[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference ssag::ask_ssag

Title:Ask the Storage Architecture Group
Notice:Check out our web page at http://www-starch.shr.dec.com
Moderator:SSAG::TERZAN
Created:Wed Oct 15 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:6756
Total number of notes:25276

6362.0. "RZ29B,PMAZ-A & InvCmd?" by UTRTSC::VISSER () Thu Jan 30 1997 06:26

	This was reported to me and it confused me:

	VAXstation 4000 model 90A with DWCTX-BX and PMAZ-A (hw rev D07)
	and RZ29B-VA's running OpenVMS V6.2

	The 2x RZ29B-VA (firmware was 0007, needs 0016) operated without
	any problems on the embedded SCSI bus of the 4090A.

	Then the Turbo Channel adapter + PMAZ-A was added and the SAME
	RZ29B-VA's move to there. 
	Now it's impossible to use them: they report OFFLINE.

	Errorlog:  command is   1A 00 3E 00 FF 00  (mode sense page 0x3E)
		   X'd sense returned:
			3  2  1  0   (byte)
			00 04 00 70
			0A 00 00 00
			00 00 00 00
			C0 00 00 24   ASC=Invalid field in CMD Descr. Block
			      02 00   -> byte #2 in error  ('3E' in cmd)

	1. Should this combination work?   or Supported!!!!!
	   (I have my sever doubts it has ever been tested) 

	2. Is page 3E a valid page in the RZ29B?  
	   Or: what is wrong in the command to the RZ29B?
	   

	Idea's appreciated, I'm stuck now,
						Jan Visser

T.RTitleUserPersonal
Name
DateLines
6362.1works herePROXY::J_EVANSFri Jan 31 1997 07:378
    Jan,
    
    The VAXstation 4000 models 96 and 60 were tested with RZ29B's on both
    the internal and TC/PMAZ bus (all drives external of course) starting
    with fw rev 14 and then 16. I'd recommend you get the drive fw upgraded
    and try it. The model 90A is just a slower version than the 90A.
    
    jim
6362.2Thanks!UTRTSC::VISSERSun Feb 02 1997 14:277
    
    Jim,
    
    Thanks!
    I know which road I am going to take.
    
    			Jan
6362.316 will not fix it.SUBSYS::VIDIOT::PATENAUDEAsk your boss for ARRAY's...Mon Feb 03 1997 07:597
Read the blitz about version 0016, you won't see that error.

I was waiting for a VMS expert to reply first, but 3E looks bad. 3F I believe is
equal to "dump all pages" but 3E does not show up as a supported page.

roger.
6362.4So: Who's confusing who?UTRTSC::VISSERMon Feb 03 1997 14:4720
    
    Roger,
    
    Thanks for confirming 3E is indeed bad (I already had that feeling).
    
    I assume the command in the errorlog-entry (which looks valid and
    sane), which tells me 3E, REALLY means to say 3E.
    
    If this is right, VMS has gone banana's somehow. Leaves the question:
    Who confuses VMS? To exclude any hardware and firmware-issues, step ONE
    was to insure highest versions and exclude buggy hardware.
    
    And fight with the remainder of the problem after this.
    
    I will sniffer around in the VMSstuff,
    
    			Jan
    
    
    
6362.5Checked DKdriverUTRTSC::VISSERWed Feb 05 1997 14:119
    
    Indeed, the OpenVMS DKDRIVER uses 0x3F (all pages) and he requests for
    255 bytes.
    Checking if VAXSCSI patch-kit brings anything (if any)....
    
    We might need the good, old SCSI analyser out of the dust...
    
    			Jan.
    
6362.6Bad TCA!UTRTSC::VISSERFri Feb 14 1997 14:5812
    
    Problem was caused by a bad TCA (Turbo Channel adapter), sitting between
    system-board and PMAZ. 
    
    The TCA (a module with intelligence) was able to confude VMS in such
    a way (that's my view), that a ILLEGAL SCSI command byte (0x3E) was
    inserted in the Mode Sense Command.
    
    So: Who fooled who?
    Answer: TCA fooled us ALL!
    
    				Jan