On Fri, 27 Jul 2007, David Mathog wrote:

The question: what is the expected motherboard battery lifetime in
systems that are continuously on?

To my rough experience (from a small and anecdotal sample), between 5
and 10 years it is not uncommon for batteries to start to give, even in
server class systems, with an increasing probability of failure starting
(from pretty much 0) around year 6.  So it might have been "marginally
unusual" to see it by year 5, but it will be "not uncommon" in another
two years and "common to universal" a year or two later.

And yeah, it is a real PITA if you have anything non-default in your
BIOS settings.  Not so much a problem for clock settings, but if you
have to set up peripherals or memory (in older motherboards) every time
you boot...

    rgb


Background:

The question arises because it was necesary to cycle the power today
on a couple of nodes by unplugging them and then plugging them
back in again.  When one came back up the BIOS settings were gone.
That wasn't evident until a keyboard and monitor were plugged in.
Losing BIOS settings when power is removed is pretty much the
classic symptom of a dead motherboard battery.

These nodes are now about 5 years old, and that is about the average
battery lifetime in PCs that are turned off every night.  However these
nodes have been on 99.99% of that time, so presumably the drain on the
battery should have been nil.  Shelf life for motherboard batteries is
typically about 10 years. There is no info on how old the batteries were
that went into these systems, but presumably Tyan goes through a lot of
batteries and these were probably fresh when installed.

So bottom line, is this just a freak event or would you expect the
other motherboard batteries to also die soon?

Thanks,

David Mathog
[EMAIL PROTECTED]
Manager, Sequence Analysis Facility, Biology Division, Caltech
_______________________________________________
Beowulf mailing list, [email protected]
To change your subscription (digest mode or unsubscribe) visit 
http://www.beowulf.org/mailman/listinfo/beowulf


--
Robert G. Brown                        http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:[EMAIL PROTECTED]


_______________________________________________
Beowulf mailing list, [email protected]
To change your subscription (digest mode or unsubscribe) visit 
http://www.beowulf.org/mailman/listinfo/beowulf

Reply via email to