Storage

EMC AX4 - A failover update

EMC's AX4 failover features work as advertised, much to the delight of Scott Lowe. Sometimes, it's nice to celebrate the small successes!

For a few months now, I've written blog postings regarding my experience with EMC's AX4 entry-level iSCSI SAN.  Westminster College purchased two AX4 arrays - a SAS unit with dual redundant controllers and an attached SATA unit - for a total of 13.8TB raw capacity.  Since installation, we have slowly begun moving applications and services to the SAN and away from direct-attached storage.  We now run our primary file server, Exchange 2007 mailbox stores, and, as of last week, three administrative database applications from the SAN as well as a number of VMware virtual machines.

Last week, while we were moving our administrative databases to the SAN, we experienced a problem with one of the controllers.  Although it was still passing traffic, it was showing up in the array management software in an "unmanaged" state.  Since we were actively performing configuration tasks for our database move, I wanted to make sure that this issue got resolved, so I contacted EMC support.

During the support session, the analyst asked if he could reboot the controller.  I asked him to tell me what the impact of this operation would be and he indicated that any connections to this controller would be severed.  I knew for a fact that some of our iSCSI targets were using the controller in question.  However, everything we have connected to the SAN is configured for high availability.  In theory, a connection that is severed for any reason will reestablish itself through the second controller.  As we were at a safe point in the day, I told the customer support rep to have at it.

Good news on two fronts:

  • The reboot corrected the problem.
  • All of the connected servers failed over to the second controller with no noticeable interruption in service.  In fact, I was remoted in to one of the connected servers when the first controller was rebooted.  As soon as the first controller went down, EMC's PowerPath software popped open and told me that redundancy to the controller had been lost and that traffic was moving through controller B now.  A beautiful thing, indeed!

Now, some of you will say, "Why so happy?  Isn't that what you expected?"  Well, yes.  But, sometimes, it's just nice to see something work the way that it's supposed to without any problems!  We spent a lot of time planning the architecture to get to this point and this was a great, safe way to test our work in real life.

About

Since 1994, Scott Lowe has been providing technology solutions to a variety of organizations. After spending 10 years in multiple CIO roles, Scott is now an independent consultant, blogger, author, owner of The 1610 Group, and a Senior IT Executive w...

16 comments
jonathan_hau
jonathan_hau

Just wait until EMC tells you that you need to upgrade the Powerpath software to support their umpteenth flare code bug fix. By the way you have to do this yourself as EMC will not do it for you...

mike_patburgess
mike_patburgess

Well, is it good, bad or mediocre? Have you ever looked at HDS and their portfolio?

cgehring
cgehring

I am surprised it worked well. I have not had good experience with EMC and failover... clumbsy at best. EMC technology is old school. You should look at the new players in the market.. Compellent, 3PAR, Pillar, LeftHand, etc. These guys understand storage, and how to make it work seamlessly and effectively. Good luck with that EMC next time you need an upgrade, or you outgrow the little guy. It will eventually look like a chopped up mess, full of hot spots.

paul_austine
paul_austine

sir i have just recieve note of lotus which i won from palm sterling and i do not know how to interpret it ,what do i do? augustine okorie junior paul_austine@yahoo.com cnn-nigeria

tonyfitz7
tonyfitz7

Strange how a big fat green NetApp Banner gets installed over the EMC article header....and you can't get rid of it...Hmmmm..TechRepublic needs to account for this...???????

Scott Lowe
Scott Lowe

I did consider some of the vendors you listed in addition to others. With serious and signficant budget constraints, I chose what I felt was the best solution for the available budget dollars. I'm under no delusion that the AX4 is a permanent solution. As we grow, we'll add expansion chassis until we hit the solutions 60TB max. Quite honestly, that alone will take us a very, very long time. After we outgrow this solution, I'm aware that we'll be in "rip and replace" mode, but with the pace of change in this market, I have no doubt that we'd be doing that regardless of selected vendor. I will admit that managing the AX4 is not as straightforward as managing a LeftHand or EqualLogic solution. That said, once you get the hang of it, it's not bad. Scott

anne.powel
anne.powel

Pop-up ad folks are getting more and more clever (and obnoxious!)

jadewith
jadewith

I am amid this very decision. I believe I have narrowed my choices down to either a Lefthand 2060 or an EMC Ax4. Both are configured with small arrays 2-3 TB and cost right around $20k. The majority of the negative comments out there that I have seen has been towards the EMC. Is that mostly because they are more prevalent in the marketplace, and so have had more chance to get under more people's skin? Or is Lefthand's solution really that much better?

