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

Conference ssdevo::hsj40_product

Title:HSJ30/40 Product Conference
Moderator:SSDEVO::EDMONDS
Created:Tue Jul 13 1993
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:1264
Total number of notes:4958

1261.0. "Battery Failure, controller failover and HSOF3.1" by KERNEL::BROOKER () Thu May 29 1997 09:21


	Is device and unit failover during battery failure supported in
	HSOF 3.1.  for HSJ40 controllers, and if so where is this documented?

	It's clearly stated in the 3.0 release notes for the HSZ40 and so 
	I suspect is also in the 3.0 HSJ40 release notes....which I don't
	think were ever released.

	Could someone confirm my suspicions please?

	Cheers,

	Clive.
T.RTitleUserPersonal
Name
DateLines
1261.1...read between the lines!!!KERNEL::LOANEComfortably numb!!Thu May 29 1997 11:2121
>	Is device and unit failover during battery failure supported in
>	HSOF 3.1.  for HSJ40 controllers, and if so where is this documented?

    It's  alluded  to  on page 11 of the Rev P01 release notes under the 
    heading "Low/Failed Battery on  Dual-Redundant  Configurations  with 
    Preferred   Paths",  although  the  content  of  this  `feature'  is 
    worrying!!

    If I read it correctly, the statement suggests that  customers  with 
    Controller-based  preferred  paths  set  will  STILL  suffer  device 
    availability problems (until they recognise the event and  turn  off 
    preferred pathing). 

    I suppose one way round this would be to:-                          
       Use FIBSCRIPT/HSDSA to set preferred pathing 
       Do some I/O to trigger device failover to the correct configuration
       Use FIBSCRIPT/HSDSA to turn off preferred pathing 

       ...and call the above at system startup and/or controller reboot

    Comments??
1261.2.....try it this way...KERNEL::LOANEComfortably numb!!Wed Jun 04 1997 08:0611
    OK...let's  ask  it a different way!! The Release notes state that a 
    cache battery transitioning to Low on a  Dual  Redundant  controller 
    with  Controller  preferred paths set "may go AVAILABLE on the OTHER 
    controller and INOPERATIVE on it".      ^
                                            |
    It's the word "may" that I'm interested in. 

    We  have  now got failover capability for Bad WBC Batteries, and I'd 
    feel uncomfortable leading customers  to  assume  that  ALL  battery 
    failures  will  result in Controller Failover if, in fact, Preferred 
    Paths could possibly prevent this!!.