VAAI status “Unknown” on EMC array using > 2TB LUNs (Update)

Today I noticed that some new created LUNs on an EMC Clariion CX4 array stuck on the VAAI status “Unknown” below the storage configuration on vSphere 5. This was quite strange, because I was creating and adding multiple LUNs and only the 2TB LUNs (on SATA drives) seems to have this problem. All other LUNs directly prompted “Supported” after adding to the ESXi hosts.

I started some tasks which use the VAAI feature (cloning, svMotion), because I remembered that I have read VAAI sometimes first get detected after a task tried to use it. After several tasks and some rescans the VAAI status didn´t change. Also I verified the VAAI status on the command line (command: esxcli storage core device vaai status get), which showed the following on the 2TB LUNs:

ATS Status: unsupported
Clone Status: unsupported
Zero Status: supported

All other LUNs showed supported on all three features. You can also control the usage of the VAAI features via esxtop. Rickard wrote a nice article about this, which you can find here.

I searched the KB and community and came across this interesting thread, where people exactly reported the same issue. The result seems to be a FLARE code upgrade, which EMC needed sometime to release for the Clariion arrays (VNX got fixed earlier). So I am going to upgrade our both arrays next week and I am pretty sure this will also help in my case. I didn´t notice this problem/bug until now, because I never created >2TB LUNs until now with vSphere 5.

After updating the FLARE code I will give a short update on this post….

Update – 27.04.2012

We just finished updating our first CX4-120 from FLARE to and this solved the problem described before.

Speak Your Mind