Storage Issues when rebooting a S2D node after may patches

Share this post:

Lately we have seen alot of Event ID 5120 with a status code of STATUS_IO_TIMEOUT or STATUS_CONNECTION_DISCONNECTED during rebooting a node.

Here is a statement from Microsoft about the issue and what to do when rebooting a node.

In the May cumulative update we introduced SMB Resilient Handles for the S2D intra-cluster network to improve resiliency to transient network failures.  This has had some side effects in increased timeouts when a node is rebooted, which can effect a system under stress.  Symptoms include event ID 5120’s with a status code of STATUS_IO_TIMEOUT or STATUS_CONNECTION_DISCONNECTED when a node is rebooted.

Until a fix is available, a workaround that addresses the issue is to invoke Storage Maintenance Node prior to rebooting a node in a Storage Spaces Direct cluster.  Let’s say when patching for example.

So, first drain the node, then invoke Storage Maintenance Mode, then reboot.

Here’s the syntax:

Get-StorageFaultDomain -type StorageScaleUnit | Where-Object {$_.FriendlyName -eq “<NodeName>”} | Enable-StorageMaintenanceMode

Once the node is back online disable Storage Maintenance Mode with this syntax :

Get-StorageFaultDomain -type StorageScaleUnit | Where-Object {$_.FriendlyName -eq “<NodeName>”} | Disable-StorageMaintenanceMode

Please note that Cluster Aware Updating does not put your nodes in Storage Maintenance mode.

 

reference: https://jtpedersen.com/2018/08/storage-issues-when-rebooting-a-s2d-node-after-may-patches/

Subscribe to our newsletter

Get the inside scoop! Sign up for our newsletter to stay in the know with all the latest news and updates.

Don’t forget to share this post!

2 thoughts on “Storage Issues when rebooting a S2D node after may patches”

    1. It depends on if you have any issues with any thing. It will not allow you to put a node in to maintenance if there is a warning in the cluster. Let’s say a disk that is unhealthy, a virtualdisk that is unhealthy or something. So make sure that there is no unhealthy systems in your setup.

      Regards
      Jan-Tore Pedersen

Leave a Comment

Scroll to Top