VCAP5-DCA Exam prep

I took the VCAP5-DCA exam some 7 months ago, so it’s not quite so fresh in my mind as the VCAP5-DCD exam. If I’d planned things better, I’d have moved straight on to it after passing the VCAP4-DCA exam (which I passed the year before), as there was a good 70-80% overlap of content, and I’d have saved myself a significant amount of study time!

As with the DCD exam, there is no specific course requirement for the DCA exam, but you must have passed one of the VCP5 certifications (any will do, it doesn’t have to be the VCP5-DCV).

I won’t beat around the bush, the DCA exam is a real trial:- 3.5 hours long, 26 scenarios with multiple configuration and/or troubleshooting tasks on each. And believe me, 3.5 hours is not enough time!

You will get credit for what you do manage to achieve, even if you haven’t fully completed a task, so it’s worth having a go, to get partial marks, even if you’re not able to finish something.

The exam uses a real lab environment with actual hosts and VMs, not some kind of Shockwave Flash emulation. This is a good thing and a bad thing. It does behave exactly how you expect, but if you screw up the configuration (say you deleted the networking completely) then you will lose access to it, just as you would in the real world.

There are 2 main things you need to prepare for this exam

  1. The Blueprint
  2. Your own lab environment

In addition to the blueprint, study guides can help, these two were the ones I found most helpful

Plenty more are available, feel free to have a browse of them, and make your own mind up. I spent most time with the Valco Labs one, printed it out and used it as the basis for my study.

The actual study…

For me this was firstly about spending time understanding all the areas in the blueprint listed as “Knowledge”. Once those areas had started to sink in, I moved on to the actual lab practice. You need to work through all the “Skills and Abilities” entries and make sure you know how to do them, and how to do them very, very quickly.

In my view, the key to this exam is how quickly you can work. I’ve seen a number of guides on this exam suggest that you should practice the 3 ways of doing something (GUI, PowerCLI, Command-Line) until you can do them all without referring to any reference guides – I think that’s probably over the top and would take an excessive amount of time to achieve.

More practically, there are some things that can *only* be done via PowerCLI or Command-line, and you do need to practice those as in an exam, you don’t have time to be looking up syntax. For example manipulating the Path Claim rules, or configuring Auto-Deploy. Other things where there is a GUI method, I personally found it to be quicker and easier to use the GUI method, however you do still need to practice this until you know instinctively where you’re going in the GUI and can perform the task rapidly.

Ultimately, studying for this exam is about practice, practice, and more practice. I spent several hours a week for 3 months, rising to a couple of hours a day in the last 2 weeks before the exam, and pretty much all day on the last few days.

On to the exam…

Using the GUI is significantly slower than you will be used to. The labs are in the US, and you are using them via a much smaller screen size than you are likely to be used to.

VMware provide a simulation of the lab environment, which is well worth checking out, it gives you a good idea of what to expect.

Another hint I picked up was to set up advanced search on the reference materials they provide – this makes it much quicker to find something when you’re not sure which document it’s in. However you cannot afford to spend time searching for how to do things, other than the odd occasion, or you will run out of time.

As mentioned earlier in this post, the 26 “scenarios” will cover configuration tasks, and troubleshooting tasks. You may also get asked to create or amend a PowerCLI script (I don’t class that as config or troubleshooting).

Sometimes the scenarios may build on a previous one – on my VCAP4-DCA this allowed me to realise a mistake I’d made on the first one and go back and correct it.

Make sure you’re adequately hydrated before you go in, but not to the extent that you may need a toilet break – you can’t afford the time. Eat something with slow energy release before the exam, and maybe take ibuprofen if you’re likely to get a headache.

I think the key to this exam is to put in the hours practicing, especially on parts of the product that you don’t normally use – for me this is quite a large proportion as my main client only has Enterprise licencing rather than Enterprise+, and secondly about being able to work fast.

Unlike most of the VMware exams, you don’t get your results straight away, you have to wait 15 working days (my VCAP5-DCA results actually took over a month to arrive!), and you get an email with your score.

I didn’t finish either the VCAP4-DCA or the VCAP5-DCA, but scored enough marks to pass both (333 on VCAP4, and 381 on VCAP5).

