topic Re: Please help me understand controller/volume redundancy in MSA Storage

Please help me understand controller/volume redundancy

atomicluis — Tue, 17 Aug 2021 11:18:33 GMT

I am new to this.

I have 2 physical nodes in a Windows 2016 Cluster.

An dual-controller MSA 1060 with a "Data" volume which is presented to both Nodes.

My question is... If I disconnect the iSCSCI cables from 1 of the controller, is the volume supposed to be available to the Active Node?

I tried this and lost access to the volume.

Are both controllers active? or is one of them supposed to be in Standby and kick in when the active one fails?

Thanks for help!

Re: Please help me understand controller/volume redundancy

Ihaqueit — Mon, 16 Aug 2021 18:44:50 GMT

As per your query,

Dual controller means data passed thru these both controller.

If any malfunction in one controller then one controller act as main active controller and its change the state like write mode cache mean data pass thru one controller.

In your case how you setup the controller?

Re: Please help me understand controller/volume redundancy

Cali — Tue, 17 Aug 2021 07:24:54 GMT

Hi,

is it cabled in this way?

Then check this: HPE MSA 1060/2060/2062 Storage Arrays Best Practices

Here Section: CONNECTIVITY BEST PRACTICES

And Section: MPIO

Cali

Re: Please help me understand controller/volume redundancy

JonPaul — Wed, 18 Aug 2021 19:10:33 GMT

I think that the problem you experienced is the number of SESSIONs that your Windows system is creating to the storage target.
By default Windows will only choose a single path: Initiator IP -> Target IP
You should create multiple sessions from the host to the target using the Initiator IPs specifically called out to the specific target IPs.
In other protocols the additional PATHs are automatic but in iSCSI they need to explicitly set.
You should see multiple paths to the LUN in the device manager or the MPIO manager.