Conference ssdevo::hsz40_product

846.0. "Dynamic disk replacement" by MSBCS::GRADY () Wed Apr 16 1997 14:30

    
    Hello,
    
    	    I have a SW800  cab with dual redundant HSZ40s' and 
    	    striped mirrorsets, no sparesets. 
    	    
    	    One disk in M1 went bad, I ran CFMENU and the bad disk showed 
    	    up in the PTL list not in the failedset area. In order to fix 
    	    this I had to delete the unit # then I could get to the stripset, 
    	    broke apart the stripeset so I could get at the mirrorset, 
    	    reconfigured the mirror to be a one member mirrorset then delete 
    	    the mirror.
    
            I then recreated M1 with two good disks, recreated the stripset,
            gave it a unit # and as soon as I hit enter the stripeset
            initialized.                               
    
            Question: Could I have physically replaced the bad disk then
            had the mirror of it just write to it? If so where do I find
    	    the command syntax?
    
            This way I would have had to move the data off the stripeset
            fix the mirror then move the data back on. (there was no data
            on these disks at the time).
            
                               			Thanks in advance
    
    							Dennis
    
    
    
    
    
    
    

Set mirrorset-name has all the legal CLI actions described.

    More than likely you have the failedset set to noautospare.  This
    problem should not occur.  If a drive failes in a 0+1 set then it
    should put this drive into a Failedset and await physical replacement. 
    At this time the drive can be removed and a new drive inserted.  No
    intervention should be required by anyone.
    
    To interogate hsz>show failedset full
    		  look and see the autospare function, if noautospare is
    set (this is the default) hsz>set failedset autospare
    
    Happy sparing.
    
    Paul