> Plus... many folks can (probably..) work out how to replace a board,
> or modular part of the system (power supply) with another one from an
> also-dead system.. but would NOT be able to actually track down the
> real problem and do component-level repairs, simply by lack of skills
The actual job of soldering in a surface mount device (or even easier a
pin-trough-hole device) doesn't take much skill. The skill is in finding
the dead component.
And alas, I will _never_ believe you can fix a machine unless you
actually know what component has failed (even if you then have to replace
the board because it's a direct-on-board chip or something horrible).
I've seen problems caused by randomly replacing parts too many times.
Look back through the archives for the stories, I don't feel like posting
them again tonight,,,
> in that field. In the service industry, time is money, to field techs
> did (and still do) board- or system-level swapping, and the (skilled)
Which is why you'd think the customer would want it fixed right first
time....
> folks in the repair departments (try to) do the rest. Not good per
> se, but also not bad. I cost $500 an hour (OK, 475, really) so me
> spending hours on locating a problem in a module from a Catalyst
> stack would quickly grow extremely expensive. Rather, I'll come in,
> determine which board to pull, and then pull it. Often, this is
> already known because of system monitoring and reporting, so I know
I have never understood how you can trust diagnostics running on a
machine that's malfunctioning (and yes, I have seen this problem too).
-tony
Received on Mon Aug 02 2004 - 23:16:56 BST
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:36:32 BST