On Tuesday, 03/12/2013 at 08:33 EDT, "Pavelka, Tomas" <[email protected]> wrote: > We have been trying to format all minidisks from Linux only and this turned out > to be problematic. I am looking for a solution that would let us stay in Linux > without having to involve CMS format for every new minidisk. Let me first > describe the problem: > When there is a record on dasd that has incorrect cylinder in the count area, > this leads to "record not found" errors when the dasd is brought online. Since > the dasd needs to be online before the problem is fixed (by formatting) the > only way around that I can see is to preformat in CMS. > If new minidisks are regularly formatted and destroyed, it is possible to run > into situation where part of the disk has the correct format and part has the > cylinder number in the count area wrong.
What you have identified isn't new, and it affects all operating systems. Any disk space that is decommissioned needs to be formatted in some way to avoid precisely this kind of problem. If you gave z/OS a volume with a valid label and VTOC, but the index pointed to garbage or near-garbage, bad things would happen. If there's something that looks like a valid label on the disk, then the OS will start looking further. At some point in that process, the OS finds a Magic Cookie and based on that decides that the disk has been formatted. It will *assume* that all internal pointers on the disk are valid and consistent. If there is no sign of a volume label, or the label indicates the hen the OS assumes the disk is NOT formatted and it won't read any further. If you're using minidisks, then you erase a decommissioned disk before you give it to another guest. You can have DIRMAINT do it automatically by putting DISK_CLEANUP=YES in your CONFIGxx DATADVH file. That way, if the old cylinder 0 gets re-used as the new cylinder 0, the guest sees an empty disk. If something other than cyl 0 get used, then the guest will see all zeroes (no label). Write garbage in the middle of a SCSI LUN and watch the OS go nuts. The biggest fear is that 'garbage' is semantically valid. Then you end up reading random data from the disk with no I/O errors - just bad data. Alan Altmark Senior Managing z/VM and Linux Consultant IBM System Lab Services and Training ibm.com/systems/services/labservices office: 607.429.3323 mobile; 607.321.7556 [email protected] IBM Endicott ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For more information on Linux on System z, visit http://wiki.linuxvm.org/
