This appendix contains information about Solaris Volume Manager files for reference purposes. It contains the following:
This section explains the files that are necessary for Solaris Volume Manager to operate correctly. With the exception of a few specialized configuration changes, you will not need to access or modify these files.
Do not edit this file. If you change this file, you could corrupt your Solaris Volume Manager configuration.
The /etc/lvm/mddb.cf file records the locations of state database replicas. When state database replica locations change, Solaris Volume Manager makes an entry in the mddb.cf file that records the locations of all state databases. See the mddb.cf(4) man page for more information.
The /etc/lvm/md.cf file contains automatically generated configuration information for the default (unspecified or local) disk set. When you change the Solaris Volume Manager configuration, Solaris Volume Manager automatically updates the md.cf file (except for information about hot spares in use). See the md.cf(4) man page for more information.
Do not edit this file. If you change this file, you could corrupt your Solaris Volume Manager configuration or be unable to recover your Solaris Volume Manager configuration.
If your system loses the information maintained in the state database, and as long as no volumes were changed or created in the meantime, you can use the md.cf file to recover your configuration. See How to Initialize Solaris Volume Manager From a Configuration File.
The md.conf configuration file is read by Solaris Volume Manager at startup. You can edit two fields in this file: nmd, which sets the number of volumes (metadevices) that the configuration can support, and md_nsets, which is the number of disk sets. The default value for nmd is 128, which can be increased to 8192. The default value for md_nsets is 4, which can be increased to 32. The total number of named disk sets is always one less than the md_nsets value, because the default (unnamed or local) disk set is included in md_nsets.
Keep the values of nmd and md_nsets as low as possible. Memory structures exist for all possible devices as determined by nmd and md_nsets, even if you have not created those devices. For optimal performance, keep nmd and md_nsets only slightly higher than the number of volumes you will use.
After changing the md.conf configuration file, you will need to do a reconfiguration reboot for the changes to take effect.
This file configures and starts Solaris Volume Manager at boot and allows administrators to start and stop the daemons.
This file checks the Solaris Volume Manager configuration at boot, starts resynchronization of mirrors if necessary, and starts the active monitoring daemon. (For more information, see mdmonitord(1M).)
The /etc/lvm/md.tab file contains Solaris Volume Manager configuration information that can be used to reconstruct your Solaris Volume Manager configuration. Solaris Volume Manager can use this file as input to the command line utilities metainit, metadb, and metahs to reconstruct a configuration. Volumes, disk sets, and hot spare pools might have entries in this file. See How to Create Configuration Files for instructions on creating this file (using metastat -p > /etc/lvm/md.tab).
The configuration information in the /etc/lvm/md.tab file might differ from the current volumes, hot spares, and state database replicas in use. It is used manually, by the system administrator, to capture the intended configuration. After you change your Solaris Volume Manager configuration, recreate this file and preserve a backup copy.
Once you have created and updated the file, the metainit, metahs, and metadb commands then activate the volumes, hot spare pools, and state database replicas defined in the file.
In the /etc/lvm/md.tab file, one complete configuration entry for a single volume appears on each line using the syntax of the metainit, metadb, and metahs commands.
If you use metainit -an to simulate initializing all of the volumes in md.tab, you may see error messages for volumes that have dependencies on other volumes defined in md.tab. This occurs because Solaris Volume Manager does not maintain state of the volumes that would have been created when running metainit -an, so each line is evaluated based on the existing configuration, if a configuration exists. Therefore, even if it appears that metainit -an would fail, it might succeed without the -n option.
You then run the metainit command with either the -a option, to activate all volumes in the /etc/lvm/md.tab file, or with the volume name that corresponds to a specific entry in the file.
Solaris Volume Manager does not write to or store configuration information in the /etc/lvm/md.tab file. You must edit the file by hand and run the metainit, metahs, or metadb commands to create Solaris Volume Manager components.
For more information, see the md.tab(4) man page.