Skip to main content

Example of responding to degraded system health

Contributors netapp-aherbin
PDFs

By reviewing a specific example of degraded system health caused by a shelf that lacks two paths to a node, you can see what the CLI displays when you respond to an alert.

After starting ONTAP, you check the system health and you discover that the status is degraded:

      cluster1::>system health status show
        Status
        ---------------
        degraded

You show alerts to find out where the problem is, and see that shelf 2 does not have two paths to node1:

      cluster1::>system health alert show
               Node: node1
           Resource: Shelf ID 2
           Severity: Major
	   Indication Time: Mon Nov 10 16:48:12 2013
     Probable Cause: Disk shelf 2 does not have two paths to controller
                     node1.
    Possible Effect: Access to disk shelf 2 via controller node1 will be
                     lost with a single hardware component failure (e.g.
                     cable, HBA, or IOM failure).
 Corrective Actions: 1. Halt controller node1 and all controllers attached to disk shelf 2.
                     2. Connect disk shelf 2 to controller node1 via two paths following the rules in the Universal SAS and ACP Cabling Guide.
                     3. Reboot the halted controllers.
                     4. Contact support personnel if the alert persists.

You display details about the alert to get more information, including the alert ID:

      cluster1::>system health alert show -monitor node-connect -alert-id DualPathToDiskShelf_Alert -instance
                  Node: node1
               Monitor: node-connect
              Alert ID: DualPathToDiskShelf_Alert
     Alerting Resource: 50:05:0c:c1:02:00:0f:02
             Subsystem: SAS-connect
       Indication Time: Mon Mar 21 10:26:38 2011
    Perceived Severity: Major
        Probable Cause: Connection_establishment_error
           Description: Disk shelf 2 does not have two paths to controller node1.
    Corrective Actions: 1. Halt controller node1 and all controllers attached to disk shelf 2.
                        2. Connect disk shelf 2 to controller node1 via two paths following the rules in the Universal SAS and ACP Cabling Guide.
                        3. Reboot the halted controllers.
                        4. Contact support personnel if the alert persists.
       Possible Effect: Access to disk shelf 2 via controller node1 will be lost with a single
 hardware component failure (e.g. cable, HBA, or IOM failure).
           Acknowledge: false
              Suppress: false
                Policy: DualPathToDiskShelf_Policy
          Acknowledger: -
            Suppressor: -
Additional Information: Shelf uuid: 50:05:0c:c1:02:00:0f:02
                        Shelf id: 2
                        Shelf Name: 4d.shelf2
                        Number of Paths: 1
                        Number of Disks: 6
                        Adapter connected to IOMA:
                        Adapter connected to IOMB: 4d
Alerting Resource Name: Shelf ID 2

You acknowledge the alert to indicate that you are working on it.

      cluster1::>system health alert modify -node node1 -alert-id DualPathToDiskShelf_Alert -acknowledge true

You fix the cabling between shelf 2 and node1, and then reboot the system. Then you check system health again, and see that the status is OK:

      cluster1::>system health status show
        Status
        ---------------
        OK