[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

6495.0. "total blocks discrepency on V27Z-2 & VMS" by PTOSS1::CYPHER () Wed Mar 19 1997 08:46

    I have a customer who has a displayed-capacity discrepancy on some of
    their HSZ40 V27Z-2 RZ28D drives. They have 6 mirrorsets, each with two
    drives. HSZ CLI "show disk full" sizes the subject drives as having
    4109470 blocks. OpenVMS 6.2-1H3 "show device full" displays total
    blocks as 4108470, 500 blocks smaller, at the operating system level.
    I know this is backwards of the case in the blitz, and I even think I 
    know how this happened. My question is "will this hurt anything"? The
    operating system seemingly will simply waste 500 blocks, because it
    doesn't know they exist. Am I correct in this theory? 
    
    Dennis
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
T.RTitleUserPersonal
Name
DateLines
6495.1Save Configuration uses 500SSDEVO::RMCLEANWed Mar 19 1997 09:552
Do you have SAVE_CONFIGURATION enabled.  That is 500 blocks that the
operating system won't see.
6495.2Reserved but not used, apparantlyPTOSS1::CYPHERFri Mar 21 1997 14:1214
    These drives were all init'd w/save_config flag. I needed to move half
    of the dual mirrorsets (3 sets) to a less crowded shelf. I did that by
    failing out a member and moving it to a new spacious location on the
    same controller, different shelf. I then added the drive back into the
    mirrorset, waited completion, and broke the other out, moving it the
    same way. When the operations were complete, SHOW DISK FULL has no
    indication of "save_config" status ("not being backed up", therefore),
    excepting the size discrepency. I definitely feel the 500 block
    save_config area is reserved, but the controller is ignorant of it's
    existance. Very interesting. This should cause no problem; just
    curious.
    
    Cheers,
    Dennis
6495.3Yup this is the same as the BLITZSSDEVO::RMCLEANMon Mar 24 1997 10:217
There is a BLITZ that covers this issue.  The problem is that if you init your
file system after you have re-booted you will get overlap between the save
configuration space and the file system.  The way you have it now is
perfectly safe. 

You really should consider upgrading to a later software version which does
not exhibit this problem.