rion@FORD-WDL1.ARPA (Rion Cassidy) (03/29/88)
After waiting about two weeks for people to respond, I have now
summarized the responses to my 'reliability questions' posting.
Out of all the (who knows how many?) Iris users there are on the
info-iris system, I recieved 12 responses, one of which was from a SGI
technician.
I've categorized and tallied the responses as follows:
"you've got a lemon -- our Iris(s) has had excellent 2
reliability"
"your story sounds familiar -- we've had similar problems" 4
"SGI is putting out leading-edge high-performance workstations 1
and you can't expect the best in reliability"
"we've had hardware problems, but not as bad as yours, and 4
would like to take this opportunity to complain about software
bugs and poor documentation"
"I'm an SGI tech and would like to make a few comments" 1
My favorite quote:
"We have two Iris workstations here, a 3020 and a 3030. In my opinion,
the worst decision we every made was to buy one of these things. The
best decision we ever made was to buy two of them, as we now have
enough parts to keep one of them running."
Comments from the SGI tech:
"Switching the system off without executing reboot could have caused
damage to the disk and subsequent problems with getting that disk to
work again. ...in every case in which a customer complained to me about
frequent problems with his disk drive, I have discovered that he (or
she) was NOT using the reboot command (they just hit the power
switch)."
"It seems that the tape drive on many 3000 series products is not
properly grounded. We have devised a method of grounding a tape when
it is inserted. You should ask your FE for the grounding clip for
your tape drive. Electrostatic discharge from a tape inserted in the
tape drive is almost always the cause of the type of system lockups
you describe. (By the way, until the clip is installed, touch your
tape to the cabinet before inserting it.)"
"I hope you will keep in mind that the nature of humans is to speak up
when things are wrong and to keep quiet when things are going well -
so your survey may be a little skewed to those with problems with
their IRIS'."
I have to agree with this last comment. I will refrain from making
*any* conclusions; the number of responses is very small, and, I'm
afraid, not indicative of anything. Another problem is that some
people obviously have a single machine, some have dozens, while others
did not specify. This makes it hard to do any meaningful statistics.
I have also purposely avoided naming the respondents.
Rion Cassidy
Ford Aerospace
rion@ford-wdl1.arpa
...{sgi,sun,ucbvax}!wdl1!rion
Disclaimer: My employer forced me to write this at gun-point.
I assume no responsibility whatsoever for what I've said here.