LSM is fault-tolerant and resolves most problems without system administrator
intervention.
If the volume daemon (vold) recognizes what
actions are being taken, it can roll a transaction forward or back.
If
vold
is unable to recognize and fix system problems, you need to
solve the problem.
This appendix describes the majority of informational, failure, and
error messages displayed by
vold
and the kernel driver.
These sections include some errors that are infrequently encountered and difficult
to troubleshoot.
Causes are included to elaborate on the situation or problem
that might have generated a particular message.
Wherever possible, a recovery
procedure (solution) is provided to locate and correct potential problems.
If you need to contact your customer support organization, these messages
are numbered for ease of reference.
7.1 Volume Daemon Error Messages
The following list contains the error messages associated with the volume
(vold) daemon.
-r must be followed by 'reset'
Cause: This message is caused by a usage error.
Solution: Correct the usage and try again.
-x
argument
: prefix
too long
Cause: The stub-mode device path prefix name supplied exceeded the maximum of 32 characters.
Solution: Select an alternate path for device files and retry the command.
-x
string
: invalid
debug string
Cause:
An unknown argument string was given to
the
-x
option to
vold.
Solution:
Select a valid string from the reference
page for
vold
and try again.
Usage: vold [-dkf] [-r reset] [-m mode] [-x
level] For detailed help use: vold help
Cause:
vold
was invoked with
an invalid set of arguments.
Solution:
Correct the usage and try again or type
vold help
for more help.
This is the full usage message from entering
vold help:
Usage: vold [-dkf] [-r reset] [-m mode] [-x level]
Recognized options:
-d set initial mode to disabled for transactions
-k kill the existing configuration daemon process
-f operate in foreground; default is background
-r reset reset kernel state; requires 'reset' option argument
-m mode set vold's operating mode
modes: disable, enable, bootload, bootstart
-x level set debugging level to <debug>, 0 turns off debugging
-R file set filename for client request rendezvous
-D file set filename for client diag request rendezvous
lsm:vold: Error: volume
volume
: Logging daemon killed by signal
signal_number
[ core dumped ]
Cause: Someone killed the logging daemon.
Solution:
If required, restart the daemon by entering
voliod logio.
lsm:vold: Error: /dev/volevent:
error_messagelsm:vold: Error: cannot open /dev/volconfig:
error_messagelsm:vold: Error: Cannot kill existing
daemon, pid=
process_id
Cause:
An attempt to kill an existing
vold
process with a SIGKILL signal has failed.
This might be due
to the process being in an unkillable kernel state perhaps because of a hung
I/O or a missing I/O interrupt.
There might be disk driver error messages
in the
/dev/osm
buffer.
Solution:
Try typing
cat /dev/osm
to see if any other messages have been sent to the console device.
If possible,
use
crash
to determine the state of the process.
If the
process is asleep waiting for an I/O completion, then any disk driver error
messages that have occurred might point to the solution.
Failing this, you
should restart the system.
lsm:vold: Error: /dev/voliod: VOL_LOGIOD_CHECK
failed
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then attempt to restart the system, possibly
followed by reinstalling the LSM software.
If this fails, contact Customer
Support.
lsm:vold: Error: /dev/voliod: VOL_LOGIOD_KILL
failed
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then attempt to restart system, possibly followed
by reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: All transactions are disabled
Cause:
This message might appear with the message
Disk group disabled by errors
if the disk group to be disabled is
the root disk group.
The continued use of the system could be dangerous because
any configuration changes required (including error handling cases) could
cause the loss of ability to perform I/O to a volume.
Because this includes
the root volume, this situation could, if uncorrected, cause the system to
hang.
Solution:
This is a fatal error.
All copies of
the bootable root disk have failed.
Recovery from this situation will require
booting from floppy or from a disk unconnected with the LSM software.
It might
then be necessary to remove the LSM bootable disk configuration by using the
volunroot
command.
See the LSM installation instructions for details.
Then you can reinitialize the root disk group to reestablish the database
and log areas.
lsm:vold: Error: Cannot get all disk groups
from the kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reinstalling the LSM software.
If this fails, contact Customer Support.
lsm:vold: Error: Cannot get all disks from
the kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Cannot get kernel transaction
state
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting the
vold
daemon.
