> Once upon a time patrick@value.net shaped the electrons to say...
> >have had similar problems and have seen messages on the list to the same
> >effect) are all of a sudden having hardware failures.
>
> Goddess...
>
> YOU ARE NOT.
Who are you to tell me what problems we have had? Why were we down a PM
and and a 5 BRI board yesterday, because of a problem we did not have? Why
was that machine RMA'd? Oh, I'm sorry, please, tell me I was just
dreaming. Tell me that I have been dreaming about the sharp increase in
problems with 3.3.2c1 necessitating netboots or RMA'ing I have noticed on
the list. Tell me I don't have have at least four seperate messages in my
mailbox from people having to at least netboot, and in one case one person
RMAing 4 out of 8 PM's.
Do the words "PANIC:Watchdog" mean anything?
> Every corrupted FLASH is not a hardware failure. But since his was an
> RMA I bet his was. We wouldn't RMA on a simple corruption because those are
> readlly fixed with a reformat and reconfigure.
Of course it isn't.
> 3.3.2 and up have had more than normal corruptions. AGAIN, we changed the
Thank you. That is all I am saying. I am not saying "Livingston is evil
and their products suck." Actually, I was pleasantly suprised to call and
say "I have a site down" and get a real live person (the lists own John
Thompson) on the line right then and there. This is a marked improvement
from past calls to support. Thank you.
> There is a fundamental change in the formatting of the FLASH so part of the
> upgrade is rearranging all of the data. If there is any oddities, hidden
> corruption, etc, in the existing format this is likely to trip over it as
> it will translate to the new format in unpredictable ways.
And this is why when I talked to support(Melissa) I asked about error
detection when attempting to upgrade. I got the impression there wasn't
any. Is it viewed as "code bloat?" To go from a functional site, to an
RMA'ed machine because of an attempted software upgrade seems a little
silly.
> hits all the machines at the site, I suspect the configuration and/or
> environment first - the odds of that happening are so ridiculously low that
> there must be a common fault native to that site. We do scores of upgrades
> every day, and people are constantly downloading the ComOS. And it is a
> very, very small number of users who report any problem. And this RMA is the
> first one I've heard of, it is just coincidence that it happened with the
> upgrade. The FLASH chip probably went bad long before and it just happened
> that the older release wasn't trying to use that bit of FLASH, 3.3.2 tried
> to access it and boom, HW failure reported. 3.3.2 has more to store, so it
> is using more FLASH areas actively.
Well, I had one, and I have people that have sent me mail with similar
experiences. Because *you* are not aware of something, does not mean that
it does not exist.
/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
Patrick Greenwell (510) 943-5769 voice
Systems Administrator (510) 210-2000 modem
Value Net, Inc. (510) 943-1708 fax
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/