HPE Storage Users Group

@Cleanur, Richard:

I was not aware 2.3.1 allowed node reboots. This is an eyeopener for me. So on that one, agreed. I was not taking the older software versions into account. Apologies for that.

On the dual node failure in version 3. Just read your own explanation again, a situation where you have to thank heavens above for the fact your 3PAR didn't crash on a dual node failure is simple not an acceptable solution. In other words: The 3PAR is not build for dual node failure, so just don't let it happen, it will go wrong.

On the rebooting part. I've witnessed an occasion where a node in the other node pair was rebooted, during the outage of node in the first node pair, and the V-class 3PAR panicked. Luckily it was a POC setup so no impact, but for the technical part it does not matter.

I don't know why HP is a bit difficult about being clear this. It is not a defect, nor a shame. Other vendor solutions (EMC, etc) are neither build to survive a double component failure.

_________________
The goal is to achieve the best results by following the clients wishes. If they want to have a house build upside down standing on its chimney, it's up to you to figure out how do it, while still making it usable.

They do seem to keep it on the down low. Back when I first discovered this "mechanic" I felt a little disappointed, perhaps misguided by the promise of "Active Mesh Architecture". However, I now understand the distinction that "Active Mesh" applies to the front end LUN ownership, not the back end PDs.

_________________
Richard Siemers
The views and opinions expressed are my own and do not necessarily reflect those of my employer.

I am getting ready to put a 7400 into service shortly. We will include a 2 controller failure test when we do our fault tollerance testing. Making sure to remove one controller from the 3,2 group and one from the 0,1 group.