VCAP5-DCD Exam prep

A number of people have shown an interest in how I prepared for the VCAP5-DCD exam (there’s also been interest in how I prepared for the DCA exam, so I may do a post on that at a later date).

It is a difficult exam to revise for, the biggest problem being that there is nowhere to practice some of the new question formats which are introduced in the DCD exam. I’m talking about the Drag-and-Drop and the Design Tool (also referred to as Visio-style) questions.

I was fortunate to find this blog post about the exam format, a few days before the exam, which helped. It’s very, very helpful, and I say a huge thank you to the author for sharing that experience.

My actual preparation started with the blueprint, and then I searched for study guides. I found the following ones the most helpful:

Virten.Net – Florian Grehl
Virtually Hyper
and the vBrownBag APAC sessions, collated by Nick Marshall, which I downloaded to my phone and listened to as podcasts while commuting. You don’t lose a lot by not seeing the slides.

Books were also a key part of my revision – the study guides don’t go into as much detail, but the books can be a bit overwhelming. I think it took me several *months* to get through the Clustering Deep Dive!

I was fortunate to be given a copy of VCAP5-DCD Official Cert Guide book – Paul McSharry when attending our company conference in Atlanta a few weeks back. This was a straightforward read, helpful on the business requirements / design objectives, for someone like myself – coming from an Engineer point of view rather than Architecture.

While I don’t own a copy of VMware vSphere Design 2nd Edition – Forbes Guthrie I was able to have a thumb through one and think I might be adding it to my (virtual) bookshelf for my work towards VCDX. It looked to be very useful for the VCAP5-DCD exam anyway.

The toughest of the books I read was Vmware vSphere 5.1 Clustering Deep Dive – Epping/Denneman. As I mentioned above, this took me a long time to read through, and was very informative on DRS and HA. The best thing about having the ebook version was that I could pull out all the “Basic Design Principles” mentioned throughout the book, and have 4 pages of A4 that had key design concepts to remember.

Coming from an engineering perspective, one of the hardest things to get my head around was the business requirements, design methodology and architectural aspects of the exam. I spent a long time getting to grips with Requirements (Functional and Non-Functional), Constraints, Assumptions and Risks. This is really fundamental to passing this exam if you come from the same background as me. I can only recommend that you do as much reading around these terms as you can. There is a certain amount of overlap between them, which can depend on perspective, and the type of design you’re doing. I only started thinking about booking the exam when I felt confident about placing something into one of these categories without second guessing myself.

One other thing to make sure you do, is to watch the demo of the design tool. I would recommend watching it several times, so you’re more comfortable with it.

And so on to the exam itself…

I had 6 design tool question, I think something like 10 drag-and-drop questions, and the rest of the 100 were the single/multiple choice.

The first 2 questions were design tool ones, which was a real tough opening to the exam, but when I got stuck into the first one, I started to feel much more confident. In fact, I didn’t really have too much trouble with the design questions (although as you don’t see your detailed marking, I can’t guarantee how well I answered them!), but it was some of the drag-and-drop ones I found hardest.
Some were straightforward – 4 choices to drop on 4 options, but some had 7-8 choices, 7-8 options, and you could use the choices multiple times, or not at all. These were really difficult as you really did find yourself second-guessing whether an option could go in 2, 3 or maybe 4 places, then taking it out again, and so on.

This comes to another tip – don’t get bogged down. At some point you have to go with your best guess.

I had a PC crash just as I’d completed one of the design tool questions – I was wiggling an object to see if the links were connected ok, and the screen froze. I got the “A script on this page may be busy or may have stopped working” error, and eventually an application error. You can read more on my previous post, but I got up and running again on another PC, albeit with that question reset back to blank. Re-doing the question took a fraction of the time it took originally though.

The 15-20 mins per design tool question, which a number of people have suggested as a rule of thumb, seems to be a good idea. There were a couple that took nearly 20 mins, but some were much shorter – one I think was only about 5 minutes.

Another exam tip I would reiterate is to read the question before the supporting material. It really helps you to narrow down your reading of what is often excess material. There were a number of questions where you really did not need to read the supporting material whatsoever!

