Around the Storage Block Blog

[#1]

By Jim Haberkorn

Question of the day: What would happen if Star Fleet used NetApp MetroCluster instead of the Kobayashi Maru as their no-win-scenario war game software?   

Scottie:  “Captain, I’ve just discovered that our primary site has six single-points-of-failure. What do we do?”

Captain:  “Failover to the secondary site immediately, you fool.”

Scottie:  “But, Captain, the secondary site also has 6 SPOF.”

Captain:  “Ha Ha, Scottie, I love a man who can joke in the face of disaster.”

Scottie:  “Captain, I’m not joking. When I bought the product I asked every question in the galaxy, but I never thought to ask about single-points-of-failure. Captain, it’s the year 2010. Who in their right mind sells HA products with single-points-of-failure?”

While you are pondering the IT implications of this scene, here are a few things to consider:

1. NetApp markets its MetroCluster solution as able to “Achieve continuous data availability for mission-critical applications at half the cost and complexity.”
2. What MetroCluster does is break up the no-single-point-of-failure NetApp system on your primary site and turn it into 2 separate systems, each with multiple single-points-of-failure, and place each system at a different location up to 100km apart. This is what passes for high-end, mission-critical HA with NetApp.
3. Your primary site now has a single NetApp filer that is orders of magnitude more likely to fail than the original system. And now, when it does fail, it will failover to a single NetApp filer which is also orders of magnitude more likely to fail. I am not making this up.
4. In other words, the NetApp idea of HA is to make your primary site more risky and then have you failover to a system that you wouldn’t trust your mother’s cookie recipes on let alone your company’s mission-critical data.

Okay, NetApp enthusiasts, your turn.  

[#2]

Hi Leon,

Thank you for your posting. I hope you are happy with NetApp MetroCluster and that it gives your company many years of reliable service.

In regards, MetroCluster, my main criticism is not that it doesn’t work, but that it is mis-marketed when NetApp claims it is for mission-critical, zero-downtime-environments. I won’t repeat all the points I’ve made previously, but here are a few:

1. At its core, NetApp MetroCluster is a system with multiple SPOFs on the primary site failing over to another system with multiple SPOFs on the secondary site
2. With the added detriment of allowing events outside the datacenter to interfere with the HA of a system inside the datacenter. For example, if the link between your sites goes down then the filer in your primary datacenter has no HA. NetApp MetroCluster effectively neutralizes your company's multi-$million datecenter investment. 
3. Also, because of the single head at each end and the built-in risk of split-brain, the failover is not totally automatic, though I have read NetApp claims that it is. Further, the failback is highly manual and complex and comes with the vendor’s warning that a misstep could cause data corruption. To me, none of this adds up to ‘mission-critical ZDT'.  

But from NetApp’s point of view MC does have two advantages:

1. If NetApp finds itself in a price war vs. a competitor then MC’s single-node architecture allows NetApp to really drop its price and still make a huge profit.    
2. If it wasn’t for MC, NetApp wouldn’t be able to offer application failover. NetApp can’t do application failover with a 2-node system at each site – that’s why they have to sell MC in mission-critical environments even with all its problems.  

 Personally, if I were a customer, I would be suspicious of a company that didn’t have a proper 2-node solution to sell in this space. I would be worried that they were pushing their single-node solution only because they didn’t have anything better to sell me.

One last point: NetApp has a 55-page white paper on MetroCluster that offers an incredible amount of detail about MetroCluster, but neglects to mention the obvious HA holes in the product. Of all the details to leave out in a mission-critical, ZDT product, it leaves out the most important ones – how strange? But I would still advise customers to read the white paper; it does have some good information, for example, the failback steps and the warning about data corruption.

Best regards,

Jim

[#3]

Very interesting battle back and forth on this.  I tend to agree with Jim, NetApp is less than forthcoming about how this architecture works in the real world, that it is definitely two individual mirrored systems - with a logical connection between.

BTW- to the gentleman who said Scott Lowe lacks credibility - Scott is very well respected in the storage community and has shown time and time again that he knows his stuff.  Attacking his credibility (especially since he has nothing to do with this article) goes a long way towards discrediting your own credibility.

Thank you to everyone for a fantastic read.

[#4]

Hey Bill - I just saw Scott in Vienna and totally agree with you that Scott is very well respected.  Funny that the person who took a swipe at him is a NetApp reseller.  But I also think it's very difficult for anyone paid by a vendor to be unbiased.  Whether we work for a vendor or not, we all have biases but who your paycheck comes from has a big effect.