Well, as they say, "This list is incredible".
Remember my problem? I had a new Fujitsu M2372K disk as a
second disk on a Xylogics 451 controller; a M2361A Super Eagle as the
original disk. The new disk arrived formatted. It failed reads
after succeeding writes on *some* label attempts. All this under OS3.5
After talking to the technician who formatted, I found out that
their 451 controller was sick that week so they used a 7053 (arghhh)
and hadn't mentioned it. What to do? Format under 3.x? Format under
4.x? Send it back?
Amongst all of the replies that I received, I managed to
formulate an opinion but it wasn't easy. For every claim that
was made, there was someone who succinctly stated the opposite.
For instance,
- Many reminded me of the well-known rumour that disks formatted
under SUN OSv4.x could never be reformatted under 3.x.
+ At least one person claimed to have done just that many times
without noticeable problems
- No one really said whether, having formatted under 4.x, one
could use the disk under 3.x. Some suggested "no".
- Everyone agreed that formatting under a 7053 controller was
incompatible with a 451.
+ Literature mentioned by some "implied" that the 7053 and the
451 are compatible; experience has shown otherwise.
- Some felt I could get away with formatting under 3.x since it
should find 90+% of the defects.
+ Others felt I had no choice but use 4.x (which preserves the
table) or send it back.
- Some mentioned the 451's 600 byte/sector minimum size.
+ Others sent their format specs which showed them using under
600 bytes/sector.
- Some said that the 451 required having identical disks with
identical configurations
+ Some had working arrangements with mixed CDC, Fujitsu and
other disks on the same 451
What did I do? Well, I set the byte/sector size to 600 and
I reformatted under 3.5 with my 451 controller. It found 4 defects,
2 of which it couldn't confirm. The original defect list showed
115 defects. I sent it back with a threat to stop payment. They have
promised it by next Tuesday.
I found that after the format, "test" (random write/read)
worked fine, I could read and write a label, partition and mount the
disk. In fact it worked in all tests thereafter. However, with 115
listed defects, I decided not to chance disaster.
My experience says:
1) 7053 controller is incompatible with 451
2) 3.x format is inefficient and should be avoided
3) The 451 can handle different disks at least of the same
company brand with different configurations.
Loki Jorgenson node: loki@physics.mcgill.ca
Physics, McGill University fax: (514) 398-3733
Montreal Quebec CANADA phone: (514) 398-6531
Thanks to: (I lost a couple of names.... sorry)
ucrmath.ucr.edu!proton!muon!baumann (Michael Baumann)
mikulska%ece@ucsd.edu (Margaret Mikulska)
andys@ulysses.att.com
del@mlb.semi.harris.com (Don Lewis)
Bill Aten <aten@afcsa.af.mil>
kwthomas@nsslsun.gcn.uoknor.edu (Kevin W. Thomas)
Richard Schultz <rich%idacrd@princeton.edu>
era@niwot.scd.ucar.EDU (Ed Arnold)
Rob ten Kroode <rtk@styx.sun.nl>
hr@rigel.astro.uiuc.edu (Harold Ravlin)
aldrich@sunrise.Stanford.EDU (Jeff Aldrich)
Fons Ullings <fons@nat.vu.nl>
celeste%coherent.com%uunet.UUCP@Larry.McRCIM.McGill.EDU
kms@sunrise.East.Sun.COM (Kevin M. Samuels - Sun Consulting NJ)
Mike Clarkson <mike@ists.ists.ca>
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:05:57 CDT