If this fails then restart the system, possibly followed
by reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Cannot get private storage
from kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Cannot get private storage
size from kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Cannot get record
name
from the kernel:
error_message
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Cannot not make directory
directory_path
Cause:
When trying to create the specified directory,
vold
got a failure.
Solution:
Try creating the directory manually and then issue the
command
voldctl enable.
lsm:vold: Error: Cannot recover operation
in progressFailed to get group
disk_group
from the kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Cannot start
usage_type
volume, no valid complete plexes
Cause:
No usable plexes remain for either the root
or swap volume.
This error is fatal and will result in the system displaying
the message
System startup failed
and then shutting down.
Solution: This is generally an unrecoverable error and will likely require that you reload the system from backups.
lsm:vold: Error: Cannot start
usage_type
volume, no valid plexes
Cause
No usable plexes remain for either the root
or swap volume.
This error is fatal and will result in the system displaying
the message
System startup failed
and then shutting down.
Solution: This is generally an unrecoverable error and will likely require a reload of the system from backups.
lsm:vold: Error: Cannot start
usage_type
volume, volume state is invalid
Cause:
The volume is not in a state that can be
recovered from.
This might be because of corruption of the databases or because
of an invalid use of the
vold
interfaces without the use
of the utilities.
Solution: This is generally an unrecoverable error and will require reloading of the system from backups.
lsm:vold: Error: Cannot store private storage
into the kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Differing version of vold
installed
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Disk
disk,
group
disk_group, device
device_name
: not updated with new host ID Error:
error_message
Cause:
If the host ID for a system is changed using the
voldctl init
command then all disks in all imported disk groups
will need to have the host ID changed to the new ID.
If the host ID for a
disk cannot be changed, then this message will be displayed.
Other problems
might also exist for this disk.
Solution: Move the contents of the disk elsewhere and reinitialize the disk.
lsm:vold: Error: Disk group
disk_group, Disk
disk
: Cannot auto-import
group:
error_message
Cause: The disk group disk_group could not be reimported after a system restart. The reason is included in the error message. Other error messages might appear that provide more information on what went wrong. Any volumes in the disk group will be unavailable until you fix the error condition and reimport the disk group.
Solution:
Clear the error condition, if possible,
and then import the disk group manually with
voldg import.
After importing, restart all volumes with
voldg -g
groupname
-sb.
lsm:vold: Error: Disk group
disk_group, Disk
disk
: Group name
collides with record in rootdg
Cause: The disk group name disk_group, for the disk group being imported from the named disk, collides with a configuration record in the rootdg disk group. Disk groups must have names that do not match any records in the root disk group.
Solution: If you want to import the disk group, rename the conflicting record in rootdg to some other name.
lsm:vold: Error: Disk group
disk_group
: Cannot recover temp database error_message
Cause: The temp database stored in the root file system could not be opened or read. Other messages will detail the error. This might happen because of an I/O error or a problem in the file system.
Solution: Restart the system and retry the operation.
lsm:vold: Error: Disk group
disk_group
: Disabled by errors
Cause: This message can appear if the last configuration database or last kernel log area for a disk group became disabled. This could have been due to an I/O error or some other condition. Other messages preceding this one are likely to highlight the root cause.
Solution: Back up any remaining active volumes. Reinitialize the disk group and add the disks back to the group to recover.
lsm:vold: Error: Disk group
disk_group
: Errors in some configuration copies:
Cause: One or more on-disk database copies were found to contain errors. As a result, the disk group could not be imported. This is probably due to a disk I/O error, or to blocks of a configuration copy being overwritten within invalid contents. Check for messages from the disk driver. Errors pertaining to specific configuration copies are listed on successive lines. These lines can be in either of the following forms:
File filename : error_message : \ Block number : error_message Disk diskname, copy copy_number : \ error_message : Block number : error_message
Lines beginning with
File
indicate an error in the special configuration copy file used for storing
nonpersistent disk group information.
Lines beginning with
Disk
indicate failure of a persistent configuration copy stored on a disk.
The
copy number indicates which of the disk's configuration copies contains the
error.
Solution:
If one or more disks for the disk group
are currently inaccessible (such as due to a cabling error), make the disks
accessible and try to import the disk group again with
voldg import.
Otherwise, the disk group is probably no longer usable and you
will have to recreate it.
