NAME
vx_emerg_start — start the VERITAS Volume Manager (VxVM) from recovery media
SYNOPSIS
vx_emerg_start
[-m]
[-r
root_daname]
hostname
DESCRIPTION
The
vx_emerg_start
utility starts the VERITAS Volume Manager from alternate media.
The usual reason for
doing so is to allow repair of a rootable Volume Manager
configuration in the event of a catastrophic failure.
Options
vx_emerg_start
takes the following options and arguments:
- -m
Mount the root filesystem contained on the the
rootvol
volume after the Volume Manager has been started.
Prior to being mounted, the rootvol is started and
fsck(1M)
is run on the root filesystem.
- -r root_daname
Specify the Disk Access name of one of the members of the desired
rootdg
disk group.
This option can be used to specify the appropriate
rootdg
when multiple generations of
rootdg
disk groups exist on the system under repair.
If this option is not specified, the default is to pick the
rootdg
with the most recent timestamp when there is more than 1
rootdg.
- hostname
Specify the system name of the host, also called
nodename,
of the system being repaired.
This name is used to allow importation of the desired
rootdg.
It must match the name of the system being repaired,
as it would most likely not be recorded
on the recovery media from which you are booted.
NOTES
After the volume manager has been started, various recovery
options can be performed.
Exactly what recovery action needs to be performed depends on what is wrong.
It is recommended that
vxprint(1M)
be used to determine the configuration state.
One common problem that can usually be repaired is when
all mirrors of the rootdisk has stale plexes.
This would appear in the configuration like this:
v rootvol root DISABLED 393216 - ACTIVE -
pl rootvol-01 rootvol DISABLED 393216 - STALE -
sd rootdisk01-02 rootvol-01 ENABLED 393216 0 - -
pl rootvol-02 rootvol DISABLED 393216 - STALE -
sd rootdisk02-02 rootvol-02 ENABLED 393216 0 - -
When this happens, the volume can usually be repaired by performing:
vxvol -g rootdg -f start rootvol
If the volume is mirrored, it will be put in read-write-back recovery mode.
Since we put the command in the foreground,
it will not complete until the recovery is complete.
Once recovery is complete, you should check the root filesystem with
fsck(1M)
and mount it before you reboot:
fsck -F vxfs -o full /dev/vx/rdsk/rootdg/rootvol
mkdir /tmp_mnt
mount -F vxfs /dev/vx/dsk/rootdg/rootvol /tmp_mnt
Printable version
