> 	I have a MicroVAX II with a Sigma RQD-11/EC ESDI controller and two
> Hitachi DK516 1 gig(ish) ESDI drives.  After using disklabel to set
> everything up, I newfs'ed the partitions that I needed to, but now when
> I access them, I get this error:
> 
> /netbsd: ra0: drive 0 soft error datagram (continuing): controller error
> (unknown subcode) (code 10, subcode 8)
> 
> I can do lots of things to the drives, and I can boot from them, but the
> error keeps coming up when the drive is accessed (not so much when it's
> mounted read-only, though). I think I looked at all of the obvious
> things (cables are fine, etc), & now I'm stuck.  If anyone knows what
> this means, I'd love to hear from you.

Some versions of the Webster/Sigma microcode (this particular series of
Sigma controllers were all designed by Webster in Australia) report 
back status information that confuses the NetBSD ra driver.  This 
extended status information is ignored by VMS, RT-11, RSX-11, and other
DEC operating systems, but all the BSD-derived MSCP drivers (including
BSD4.2, BSD4.3, BSD2.11, and NetBSD) are confused by this information
and cause the log message to the console that you are seeing.

Two solutions:

1.  Ignore subocde 8 errors (these are complaints that the unit is
"already online") in the driver.  This is what I've done in my BSD2.11
system.

2.  Use a different version of microcode on the disk controller.  When
you go into the controller config/diagnostics, what version of Wombat
does it say it is running?  If you've got a EPROM burner and some
spare 27512's, upgrading to a different version is quite straightforward.

Tim. (shoppa@triumf.ca)