All volume configuration information for the disk
group is lost.
lsm:vold: Error: Disk group
disk_group
: Reimport of disk group failed:
error_message
Cause: The reload of a disk group into the kernel failed. This could be because the log size for the kernel might not be set or because of some other error in the import procedure. Other messages should indicate the true cause of the failure.
Solution: Retry the operation unless some other error message leads to a suggested course of action. If this fails, restart the system.
lsm:vold: Error: Disk group
disk_group
: update failed:
error_message
Cause: This message occurs because a database update failed completely. No complete copy of the database could be written for the disk group. The disk group will be disabled and further access for configuration changes will be disallowed. If this error occurs for the root disk group, it will probably be necessary to reinstall the system.
Solution: Back up any volumes still active in the disk group. Reinitialize the disk group and add the disks back to it.
lsm:vold: Error: Exec of /sbin/voliod failed
Cause:
An
exec
of
/sbin/voliod
failed.
Solution:
Check the existence and permissions of
the
/sbin/voliod
command.
Try executing the command manually
to ensure that it can be run.
lsm:vold: Error: Failed to store commit status
list into kernel:error_message
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Fork of logio daemon failed
Cause: The creation of a process that could then be used as a logging daemon failed.
Solution:
Check for messages explaining the reason
that a
fork(2)
call failed.
Retry the operation.
lsm:vold: Error: GET_VOLINFO ioctl failed:
error_message
lsm:vold: Error: Version number of kernel does not match vold
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Get of current rootdg failed
Cause:
An attempt to retrieve the rootdg from the
kernel failed.
This might be because of a kernel
vold
inconsistency
or could also be because of a version difference between
vold
and the kernel.
Solution:
Check that the correct version of
vold
and the kernel are installed.
Other messages might suggest
other problems in a prior attempt at loading a configuration and possible
courses of action.
If this fails, contact Customer Support.
lsm:vold: Error: No convergence between root
disk group and disk list Disks in one version of rootdg:
disk
type=
disk_type
info=
disk_info
Disks in alternate version of rootdg:
disk
type=
disk_type
info=
disk_info
Cause:
This message can appear when
vold
is not running in autoconfigure mode (see the
vold(8)
reference
page) and when, after several retries, it cannot resolve the set of disks
belonging to the root disk group.
The algorithm for disks that are not autoconfigured
is to scan disks listed in the
/etc/vol/volboot
file and
then examine the disks to find a database copy for the rootdg disk group.
The database copy is then read to find the list of disk access records for
disks contained in the group.
These disks are then examined to ensure that
they contain the same database copy.
As such, this algorithm expects to gain
convergence on the set of disks and the database copies contained on them.
If a loop is entered and convergence cannot be reached, then this message
will appear and the root disk group importation will fail.
Solution: Reorganizing the physical locations of the devices attached to the system might break the deadlock. If this fails, contact Customer Support.
lsm:vold: Error: Open of directory
directory_path
failed
Cause:
When
vold
was trying
to create node files for the volumes, it was unable to open the directory
in which the nodes were to be created.
Solution:
Check for other errors that suggest why the directory
might be missing or if the permissions might be incorrect.
Fix the condition
to allow
vold
to open or create the directory, then issue
the command
voldctl enable.
lsm:vold: Error: Read of directory
directory_path
failed
Cause:
The node directory could not be read when
vold
was trying to scan for volume nodes.
Solution:
Check for other messages that might suggest
why the directory is inaccessible.
Try reading the directory manually if the
directory is corrupted, then try removing and recreating it and then restarting
vold.
lsm:vold: Error: Unexpected configuration
tid for group
disk_group
found in kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Unexpected error during
usage_type
volume reconfiguration:
error_message
Cause: A record lock for the volume could not be acquired as part of the initial volume setup for either a root or swap volume. This is most likely to occur under low memory conditions.
Solution: Other messages might suggest an alternate course of action. Otherwise, this is generally an unrecoverable error and will require that you either boot off an alternate root device or reload the system from backups.
lsm:vold: Error: Unexpected error fetching
disk for
usage_type
volume:
error_message
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: Error: Unexpected values stored
in the kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: Error: VOL_RESET_KERNEL failed:
a volume or plex device is open
Or:
lsm:vold: Error: VOL_RESET_KERNEL failed:
error_message
Cause:
An attempt at resetting the kernel state
with a
vold
-r
reset
command failed because
all the LSM objects in the kernel were not closed.
