|
APIs in Category: disk |
Cluster-Mode API version 1.15 |
disk-fail |
| disk-fail | [top] |
Fail a file system disk. Removes the specified file system disk from the RAID configuration, spinning the disk down when removal is complete. disk fail is used to remove a file system disk that may be logging excessive errors and requires replacement. Note that if optional input parameter 'is-immediate' is true, the specified disk will be immediately failed out, and the RAID group to which the disk belongs will enter degraded mode (meaning a disk is missing from the RAID group). If a spare disk at least as large as the disk being removed is available, the contents of the disk being removed will be reconstructed onto that spare disk. If 'is-immediate' options is false or not specified, system will prefail the disk and its content will be copied to a replacement disk if a suitable spare disk is available, and afterwards the prefailed disk will be failed out. This process can be observed by polling disk-list-info for this disk and tracking values of elements copy-destination and copy-percent. Same can be done using 'storage-disk-get-iter' when requested from Admin Vserver LIF. The disk being failed is marked as ``broken'', so that if it remains in the disk shelf, it will not be used by the filer as a spare disk. If the disk is moved to another filer, that filer will use it as a spare. This is not a recommended course of action, as the reason that the disk was failed may have been because it needed to be replaced. NOTE: Data ONTAP 7.0 and earlier releases don't indicate failure code properly.
Input Name Range Type Description disk string
Name of the disk to fail, e.g. "v0.1". On Admin Vserver LIF, this will be disk path name with format of " : ". e.g. "Filer01:v0.1" is-immediate boolean
optional
Specify 'true' if disk is to be failed out immediately. If disk is to be prefailed, specify 'false'. Default value is 'false'.
Errno Description EINTERNALERROR EDISKNOTFOUND ESWAPINPROGRESS ENOTDISKOWNER EOPNOTSUPPORTED EDISKINRAIDZEROVOL ENOTSTARTED EDISKISSPARE ESAVECOREDISK
| disk-remove | [top] |
Remove a spare disk. Removes the specified spare disk from the RAID configuration, spinning the disk down when removal is complete. You can use disk remove to remove a spare disk so that it can be used by another filer (as a replacement for a failed disk or to expand file system space). NOTE: Data ONTAP 7.0 and earlier releases don't indicate failure code properly.
Input Name Range Type Description disk string
Name of the disk to remove, e.g. "v0.1". On Admin Vserver LIF, this will be disk path name with format of " :Disk Name". e.g. "Filer01:2a.32"
Errno Description EINTERNALERROR EDISKNOTFOUND ENOTSTARTED EDISKNOTSPARE ESAVECOREDISK
| disk-sanown-assign | [top] |
Changes ownership on a disk. The normal usage is when there is no current owner or to assign a disk to a pool.
Input Name Range Type Description auto boolean
optional
Auto-assign unowned disks which are on loops where only 1 filer owns the disks and the pool information is the same. No other input parameters are allowed when specifying this option. checksum string
optional
Assign checksum type to disk. Option may only be specified on a RAID array LUNs. Possible values: 'block' or 'zoned'. disk string
optional
Name of disk to assign. Wildcarding for disk string is supported. force boolean
optional
Force flag need to be set to 'true' if assigning disks which are already owned a Filer. However, if that Filer is up and has put a reservation on the disk, even the force option won't work. owner string
optional
Assign disk to specific owner.Either owner or owner-id is mandatory owner-id integer
optional
Assign disk to specific owner ID (NVRAM ID or serial number). Either owner or owner-id is mandatory Range : [0..2^32-1]. pool integer
optional
Assign disk to specific pool, e.g. '0' or '1'.
Errno Description EINTERNALERROR EDISKNOTFOUND EINVALIDOWNER EINVALIDINPUTERROR
| disk-sanown-filer-list-info | [top] |
Get sanown filer information.
Input Name Range Type Description node string
optional
Node name for which info is requested Output Name Range Type Description disk-sanown-filer-details disk-sanown-filer-detail-info[]
List of filers and their information.
Errno Description EINTERNALERROR ESANOWNNOTENABLED
| disk-sanown-list-info | [top] |
Get sanown disk information.
Input Name Range Type Description disk string
optional
Get ownership status for given disk, if not supplied, get ownership for all disks or if onership-type is supplied, get ownership info for disks of specific type. '6a*' wildcarding is supported. node string
optional
Node name for which info is requested ownership-type string
optional
Possible values are 'all' which will list all disks. 'unowned' which will list all disks without owners. 'owned' which will list all disks with owners. 'visible' which will list all disks belonging to the local and partner filer. Default is 'all'. If specific disk is specified, ownership-type will be ignored. Output Name Range Type Description disk-sanown-details disk-sanown-detail-info[]
List of disks and their sanown information.
Errno Description EINTERNALERROR EDISKNOTFOUND
| disk-sanown-reassign | [top] |
Changes ownership on disks already belonging to an owner.
Input Name Range Type Description force boolean
optional
Force flag need to be set to 'true' if reassigning disks which are owned by another Filer. However, if that Filer is up and has put a reservation on the disk, even the force option won't work. In this case reassign will need to be run on the Filer which owns the disks. new-owner string
optional
Name of new owner. This form will assign all the disks belonging to the old owner to the specific owner. new-owner-id integer
optional
ID of new owner. This form will assign all the disks belonging to the old owner to the specific owner ID. Either new-owner or new-owner-id (or both) must be specified. Range : [0..2^32-1]. old-owner string
optional
Name of old owner. This form takes all disks currently belonging to the specific old owner, and reassigns them to a new owner. old-owner-id integer
optional
ID of old owner. This form takes all disks currently belonging to the specific old owner ID, and reassign them to a new owner. Either old-owner or old-owner-id must be specified, but not both. Range : [0..2^32-1].
