|  | 	Last I knew, for the older version the SPD defined what
	was supported...  It might be useful to look through all
	the systems listed to see if it is supported on any of
	them.  If you can find one, then lack of support on one
	may be the lack of qualification instead of not being able
	to function correctly.  Also look in /sys/data/cam_data.c
	to see if it has an entry for the device.  If not, then
	you're dealing with what is basically a 3rd party device,
	even if it has Digital logo on it.
	I would start by looking at the detailed error logs to see
	if the problem is hardware or media.  Supported or not,
	devices to break now and then.  If it doesn't seem to be
	a hardware problem and the cam_data.c entry doesn't exist
	try writing one.  The comments in the file should be sufficient
	if you have good enough drive documentation (though I've never
	tried it).  If that doesn't help, check the USG hardware group's
	web page to see if there are any known problems with the device.
	After that either get out the SCSI analyzer and see whether the
	device is misbehaving or the host, or replace it with a supported
	device.
 | 
|  | 
.1>
        Last I knew, for the older version the SPD defined what
        was supported...  It might be useful to look through all
        the systems listed to see if it is supported on any of
        them.  If you can find one, then lack of support on one
        may be the lack of qualification instead of not being able
        to function correctly.  Also look in /sys/data/cam_data.c
        to see if it has an entry for the device.  If not, then
        you're dealing with what is basically a 3rd party device,
        even if it has Digital logo on it.
Thanks, Alan. I've looked through the SPD's for UNIX 3.2 - 4.0B and it's 
not listed. I've modified cam_data.c to add support for the real Exabyte 
8505 many times w/o problems. Also, the Systems and Options catalog for the 
AlphaServer 400 shows the TKZ15 as a "supported" device on 3.2C of Digital 
UNIX. The TKZ15 has never appeared in cam_data.c by default on Digital UNIX.
        I would start by looking at the detailed error logs to see
        if the problem is hardware or media.  Supported or not,
        devices to break now and then.  If it doesn't seem to be
        a hardware problem and the cam_data.c entry doesn't exist
        try writing one.  The comments in the file should be sufficient
        if you have good enough drive documentation (though I've never
        tried it).  If that doesn't help, check the USG hardware group's
        web page to see if there are any known problems with the device.
        After that either get out the SCSI analyzer and see whether the
        device is misbehaving or the host, or replace it with a supported
        device.
.2> 
The TKZ15 was a CSS device, so you might need to contact
them to find out what the support story is.
Thanks, Fred!!  I thought this maight be a CSS "special". I'm not actually
working this issue but the person that is will try CSS allright - by 
launching a red-hot IPMT their way!! This is one case where just a little 
effort by someone could have prevented Digital from looking pretty bad to 
this customer. 
If you buy an Exabyte 8505 from Exabyte, it works fine. Buy it from Digital,
to run on Digital hardware, and it does not work. Go figure! 
Thanks All,
Lee
    
 |