Drive lifecycle states

Edit online

Manager monitoring reports states for data drives, healthy or not, per Slicestor® device. Alerts appear in the Event Console when user intervention is needed.

Table 1. Drive Lifecycle States
Drive State	Status LED Color	Disk Condition
ONLINE	Green	Healthy, usable.
MIGRATING	Yellow	Failed manually after quarantine. Triggers data move to other disks.
DIAGNOSTIC	Yellow	Temporarily inaccessible. Can be used for running diagnostics, tests, and so on.
FOREIGN	Yellow	Identified with a different Slicestor device and cannot be used in this Slicestor device.
OFFLINE	Gray	Unavailable to use, either physically removed or undetected by the system due to inconsistent hardware state.
UNUSABLE	Red	Kernel cannot initialize for use.
FAILED	Red	Disk is failed and needs to be replaced. It shows in the Failed FRU Report.
UNKNOWN	Gray	State undetermined. Call IBM® Customer Support for more assistance.
INIT	Gray	Formatting or reformatted. Can also appear briefly during upgrade or when a Slicestor device or process is restarted.

Drives can change through various states over the course of their useful life.

Note:

Disk Lifecycle operations should be performed if the dsnet-core process is running. Check the Monitor page for the Slicestor device. If the following message appears:

dsnet-core is expected to be running

but it is not with a red "x", wait for 15 minutes.

If this message does not go away after 15 minutes, contact IBM Customer Support.

Drive Impact on Device and Vault Health

As part of disk management, Slicestor disk health thresholds can be configured at the warning and error levels. The configuration of these thresholds can be performed for each model globally via the Settings tab or at the individual Slicestor device level in the Manager Web Interface. Defaults are defined for each model. The device and vault health are a function of these thresholds.

The following disk states contribute to the warning and error thresholds:

DIAGNOSTIC
MIGRATING
OFFLINE
FOREIGN
UNUSABLE

If the total number of non-healthy drives is greater than or equal to the warning threshold but less than the error threshold, the device health is set to yellow. If the total is greater than or equal to the error threshold, the device health is set to red.

Note: The vault health is affected if the error threshold is achieved.

Thus, the warning threshold does not impact the vault health.

With the aid of IBM Support, caching (1-wide) vaults can be created. The vaults are intended to be used along with a high reliability vault configuration within the same storage pool (although they can be used in a separate storage pool) for improving read performance. If the device health for a caching vault becomes red (due to disk failures or other issues), an incident appears stating the vault is unusable due to the number of accessible Slicestor devices that are falling below read/write threshold. Although data on the unhealthy drives is lost, it can still be obtained from the high reliability vault. Contact IBM Customer Support for more information, including reliability and performance tradeoffs.