System Pool Configuration, Modes, and Alerts in OneFS 9.15+

1) All-Flash System Pool Recommendation Alert

This alert appears when:

• The cluster contains SSD‑capable node groups, but the system pool is currently using hybrid or HDD‑based node groups instead.

2) System Pool Capacity Threshold Alert

This alert appears when:

• The system pool's available capacity share falls below the minimum recommended threshold (typically 2.8% for SSD and 1.5% for HDD).
• The capacity share is the ratio of the available space in system pool to the total available space in cluster.
  i.e.

3) Degraded System  Pool Alert

This alert appears when:

• when the system is relying on a group of nodes that no longer has enough nodes to operate safely (fewer than three).

• This may happen when:

  • A node in a system pool has been smartfailed or removed

  • Node hardware failures reduce the active node count in a system pool below the minimum threshold

1) SSD System Pool Recommendation Alert

   Review the active mode:

   isi storagepool settings view

   To prioritize SSDs, switch to:

   isi storagepool settings modify --system-pool-mode=all-flash

When the System Pool Strategy is set to all-flash, OneFS will automatically select flash nodes in the system pool when available, and the alert gets resolved on the next CELOG evaluation cycle.

Note: If administrators intentionally choose to keep non‑flash nodes in the system pool, the alert can be acknowledged or ignored in the WebUI.
(WebUI → Cluster Management→ Events and alerts→ Locate the event in the list, then use the available Resolve or Ignore actions)


 

2) System Pool Capacity Threshold Alert 

   Verify nodepool capacity (WebUI):

       Navigate to:

      WebUI → File System → Storage Pools → SmartPools

      Under the listed nodepools, review Capacity:

• Used capacity
• Available capacity
• Nodepool type

Add capacity, add nodes, or rebalance data where needed to restore healthy available capacity levels. 
Also review current system‑pool mode to verify whether a different mode better suits cluster's layout and capacity distribution:

  isi storagepool settings view

For example:

• Switching to all node may provide a larger set of eligible nodepools when SSD/hybrid pools are low on space.
• Keeping the mode as all-flash ensures performance prioritization but may require additional SSD capacity.

If a nodepool within the System Pool is nearing capacity, administrators should take appropriate capacity‑management actions (such as expansion or rebalancing).
For detailed guidance on handling capacity‑related issues, refer to the appropriate Dell Knowledge Base article: https://www.dell.com/support/kbdoc/en-us/000356789/clustername-1-cluster-is-running-low-on-free-space-free-space-1-99xxx-job-config-core-free-blocks-pct-threshold-lo-2?lang=en

3) Degraded System Pool Alert

   Check node health and cluster state:

   isi status

Restore or replace failed/smartfailed nodes, or add nodes of the same type to return the group to at least 3 nodes.

Additionally, review the active system‑pool mode:

   isi storagepool settings view

• If certain pools remain degraded or specific node types are unavailable, consider all node to let OneFS select healthier, non‑degraded pools across media types.
• If staying with ssd or hybrid for policy/performance reasons, prioritize restoring nodes or adding capacity in those media classes.

If the node is unavailable, add a replacement node to the cluster using the documented procedure:
   KB: Isilon: How to reset and add repurposed Isilon nodes to an existing cluster
Isilon: How to Reset and Add Repurposed Isilon Nodes to an Existing Cluster

If assistance is required or if the node cannot be recovered or joined successfully, contact Dell Technical Support.