If any volumes are in use,
then the reset cannot be performed.
This might also happen if a reset was
requested on a system with root volumes.
Root volumes are, by definition,
never closed and so a reset cannot be performed.
Solution: If you really want to reset the kernel, then check the state of the volumes and any mounted file systems to display information about who might have them open. Unmount all volumes and kill any processes accessing the volumes, then reset the kernel.
lsm:vold: Error:
mode
: Unrecognized operating mode
Cause: An unknown mode string was entered following a -m option.
Solution:
Select a valid mode from the
vold(8)
reference page and try again.
lsm:vold: Error: cannot open /dev/voliod:
error_message
Cause:
The open of the
/dev/voliod
file can fail only if the device node is missing or has an incorrect major
or minor number.
Solution: Check the existence and values of the file and make sure that the LSM software was correctly installed.
lsm:vold: Error: cannot open
argument
:
error_message
Cause: The tracefile specified on the command line could not be opened in append mode. The error message supplied should explain the reason.
Solution: Select an alternate tracefile name that can be created or appended to.
lsm:vold: Error: cannot open
volconfig_device
: Device is already open
Or:
lsm:vold: Error: cannot open
volconfig_device
:
error_message
Cause
The exclusive open device (/dev/volconfig) is already open.
Only one
vold
process can
be active on the system at one time.
Subsequent attempts at starting
vold
or opening the device will result in this message.
Solution:
Check for other running
vold
processes.
The
voldctl mode
command will report if
vold
is currently active.
lsm:vold: Error: enable failed:
error_message
Cause:
This message might occur during an initial
startup of
vold.
If changing to enabled mode when this
error occurs, failures could be due to problems with the creation of the portal
or with connection to the kernel.
If changing from an enabled state to a disabled
state, then problems could occur with removing the disk groups from the kernel
because of such things as volumes in use.
Solution: Evaluate other error messages occurring with this one to determine the root cause of the problem. Make changes suggested by the other errors and then retry the command.
lsm:vold: Error: failed to create daemon:
fork failed:
error_message
Cause:
The call to
fork(2)
to generate
a background
vold
process failed.
Solution:
Check for messages explaining the reason
that a
fork(2)
call failed.
Retry the operation.
lsm:vold: Error: volume
volume
: Wait for logging daemon failed
Cause:
The wait called to wait for the existence
of the daemon process did not execute correctly.
This can happen only if the
ioctl
does not correctly match the command required, perhaps because
of a mismatch between the
voliod
command and the kernel
versions or perhaps because of an incorrect minor number for the
/dev/voliod
device.
Solution:
Check the existence and permissions of
the
/dev/voliod
device.
lsm:vold: FATAL Error: Disk group rootdg:
Inconsistency -- Not loaded into kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: FATAL Error: Group
disk_group
: Cannot update kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: FATAL Error: Interprocess communication
failure:
error_message
Cause:
The portal to client utilities has returned
a failure.
This is a fatal error because without a portal to clients,
vold
cannot do anything useful.
Solution:
Check for other errors suggesting the
reason for portal failure.
Restart
vold.
If problems persist,
restart the system.
lsm:vold: FATAL Error: Invalid status stored
in kernel
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring of LSM.
If this fails, contact Customer Support.
lsm:vold: Warning: Cannot create device
path
:
error_message
Cause:
The
mknod(2)
call made by
vold
to create a device node failed.
The reason for the error should be displayed.
Solution:
Fix the reason indicated for node creation failure and
then issue the command
voldctl enable.
lsm:vold: Warning: Cannot exec /sbin/rm to
remove
directory_path
:
error_message
Cause:
An
exec
of
/sbin/rm
failed.
Solution: Ignore the error. It is not serious if the directory cannot be removed.
lsm:vold: Warning: Cannot fork to remove directory
directory_path
:
error_message
Cause:
The call to
fork(1)
to generate
a process that could then
exec
rm(2)
failed.
Solution: Ignore the error. It is not serious if the directory cannot be removed.
lsm:vold: Warning: Disk
device_name
in kernel not a recognized type
Cause:
The disk type of a disk in the kernel does
not match any known disk type.
