metattach is used to add submirrors to a mirror, add logging devices to trans devices, grow metadevices, or grow soft partitions. Growing metadevices can be done without interrupting service. To grow the size of a mirror or trans, the slices must be added to the submirrors or to the master devices.
Solaris Volume Manager supports storage devices and logical volumes greater than 1 terabyte (TB) when Solaris is running a 64-bit kernel. Support for large volumes is automatic. If a device greater than 1 TB is created, Solaris Volume Manager configures it appropriately and without user intervention.
If a system with large volumes is rebooted under a 32–bit Solaris kernel, the large volumes are visible through metastat output. Large volumes cannot be accessed, modified or deleted, and no new large volumes can be created. Any volumes or file systems on a large volume in this situation are also be unavailable. If a system with large volumes is rebooted under a version of Solaris that lacks large volume support, Solaris Volume Manager will not start. You must remove all large volumes before Solaris Volume Manager runs under another version of the Solaris Operating Environment.
Solaris Volume Manager supports one-to-four-way mirrors. You can only attach a metadevice to a mirror if there are three or fewer submirrors beneath the mirror. Once a new metadevice is attached to a mirror, metattach automatically starts a resync operation to the new submirror.
Attaching a new logging device to a busy trans metadevice is allowed, although a trans metadevice starts using its new logging device only after the trans is idle (after it is unmounted, for example). The busy trans is in an Attaching state (metastat) until the logging device is actually attached. Attaching a logging device in the Hard Error or Error state (metastat) is not allowed.
metadetach is used to detach submirrors from mirrors and to detach logging devices from trans metadevices.
When a submirror is detached from a mirror, it is no longer part of the mirror, thus reads and writes to and from that metadevice by way of the mirror are no longer performed through the mirror. Detaching the only existing submirror is not allowed. Detaching a submirror that has slices reported as needing maintenance (by metastat) is not allowed unless the -f (force) flag is used.
metadetach also detaches the logging device from a trans. Once detached, the logging device is no longer part of the trans, thus the trans is no longer logging and all benefits of logging are lost. Any information on the logging device that pertains to the master device is written to the master device before the logging device is detached.
Detaching the logging device from a busy trans device is not allowed unless the -f (force) flag is used. Even so, the logging device is not actually detached until the trans is idle. The trans is in the Detaching state (metastat) until the logging device is detached.
Root privileges are required for all of the following options except -h.
The following options are supported:
Force the detaching of metadevices that have components that need maintenance or are busy. You can use this option only when a mirror is in a maintenance state that can be fixed with metareplace(1M). If the mirror is in a maintenance state that can only be fixed with metasync(1M) (as shown by the output of metastat(1M)), metadetach -f has no effect, because the mirrors must be resynchronized before one of them can be detached.
Display a usage message.
Specify the interlace value for stripes, where size is a specified value followed by either k for kilobytes, m for megabytes, or b for blocks. The units can be either uppercase or lowercase. If size is not specified, the size defaults to the interlace size of the last stripe of the metadevice. When an interlace size change is made on a stripe, it is carried forward on all stripes that follow.
Specify the name of the diskset on which the metattach command or the metadetach command works.. Using the -s option causes the command to perform its administrative function within the specified diskset. Without this option, the command performs its function on local metadevices.
The following operands are supported:
The logical name for the physical slice (partition) on a disk drive, such as /dev/dsk/c0t0d0s2, being added to the concatenation, stripe, concatenation of stripes, or RAID5 metadevice.
The metadevice name of the concatenation, stripe, or concatenation of stripes.
The metadevice name of the logging device to be attached to the trans metadevice.
The metadevice name to be attached to the mirror as a submirror. This metadevice must have been previously created by the metainit command.
The name of the mirror.
The metadevice name of the RAID5 metadevice.
The amount of space to add to the soft partition in K or k for kilobytes, M or m for megabytes, G or g for gigabytes, T or t for terabytes, and B or b for blocks (sectors).
The metadevice name of the existing soft partition.
The metadevice name of the submirror to be detached from the mirror.
The metadevice name of the trans metadevice (not the master or logging device).
This example concatenates a single new slice to an existing metadevice, d8. Afterwards, you would use the growfs(1M) command to expand the file system.
# metattach d8 /dev/dsk/c0t1d0s2
This example detaches the logging device from a trans metadevice d9. Notice that you do not have to specify the logging device itself, as there can only be one.
# metadetach d9
This example expands a RAID5 metadevice, d45, by attaching another slice.
# metattach d45 /dev/dsk/c3t0d0s2
When you add additional slices to a RAID5 metadevice, the additional space is devoted to data. No new parity blocks are allocated. The data on the added slices is, however, included in the overall parity calculations, so it is protected against single-device failure.
The following example expands a soft partition, d42, by attaching 150 Mbytes.
# metattach d42 150M
When you add additional space to a soft partition, the additional space is taken from any available space on the slice and might not be contiguous with the existing soft partition.
This example adds space to a two-way mirror by adding a slice to each submirror. Afterwards, you would use the growfs(1M) command to expand the file system.
# metattach d9 /dev/dsk/c0t2d0s5 # metattach d10 /dev/dsk/c0t3d0s5
This example detaches a submirror, d2, from a mirror, d4.
# metadetach d4 d2
This example adds four slices to an existing metadevice, d9. Afterwards, you would use the growfs(1M) command to expand the file system.
# metattach d9 /dev/dsk/c0t1d0s2 /dev/dsk/c0t2d0s2 \ /dev/dsk/c0t3d0s2 /dev/dsk/c0t4d0s2
See attributes(5) for descriptions of the following attributes:
|ATTRIBUTE TYPE||ATTRIBUTE VALUE|
metaclear(1M), metadb(1M), metadetach(1M), metahs(1M), metainit(1M), metaoffline(1M), metaonline(1M), metaparam(1M), metarecover(1M), metareplace(1M), metaroot(1M), metaset(1M), metastat(1M), metasync(1M), md.tab(4), md.cf(4), mddb.cf(4), attributes(5)
Solaris Volume Manager Administration Guide
This section provides information regarding warnings for devices greater than 1 TB and for multi-way mirrors.
Do not create large (>1 TB) volumes if you expect to run the Solaris Operating Environment with a 32-bit kernel or if you expect to use a version of the Solaris Operating Environment that lacks large volume support.
When a submirror is detached from its mirror, the data on the metadevice may not be the same as the data that existed on the mirror prior to running metadetach. In particular, if the -f option was needed, the metadevice and mirror probably do not contain the same data.