Errno Description EINTERNALERROR EDISKFAILED
| disk-sanown-remove-ownership | [top] |
Removes ownership information on a disk. Default usage is to remove ownership information for all disks owned by the local node, in maintenance mode.
Input Name Range Type Description all boolean
optional
If 'true' remove ownership information from all disks. (maintenance mode only) disk-list disk-name[]
optional
List of disks to remove ownership information from. Example: 4a.18 5b.16 switch1:10.126L4 force boolean
optional
If 'true' override the RAID checks. Eg: Disk is part of aggregate. Otherwise defaults to false. owner string
optional
Remove ownership of all disks owned by owner. Either owner or owner-id is mandatory owner-id integer
optional
Remove ownership of all disks owned by this owner-id (maintenance mode only). Range [0..2^32-1]
Errno Description EINVALIDINPUTERROR EDISKNOTFOUND EINVALIDOWNER EINVALIDRESERVATION ENODISKSFOUND EMUSTBEINMAINTMODE EINTERNALERROR
| disk-unfail | [top] |
Unfail a disk in the broken pool, by clearing its FDR (Failed Disk Registry) entry and unfailing it at the Storage Layer, as necessary. If the "make-spare" option is set to B_TRUE, the disk is returned to the spare pool. Otherwise, label assimilation will bring the disk back according to its on-disk labels, with one of four possible outcomes. 1. Disk becomes a spare. This is the common case. The disk becomes a spare upon unfail, because its parent volume is complete and online. 2. Disk is assimilated into former volume. This is a recovery scenario. The disk is brought back into an existing volume, which may result in this volume coming back online. 3. Disk is assimilated into a new partial volume. This may occur in the rare case that the disk's former volume was destroyed or moved. 4. Disk returned to broken pool. This is the case if a fatal error occurs in process of unfailing the disk.
Input Name Range Type Description disk string
Name of the disk, e.g. "v0.1". On Admin Vserver LIF, this will be disk path name with format of " :Disk Name". e.g. "Filer01:2a.32" make-spare boolean
optional
Specify 'true' to force the disk to become a spare upon unfail. Default value is 'false'.
Errno Description EINTERNALERROR EDISKNOTFOUND
| disk-update-disk-fw | [top] |
Start disk firmware download process to update firmware on disks. This operation is asynchronous, and therefore returns no errors that might occur during the download process. This operation will only update firmware on disks that do not have the latest firmware revision. The firmware revision on the disk can be monitored via the disk-list-info API.
Input Name Range Type Description disk-list disk-name[]
optional
List of disks to be updated. If not specified, all disks are updated. Example: 4a.18 5b.16 switch1:10.126L4 node-name string
This field is only valid when the request is sent to the Admin Vserver LIF. This is the storage system name the command will be executed on.
Errno Description EINTERNALERROR EDISKNOTFOUND EINVALIDINPUTERROR ENODENOTFOUND
| disk-zero-spares | [top] |
Set up all non-zeroed spares owned by the filer to start zeroing. This operation is asynchronous, and therefore returns no errors that might occur when the zeroing operation actually starts, which could be several seconds after this API operation completes. Zeroing progress can be monitored via the disk-list-info API. Same can be done using 'storage-disk-get-iter' when requested from Admin Vserver LIF. The "zeroing-percent" element of disk-detail-info is returned if disk zeroing has started, and "is-zeroed" returns TRUE once the zeroing has completed (or, if zeroing wasn't necessary in the first place).
Input Name Range Type Description node-name string
optional
Name of the node on which spare disks to be zeroed When requested from Admin Vserver LIF and node name is not specified, spare disks from all nodes in the cluster will be zeroed.
Errno Description EINTERNALERROR
| Element definition: disk-name | [top] |
Disk path name
[none]
| Element definition: disk-sanown-detail-info | [top] |
Disk sanown information.
Name Range Type Description checksum string
optional
Only returned on RAID array LUNs. The possible values:
- "advanced_zoned" - advanced_zoned checksum (azcs),
- "block" - block,
- "zoned" - zoned.
home string
optional
The home of the disk. On NG will be different than owner if the home node has been taken over. On Classic this will be the same as owner. If the disk has no owner, this will not be returned. home-id integer
optional
ID (NVRAM ID) of disk home. Range : [0..2^32-1]. is-failed boolean
'true' if the disk is failed, 'false' otherwise. name string
Name of the disk, e.g. v1.1 Wildcarding for disk string is suuported, e.g. v1.* owner string
optional
Current owner of the disk. If disk has no owner, this will not be returned. owner-id integer
optional
ID (NVRAM ID) of owner if there is one. Range : [0..2^32-1]. pool integer
optional
Pool the disk belongs to, if it has a owner. reserved-by integer
ID (NVRAM ID) of node with the reservation on this disk, 0 if there is none. Range : [0..2^32-1]. serial-no string
Serial number of the disk. type string
optional
Indicates the type of disk or device being being reported. Values are one of following: "SCSI", "FCAL", "D_ATA", "E_ATA", "LUN", "SSD", "UNKNOWN"
| Element definition: disk-sanown-filer-detail-info | [top] |
Disk sanown filer information.
Name Range Type Description filer-id integer
ID (NVRAM ID) of owner. Range : [0..2^32-1]. filer-name string
optional
Name of the filer. If the filer name hasn't been written to the disks, this will not be returned.