Re: RANT: Why Sun is losing to Linux
- From: Josh McKee <jtmckee@xxxxxxxxxxxxxxx>
- Date: Sat, 23 Dec 2006 08:25:41 -0700
In article <458d3a9d$0$327$e4fe514c@xxxxxxxxxxxxxx>,
Casper H.S. *** <Casper.***@xxxxxxx> wrote:
Josh McKee <jtmckee@xxxxxxxxxxxxxxx> writes:
The reason for the version downgrade was not to use a version that
doesn't "whine" about the component. The reason for the downgrade is
that version 5.12 was the last version which does not have the service
mode for which I do not have the password. Therefore I have full control
over my system.
It's the last version which does not does the auto diagnostics
such as described in
http://www.sun.com/blueprints/1003/817-3910.pdf
and that firmware does not look at the "this component is faulty"
status.
This isn't about trying to get a bad piece of hardware to work. It's
about trying to get the replacement for said piece of hardware to work.
The document quoted above claims that the CHS data is stored with
the component; so either the system marked it as bad immediately or
someone shipped you components which were marked as bad before.
The problem turned out to be that I didn't replace the faulty DIMM. The
way the output from the "showcomponent" command and the documentation
for the system board I had led me to replace the incorrect DIMM.
Replacing the entire bank with new memory did re-enable the bank
automatically.
Or, as is the reason for my having started this thread, they will look
to alternatives which don't require payment to the vendor in order to
enable full system functionality after having replaced faulty hardware.
It's not supposed to work that way.
I'm glad to say that it appears that it's not working that way.
If Sun is going to require me to purchase a service contract or pay
hourly to obtain this password you can bet that said money will instead
be applied towards a non-Sun system. Sun probably won't care about my
isolated case. Hence the reason Linux is encroaching on Suns territory.
Hence the reason for my having started this thread.
I'm not sure where Linux comes in here; Solaris runs fine on non-Sun
(x86, AMD or SPARC) hardware too.
The animosity generated by forcing someone to pay to re-enable their
hardware would result in that person avoiding all Sun solutions. If Sun
were to requirement me to pay to fix this problem, which I am happy to
say doesn't appear to be the problem I thought it was, I wouldn't want
to do business with a company like that.
We do care about keeping our customers; but the design point of these
systems and the way they are serviced is not the single person at home
without a service contract.
While not intended for such service these systems have come down
significantly in price. You can pick up a base configuration from Ebay
for about $600. Therefore they're going to start finding there way
outside of the intended area of service. Owners will not look favorably
at Sun if they have to pay in order to obtain a password that would
allow them to fix a system once a part has been replaced. I know that's
how I was feeling yesterday. Hence my rant.
Anyway, I would not bet on the replacement components being "good".
It seems that there were a number of things which led me to an incorrect
conclusion about how these system recover from a fault. I am happy to
say that once I finally replaced the actual fault DIMM the system
brought the memory bank back online as I would expect. Already the
purchase of this system has paid off.
Josh
.
- Follow-Ups:
- Re: RANT: Why Sun is losing to Linux
- From: Casper H . S . ***
- Re: RANT: Why Sun is losing to Linux
- References:
- RANT: Why Sun is losing to Linux
- From: Josh McKee
- Re: RANT: Why Sun is losing to Linux
- From: Tim Bradshaw
- Re: RANT: Why Sun is losing to Linux
- From: Josh McKee
- Re: RANT: Why Sun is losing to Linux
- From: Casper H . S . ***
- RANT: Why Sun is losing to Linux
- Prev by Date: Re: RANT: Why Sun is losing to Linux
- Next by Date: Re: RANT: Why Sun is losing to Linux
- Previous by thread: Re: RANT: Why Sun is losing to Linux
- Next by thread: Re: RANT: Why Sun is losing to Linux
- Index(es):