< Previous | Next > | |
Product: Volume Manager Manual Pages for Storage Foundation | |
Manual: Maintenance Commands (1m) |
vxrelayoutNAMEvxrelayout - convert online storage from one layout to another SYNOPSISvxrelayout [-f] [-g diskgroup] [-o useopt] [-R request_portal] [-t tasktag] [-T taskid] [-U usetype] reverse volume vxrelayout [-f] [-g diskgroup] [-o useopt] [-R request_portal] [-t tasktag] [-T taskid] [-U usetype] start volume vxrelayout status volume vxrelayout help | -H DESCRIPTIONOnline relayout is a process where storage in a volume or a plex is converted from one layout to another maintaining data availability at all times. You can convert any supported VERITAS Volume Manager (VxVM) layout to another and change the stripe width or the number columns. You can also perform only stripe width and column changes. The transformation of a volume or a plex must be initiated using the vxassist(1M) utility. vxassist creates necessary infrastructure and storage needed to perform the layout transformation and invokes vxrelayout internally to actually perform the transformation. Note The relayout operation internally generates unique data object names (for example, v2-Dp02) when it converts a volume to a specified destination layout. Because vxrelayout depends on vxassist to do all the setup before a layout transformation can begin, the vxassist utility must be used to initiate any layout transformation. The vxtask(1M) utility can monitor the state of an ongoing transformation. Because these transformations are time consuming (mainly depending on the volume size and chunk size), they can be interrupted because of a system crash, I/O failure, or the user voluntarily stopping the transformation process either by aborting the task associated with the transformation using vxtask or by killing the transformation process. It is not advisable to use kill -9 to stop the transformations. The vxrelayout utility can be used to continue or reverse a transformation process because of such interruptions. VERITAS Volume Manager responses to possible interruptions are: vxrelayout utility exits with an error on STDERR. The vxrelocd(1M) daemon tries to relocate the failed subdisks. If it succeeds, it invokes vxrecover(1M), which in turn invokes vxrelayout to continue the transformation. An I/O failure is transparent to the user if the relocation succeeds. If relocation fails, the transformation process is not restarted. On a reboot after a system crash, vxrecover starts all startable volumes, invoking vxrelayout on appropriate volumes and plexes. If the user stops vxrelayout, then the transformation process can be reversed or continued by invoking vxrelayout manually. The first operand is a keyword that determines whether a layout conversion is to be continued, reversed, or to display the status of any existing transformation in the volume. For example, if a RAID-5 volume is being converted to STRIPED volume, reversal of such an operation will revert all data and storage in STRIPED format to the RAID-5 format. A reversal of a reversal is equivalent to a continue. Even though such a capability of direction change can be used ad infinitum, it has been provided with an objective to react to unforeseen situations. It is advisable that such direction changes are done with care. The direction of a transformation can be determined using the status operation provided in the vxrelayout utility. This command will determine if any plex in the volume is being transformed and display its characteristics. The second operand is the name of the topmost volume to which the plex undergoing relayout is attached. If there are multiple plexes in the volume and only one of them is being transformed, the vxrelayout utility will select the appropriate plex using the supplied volume name. It should be noted that there can only be one relayout operation happening in a volume at any instant of time. A specific disk group can be forced with -g diskgroup. This operation can be applied only to disk groups with version 60 or above (see vxdg(1M)), and only to one disk group at a time. The supported storage layouts are:
See the VERITAS Volume Manager Administrator's Guide for more information. KEYWORDSDisplays information on vxrelayout usage. Reverses a discontinued layout conversion. This operation determines the current state and direction of a layout transformation and reverse where it was left off since it was last stopped. It is imperative that any existing layout transformation process is stopped before this operation is applied. Continues a discontinued layout conversion. This operation determines the current state and direction of a layout transformation and continue where it was left off since it was last stopped. It is imperative that any existing layout transformation process is stopped before this operation is applied. Displays the status of a discontinued or an ongoing layout conversion. This operation displays the characteristics of the source and the destination layout, such as the layout, number of columns, and stripe width. It also displays whether the transformation is ongoing or stopped, and the actual relayout percentage completed. OPTIONSForces a continue or reverse of a layout transformation. This is considered potentially dangerous or questionable to use as using this flag may result in loss of data. This permits a limited set of operations that would otherwise be disallowed:
Specifies the disk group for the operation, either by disk group ID or by disk group name. By default, the disk group is chosen based on the name operands. Passes in usage-type-specific options to the operation. A certain set of operations are expected to be implemented by all usage types: Reduces the system performance impact of copy operations. Copy operations are usually a set of short copy operations on small regions of the volume (normally from 16 kilobytes to 128 kilobytes). This option inserts a delay between the recovery of each such region. A specific delay can be specified with iodelay as a number of milliseconds; otherwise, a default is chosen (normally 250 milliseconds). Performs copy operations in regions with the length specified by size, which is a standard VERITAS Volume Manager length number (see vxintro(1M)). Specifying a larger number typically causes the operation to complete sooner, but with greater impact on other processes using the volume. The default I/O size is 1 megabyte. Note These parameters can also be changed using the vxtask utility during an online relayout. Run the vxrelayout operation in the background. Specifies a rendezvous file path name for regular configuration and query requests. /etc/vx/vold_request is the default. If any tasks are registered to track the progress of the operation, marks them with the tag tasktag. The tag specified by tasktag is a sequence of up to 16 alphanumeric characters. Associates new tasks with the specified parent task ID. Limits the operation to apply to this usage type. Attempts to affect volumes with a different usage type will fail. EXIT CODESvxrelayout exits with a non-zero status if the operation fails. A non-zero exit code is not a complete description of the problem; it only indicates the first condition that prevented further execution. See vxintro(1M) for a list of standard exit codes. EXAMPLESCheck the progress of conversion of the volume homevol: vxrelayout status homevol Restart conversion of the volume homevol at a slow rate in the background (layout conversion having previously been stopped by using the pause keyword in vxtask): vxrelayout -o slow,bg start homevol Reverse conversion of the volume homevol (layout conversion having previously been stopped): vxrelayout reverse homevol SEE ALSOkill(1), vxassist (1M), vxdg (1M), vxintro (1M), vxrecover (1M), vxrelocd (1M), vxtask (1M) VERITAS Volume Manager Administrator's Guide |
^ Return to Top | < Previous | Next > |
Product: Volume Manager Manual Pages for Storage Foundation | |
Manual: Maintenance Commands (1m) | |
VERITAS Software Corporation
www.veritas.com |