Greetings,

We have a DSSv6 with 16 1TB SATA disks set up with RAID 6, we have a single LUN connected to a 3 server 2008 R2 cluster.
Our SAN has 4 1Gbps network cards, 1 for management and 1 connected to each host via a cross-over cable.
The SAN is used to store VHD files for our Hyper-V virtual servers, we have 9 servers on the SAN (3 on each host).

Recently after being up for about 40 days the SAN performance went from good to terrible, checking the statistics page showed me that the load had increased from an average of between 2-4 up to 4-8.

I downloaded the logs and opened up the tests file, hdparm was reporting speeds of around 10MBps on the RAID.

I shut down all of the virtual machine as soon as I had a maintainance window and restated the SAN, hdparm was then reporting around 400MBps on the RAID.
I booted the virtual servers and performance was good again.

This lasted for 7 days until the issue re-occured, I restated the SAN again last night and performance is now fine again.

I've raised a support ticket but haven't heard much back so far.

Anyone else had this or know what the problem might be?
Restarting the SAN once a week or even once a month really isn't a viable solution for me.

Many thanks,

Jon