This can occur only if
vold
and the kernel are in an inconsistent state.
Solution:
Try stopping and restarting
vold.
If this fails then reconfigure LSM.
If this fails, contact
Customer Support.
lsm:vold: Warning: Disk
disk
names group
disk_group, but group
ID differs
Cause: As part of a disk group import, a disk was discovered that had a mismatched disk group name and disk group ID. This disk will not have been imported. This can happen only if two disk groups of the same name exist that have different disk group ID values. In that case, one group will be imported along with all its disks and the other group will not. This message will appear for disks in the unselected group.
Solution: If it turns out that the disk should be imported into the group, add the disk to the group later. It will not happen automatically as part of the import. All configuration information for the disk will also be lost.
lsm:vold: Warning: Disk group
disk_group
is disabled, disks not updated with new host ID
Cause:
If the host ID for a system is changed using the
voldctl init
command then all disks in all imported disk groups
will need to have the host ID changed to the new ID.
If a disk group is found
in the imported but disabled state, then the host ID will not be changed.
Solution:
Clear the host ID using the
voldisk clearimport
command for each disk, and then reimport the
disk group.
lsm:vold: Warning: Disk group
disk_group
: Disk group log may be too small Log size should
be at least
number
blocks
Cause: The log areas for the disk group have become too small for the size of configuration currently in the group. This should usually never happen without first displaying a message about the database area size. This message occurs only during disk group import; it occurs if the disk was inaccessible while new database objects were added to the configuration, and the disk was then made accessible and the system restarted.
Solution:
If this situation does occur, then reinitialize
the disks in the group with larger log areas.
See the
voldisk(8)
reference
page for more information.
To reinitialize all the disks, detach them from
the group with which they are associated, and then reinitialize and readd
them to the disk group.
Deport and reimport the group for the changes to the
log areas for the group to take effect.
lsm:vold: Warning: Disk group
disk_group
: Errors in some configuration copies:
Cause:
One or more on-disk database copies were
found to contain errors.
As a result, the disk group could not be imported.
This is most likely to be due to a disk I/O error, or to blocks of a configuration
copy being overwritten with invalid contents.
Check for messages from the
disk driver.
Providing that other copies of the database can be successfully
read, the system will continue and the disk group import or initial
vold
enable operation should succeed.
If the database copy can subsequently
be written to, then this message will not recur.
Errors pertaining to specific
configuration copies are listed on successive lines.
These lines can be in
either of the following forms:
File filename : error_message : Block number : error_message Disk diskname, copy copy_number : error_message : \ Block number : error_message
Lines beginning with
File
indicate an error in the
special configuration copy file used for storing nonpersistent disk group
information.
Lines beginning with
Disk
indicate failure
of a persistent configuration copy stored on a disk.
The copy number indicates
which of the disk's configuration copies contains the error.
Solution: This message is likely to occur once due to an I/O failure and then not reoccur. If it does reoccur, then it might be necessary to remove the disk and reinitialize it to clear the condition. If all configuration copies for a disk group become unusable, then the disk group itself becomes unusable and must be recreated. If the rootdg disk group becomes unusable, you might need to reconfigure LSM. In this case, if the root file system is on a volume, then you might need to reinstall the operating system itself.
lsm:vold: Warning: Error in volboot file:
error_message
Entry: disk
disk disk_type disk_info
Cause:
This message occurs when an entry in the
volboot
file does not contain the correct information to define
a valid disk access record.
Solution:
Remove the entry using the
voldctl rmdisk
command and add it again using
voldctl adddisk.
lsm:vold: Warning: Failed to update voldinfo
area in kernel:
error_message
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Warning: Field too long in volboot
file: Entry: disk
disk disk_type
disk_info
Cause:
The
volboot
file is maintained
by
vold
and
voldctl
and should never
normally exhibit this problem.
This problem might indicate some corruption
of the
volboot
file or could also be the result of manual
editing of the file.
Solution:
Try to remove the offending entry with
the
voldctl rmdisk
command.
If this fails, you might have
to reinitialize
volboot
using the
voldctl init
command.
lsm:vold: Warning: Get of record
record_name
from kernel failed:
error_message
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Warning: Plex
plex
for
usage_type
volume is stale or
unusable
Cause: This message is output to alert you to the failure of one or more plexes of either the root or swap volume. The system might be able to continue depending on the existence of other usable plexes for the volume.
