof of 863, and
'B' has zero records, not the zero and five that is supposed to mean we are
ok.

But what really fakes me out is the file dates.  Both files are open, with
access and modify dates a bit over a year ago.  But we have rebooted at least
4 times since then, including once this month.  So how did the system open
them after the reboot without changing the access date ?

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