cgehring
cgehring

...I inherited them. I recently took this position and inherited a system that has been in place for about 3 years. We will be putting together an RFP to replace our SAN in the coming weeks. EMC is most likely not going to make it to the final cut. I did an RFP 3.5 years ago, and compared several vendors, and the final tally came down to EMC, NetApp, 3PAR, and Compellant. 3PAR and Compellant blew the others away when it came to technology built in. 3PAR won mainly because they were more established at the time. I love to talk storage, and I can show you real documentation from our current EMC system and my past 3PAR to prove any points you might be interested in. It is not hard to prove that new technology is better than old technology. EMC has not had an inovation in their arrays in many years. If you are truly in the market for a SAN, I suggest you get a VAR that knows the market to show you the light. EMC is legacy technology, and they can't bury it because HUGE companies rely on them to remain consistent. Their model of dedicating disks to servers is wasteful and pins you into a corner as soon as you need to move. Other legacy systems have the same problems. If you must buy a big name, look at HP. They are the closest big name vendor to true virtualized storage.. though they are simply layering magic over their legacy. I hear it works though. FYI - Before you ask... I have no ties with any VARs or storage companies... I was simply spoiled in the past, and now have to take a huge step back. This will cause anyone pain, and make them want to spout off about it. For current EMC customers that love their systems, that's cool, but I would simply argue that you don't know what you are missing. I could go on... if anyone cares :-)

tonyfitz7
tonyfitz7

Cool....assuming you would agree that PCs don't generally "create" spurious banners, I'd love to know why it's now gone...?? TechRepublic...any idea?

Scott Lowe
Scott Lowe

It is true that EMC has legacy baggage even when it comes to releasing new solutions such as the AX4. It's likely that they want the overall management experience to be similar from platform to platform so that skill sets don't need to be adjusted when an organization grows to a new platform. Lefthand definitely has the newcomer advantage here with a cleaner, easier management experience for you as well as a better feature set for future use. For example, the iSCSI AX4 doesn't include array replication, making it unsuitable for disaster recovery scenarios. I've been told that this feature is to be added at some point, but not sure when that will come. Keep growth in mind, too. With the Lefthand solution, as you add more NSMs, you get more data transfer capacity. The same is not true of the AX4. As I add space to my array, I will not get a corresponding increase in iSCSI ports. On the flip side, adding more space to an existing AX4 array is much less expensive than adding the same space to a Lefthand solution. If you're going to stay relatively small, this AX4 limitation shouldn't worry you too much. I can add another 9TB of SAS space to my AX4 for around $10K. Quite frankly, I was amazed that I ended up with an EMC solution; I never thought I'd see anything from them that my budget could support. They are a force in the storage market for a good reason - they do make good products and provide good support. Lefthand does as well. Scott

cgehring
cgehring

It all depends on your requirements. Just like the EMC might be the best choice for Scott's problem, the 3PAR was good for my last company, and something else entirely might work for the problems I am trying to solve. I recommend that for any purchase you plan to make, you force vendors into an RFP process. Many times, we are so overwhelmed with our daily jobs, we simply take the option in front of us instead of really digging to see what might be the best fit. Crappy answer right? Best advice I can give you is this... Don't even think about talking to a vendor or looking for a solution until you have listed and ranked your requirements. Give the requirements a weight, and come up with a system for scoring, then vendor by vendor see how they meet your requirements. If you are looking at small, simple storage with a single purpose, just about any solution will do. Regardless of what any SAN vendor tells you, SANs are simply building blocks made of the same core component... disks. No matter how much cache they throw on a system or how fast they say their fiber channel ports are, the disks are the bottleneck, and they are all the same, regardless of vendor. They only start to separate when you talk about scale (going really big, or really fast) or application integration (VMware, Oracle, SQL, replication, snapshots, de-duplication, backups, etc.) When you are talking about 2-3TB of storage, you are only talking about 7-14 disks depending on vendor and disk type. At that small level no vendor is going to be able to do anything spectacular. A million dollar system will be no faster than putting the disks in a direct attached configuration.... the limited number of disks is means performance will be pretty much guaranteed equal across all vendors. The questions you need to ask are: - What will I be attaching? VMware? Oracle? SQL? File Sharing? NAS? - What other things do I want to do with the storage? Replication? SnapShots? SRM? From there, you can start to look at where the vendors excel. At that level, since performance is not a question (can't be really) then your only concern is "How cool is their software, and what can it do to help me solve my problem?" Storage is fun :-)

The 'G-Man.'
The 'G-Man.'

Better start checking your systems. Having things like that popping up means your security is not as good as it should be.

gcattley
gcattley

It flashed up for around 5 seconds, and dissapeared...