Solution: Repair the failed plex by either reattaching the plex to the volume once the system is booted, or by evacuating and replacing the disk on which the failed plex resides if you think that the disk is going bad.
lsm:vold: Warning: cannot remove group
group_id
from kernel:
error_message
Cause:
Some inconsistency between
vold
and the kernel has caused an
ioctl
to fail.
This could be caused by the use of older versions of
vold
or the kernel, or it could be due to a bug in LSM.
Solution:
Try stopping and restarting
vold.
If this fails then restart the system, possibly followed by
reconfiguring LSM.
If this fails, contact Customer Support.
lsm:vold: Warning: response to client
client number
failed:
error_message
Cause:
The portal to client utilities has returned
a failure.
This is a fatal error because without a portal to clients,
vold
cannot do anything useful.
This could be caused by a STREAMS
error or some other communications problem with the client.
Solution:
Check for other errors suggesting the
reason for portal failure.
Restart
vold.
If problems persist,
restart the system.
The following are the kernel level error messages.
NOTICE:
message
on volume device
hex_device_number
(
volume
) in diskgroup
disk_group
Cause:
This is caused by a driver above the LSM
level calling the LSM
volprint
function.
This usually happens
when a driver detects some error condition in LSM and want to display the
error.
Solution: No action necessary, unless specified in a supplied string.
NOTICE: io/vol.c(volerror): Correctable
type
error on volume
volume, plex
plex, block
block_number
Cause: A correctable I/O error was detected and corrected. A correctable I/O error is one where a read error from an underlying device driver could be corrected by reading the data from an alternate mirror copy and then writing it back to the failed mirror.
Solution:
If the I/O could have been completed
by reading from an alternate mirror but the writeback to the failed mirror
still failed, the mirror will be detached.
This failure will cause the exception
handling code to recover the volume according to its error recovery policy.
This usually results in either a mirror or the volume becoming detached.
You
must reattach the mirror (volplex att), to bring back the
failed mirror copy.
If the volume was detached, then the data contained on
it is unrecoverable and will have to be restored from backups.
NOTICE: io/vol.c(volerror): Uncorrectable
type
error on volume
volume, plex
plex, block
block_number
Cause: Following an I/O error from one mirror, an attempt to reread the data from an alternate mirror failed. This could be because no other mirrors exist or could be because the other mirrors also had I/O failures.
Solution:
This failure will cause the exception
handling code to be entered, which will result in the volume's error recovery
policy being followed.
This can have effects ranging from detaching a mirror
to disabling the volume.
You must reattach the mirror (volplex att), to bring back the failed mirror copy.
If the volume was detached,
then the data contained on it is unrecoverable and you will have to restore
it from backups.
NOTICE: lsm: Can't open device
disk, device busy or inaccessible.
Cause: The named disk cannot be accessed.
Solution: Turn on the drive.
WARNING: io/vol.c(volexcept): No volume error
daemon - Cannot Log plex detach, detaching volume
Cause:
No
voliod
process was
running and able to log a detach record for a mirror that is being detached
due to an I/O error.
This is a fatal error that causes future access to the
volume to be rejected, because any system failure coming after additional
I/O would not be able to detect the failure of the mirror and mirror inconsistencies
might then occur.
Solution:
Although it is too late to rescue this
volume, you should start at least one
voliod
process as
soon as possible (using
voliod set 2).
Stop and restart
the failed volume, then restore the data from backups.
Mirrors will have become
inconsistent and so any attempt at using the data on the volume could prove
disastrous.
WARNING: volklog_dgfree: Can't clear group
commit log record for group
disk_group
Cause: This can occur if a log flush to disk could not be performed because no valid log copies remained. This is likely to compromise the ability of the LSM to recover from any further I/O errors.
Solution: When you add disks to the system, ensure that new viable logging areas can be generated. Alternatively, remove failed disks and replace them with working devices.
WARNING: volklog_dgfree: Can't free kernel
logging area for vol_reset_kernel of group
disk_group
Cause: A free of the logs for a disk group failed because either no valid log areas remained for flushing or some log records remained in the log before the clear operation was requested.
Solution: There is no action to take here; this is an LSM internal error. Contact Customer Support.