EQLDISK-MIB :: eqlDiskStatus

MIB Reference — IPNetwork Monitor · Updated October 05, 2026

All MIBs › EQLDISK-MIB › eqlDiskStatus

eqlDiskStatus

Module: EQLDISK-MIB

OID (symbolic): EQLDISK-MIB::eqlDiskStatus

OID (numeric): 1.3.6.1.4.1.12740.3.1.1.1.8

Node type: OBJECT-TYPE

Type: INTEGER

Access: read-only

Description: Disk status will be on-line when the disk is being used in the raid set. Disk status will be spare when the disk is not currently being used by raid configuration but may be used when status of a currently on-line disk changes. Disk status will be failed when either it is faulted or when there is no power. Disk status will be offline when the disk doesn't fall into any of these other categories. Disk status will be alt-sig (foreign) when the disk belongs to a RAIDset from another EqualLogic system. The disk can be converted to a spare. Disk status will be too-small if drive is too small to be used. The drive cannot be converted to spare. Disk status will be history-of-failures if drive has a history of failures. The drive can be converted to spare. Disk status will be unsupported-version if drive label is from a later release. The drive can be converted to spare. Disk status will be unhealthy if drive is unhealthy. Disk status will be replacement if drive is a replacement drive. Disk status will be encrypted if the drive is encrypted and cannot be decrypted. Disk status will be notApproved if the drive is not DELL approved. Disk status will be preempt-failed if the drive is off-line due to excessive errors.

What is eqlDiskStatus?

The disk's current operational status: on-line(1) if actively part of the RAID set, spare(2) if held in reserve, failed(3), off-line(4), alt-sig(5) if it's a foreign disk from another EqualLogic system, too-small(6), history-of-failures(7), unsupported-version(8), unhealthy(9), replacement(10), encrypted(11) if it can't be decrypted, notApproved(12) if not a Dell-approved drive, or preempt-failed(13) if taken offline for excessive errors. Admins monitor this to catch a failed(3) or unhealthy(9) drive before it causes a RAID rebuild under pressure, and to recognize an alt-sig(5) drive carried over from another system before reusing it.

Examples

Walk all instances (SNMPv2c):

snmpwalk -v2c -c public <target> 1.3.6.1.4.1.12740.3.1.1.1.8
snmpwalk -v2c -c public <target> EQLDISK-MIB::eqlDiskStatus

Get a specific instance (index 1):

snmpget -v2c -c public <target> 1.3.6.1.4.1.12740.3.1.1.1.8.1
snmpget -v2c -c public <target> EQLDISK-MIB::eqlDiskStatus.1

Start monitoring Dell EqualLogic PS-series iSCSI SAN storage array (physical disk inventory/health) with a free 30-day trial of IPNetwork Monitor. Create custom SNMP monitor using the EQLDISK-MIB::eqlDiskStatus OID value, configure state conditions and alerts, and monitor any Dell EqualLogic PS-series iSCSI SAN storage array (physical disk inventory/health) from a single console.

OID Breakdown

Upper-level ancestors (7 from the standard OID tree / other modules)
Numeric OIDNameModule
1isoLANART-AGENT
1.3orgAirPair-MIB
1.3.6dodAirPair-MIB
1.3.6.1internetAirPair-MIB
1.3.6.1.4privateAirPair-MIB
1.3.6.1.4.1enterprisesAirPair-MIB
1.3.6.1.4.1.12740equalLogicEQUALLOGIC-SMI
Numeric OIDNameModule
1.3.6.1.4.1.12740.3eqldiskModuleEQLDISK-MIB
1.3.6.1.4.1.12740.3.1eqldiskObjectsEQLDISK-MIB
1.3.6.1.4.1.12740.3.1.1eqlDiskTableEQLDISK-MIB
1.3.6.1.4.1.12740.3.1.1.1eqlDiskEntryEQLDISK-MIB
1.3.6.1.4.1.12740.3.1.1.1.8eqlDiskStatusEQLDISK-MIB