We have a customized JumpStart process which does the following: - installs Solaris 10 FCS on V880's - creates metadevices of all slices on root disk - installs our software bundle - finally attach the mirrors
After the setup of metadevices i.e. after metadb's/metadevices created, and after running metaroot, a subsequent reboot produces the following error and stops the boot process: ------------------------------------------------------------------------------- mount: /dev/md/dsk/d503 is already mounted, /var is busy or the allowable number of mount points have been exceeded -------------------------------------------------------------------------------- I can successfully do a "mount /var" at the console i.e. don't have to perform an fsck or anything. This error occurs only after the first reboot after metadevices are created, subsequent reboots are fine.The above failure is repeatable (on mutiple servers) and it's not an abberation. My questions are: * Why does the boot process think that /var is mounted when we clearly know it is not? * How do I debug this? boot -m verbose? * If /var was corrupted, it should error out stating that, right? Thanks in advance. -- prasad This message posted from opensolaris.org