You can once again review the questions at the end of the exam. This was removed for a while due to problems with some answers being lost, but I was able to go back and check that my first few questions (done on the 1st PC) were still there.

Then all you have to do is hit “End Exam” and after a few seconds you get your results. All I saw was “Congratulations…” and I was out of the room!

Finally I’d like to thank all those who have contributed to the community, especially those I’ve linked to above. Without their help, this exam would have been a lot tougher to revise for. And good luck to anyone reading this, I hope I’ve helped you.








VCAP5-DCD exam

I travelled up to Leeds today for my first attempt on the VCAP5-DCD.

My last visit to Global Knowledge was for the VCAP5-DCA, and I’d intended to go up on the train, only to get to my local station to hear that the train I wanted had been cancelled. The next one would have arrived too late, so I had to high-tail it back home and jump in the car.

This time, I turned on the TV for the local news to announce that the M1 Northbound was closed near Barnsley, and would be shut for at least the next hour. I’d intended to get the train anyway, but this meant no fallback situation of driving, so I dashed to the station to get the half-hour earlier train, to give me a bit more scope for problems!

Being this early meant of course that there were no problems so I was at GK an hour before the exam start time! I had time for a little more revision, and a drink of water, plus my pre-exam banana (one of the hints I picked up from an exam blog).

Checking in to the exam took longer than usual – not sure why – but soon I was at the PC and going through the pre-exam survey.

Then on with the exam! I obviously can’t say much about it due to NDA, but my first 2 questions were the visio-style ones. Talk of a tough start! After a brief panic, I read the first one properly, and it seemed achievable, and so I continued.

About half-way through, I’d just about completed #3 or #4 of the visio-style questions and was giving all the objects a little wiggle to make sure things were connected properly when the PC froze up. After 30 secs or so it displayed the “A script on this page may be busy or may have stopped working” message, I clicked wait and then the screen updated again. Phew I thought, then it froze again. This time I also got “An application error has occurred”.


I put my hand up. No exam administrator running to my rescue.

I waved my arms around still nothing.

Eventually I knocked on the wall (I could hear them in the next room) and one of them came through. After about 10 mins I was up and running on another PC (apparently the camera system had frozen so it looked like we were all working away fine), but it had lost all the work on that question, so I had to repeat it.

I requested they contact VMware for extra time to make up for the lost time, and they did so (in the end I decided I didn’t need it, and would have had to stop so they could add the time. Was good to know that the option was there if I was running out of time though).

Anyway, the rest of the exam went without a hitch, although the multiple choice and drag&drop questions seemed to get harder, and I reached question 100 with 1h 15m remaining on the clock. I’m not one to go second guessing my self, so I just used the review page to make sure the first couple of visio-questions were intact, then hit the End Exam button.

I was delighted to see “Congratulations” on the page – I’ve no idea what else it said, as I was straight out of the room!


So next step is the VCDX, but before that I have a stag night, a wedding, a honeymoon, a family holiday. Probably won’t even remember what a VM is after all that!

Oh and I’ll try and get another post out soon about which resources I found most useful in preparation for the exam.

The case of the disappearing datastore

I was asked this week to investigate an issue where out of multiple hosts and multiple datastores, there was one host which couldn’t access a single datastore.

In the past, I’ve seen issues where a datastore was only visible to a single host in the cluster, or a host had lost access to all datastores, but never a single datastore unavailable on a single host.

Talking someone else through diagnosing something obscure like this is never easy, but getting the end of the vmkernel.log was enlightening:

2014-06-17T11:03:05.457Z cpu31:64368)WARNING: HBX: 1889: Failed to initialize VMFS3 distributed locking on volume 529dc221-e68ab3c4-3e52-b499baa3e4c6: Not supported
2014-06-17T11:03:05.457Z cpu31:64368)FSS: 890: Failed to get object f530 28 1 529dc221 e68ab3c4 99b43e52 c6e4a3ba 0 0 0 0 0 0 0 :Not supported
2014-06-17T11:03:05.457Z cpu31:64368)WARNING: Fil3: 2034: Failed to reserve volume f530 28 1 529dc221 e68ab3c4 99b43e52 c6e4a3ba 0 0 0 0 0 0 0
2014-06-17T11:03:05.457Z cpu31:64368)FSS: 890: Failed to get object f530 28 2 529dc221 e68ab3c4 99b43e52 c6e4a3ba 4 1 0 0 0 0 0 :Not supported
2014-06-17T11:03:05.583Z cpu36:64370)HBX: 676: Setting pulse [HB state abcdef02 offset 3710976 gen 35 stampUS 85863025463 uuid 539ed180-c09ff563-6044-e4115b10555a jrnl <FB 0> drv 14.54] on vol ‘Datastorename’ failed: Not supported
2014-06-17T11:03:05.583Z cpu36:64370)WARNING: FSAts: 1263: Denying reservation access on an ATS-only vol ‘Datastorename’

A bit of a google found this

Basically, what appears to have happened is that when the datastore was created, Hardware Assisted Locking was available on the storage array. Because it was available, the VMFS filesystem was created with the flag set to use it (ATS-Only)

At some point since, it was no longer supported, and it would seem that this host had lost access and attempted to reconnect (maybe a reboot) to the datastore and failed, because ATS-Only was set, and the array no longer supported that as a locking mechanism.

A few days later, and before the fix was implemented, a power outage took out half the hosts in the farm, when these came back, none of them could access the datastore (unsurprisingly).

Implementing the fix in that link (vmkfstools –configATSOnly 0 /vmfs/devices/disks/device-ID:Partition) removed the setting and restored access to the datastore for all hosts.

SMBv1 and Hide NAT

I was asked yesterday to investigate an issue where client PCs were getting disconnect messages from a share on a virtualised server.
It was thought that there was some problem with the virtual switch, and/or the VirtualConnect config of the blade enclosure.

A network capture was included, as was a visio diagram of the network configuration, which was very nice to have! It’s great to deal with experienced techies who know you’re going to need some background material, and think to include it with the request for help 🙂

What struck me as odd when looking at the capture was that the offending RST packets always appeared to coincide with a new TCP session being set up to port 445. ie New TCP session always followed by a RST of an old one, two new TCP sessions -> 2 RST of old connections.

A bit of digging pulled up this link, something I’d never come across before. Basically, SMBv1 does not support Hide NAT. It regards a connection between 2 IPs as a single client and a single server, and closes down any old connections between those two IPs whenever a new connection is made.

The main workaround is described in the link, namely disabling communications over port 445 for Windows XP. This forces it to use NetBIOS over TCP/IP. Woo-hoo :/

Windows Vista/7/8 talking to Windows 2008 Server and above can use SMBv2 which was written to cope with Hide NAT.

Testing so far has shown this workaround for Windows XP to be functional.

Anyway, chalk another issue down as “nowt to do with VMware” 😉

PowerCLI adjust ResPool share values

One thing that Frank Denneman (@frankdenneman) regularly bangs on about is not using Resource Pools as folders. And with good reason – setting up resource pools without understanding the effect on performance can cause significant issues.

We we first started working with RPs in ESX 3.x, I was blissfully unaware of this, and we have quite a lot of RPs these days – we use them to group application instances for example.

This isn’t an issue generally, as there just isn’t the resource contention to cause share based rationing to be invoked. Having said that, if we did hit a level of contention, the default number of shares per RP probably wouldn’t give us the performance we were after.

Ideally, we need some way of setting the RP shares to a more sensible value, based on the number of vCPUs and amount of vRAM allocated to the VMs within. Enter PowerCLI, stage left…

$rps = get-cluster "Cluster Name" | get-resourcepool | where-object  {$_.Name -ne "Resources"}
foreach($rp in $rps) {
 $cpucount = 0
 $memcount = 0
 $vms = $rp | get-vm
 foreach ($vm in $vms) {
  $cpucount += $vm.numcpu * 1000
  $memcount += $vm.memorygb * 10240
 set-resourcepool $rp -numcpushares $cpucount -nummemshares $memcount

This will list all the RPs in a given cluster, count the vCPUs and vMEM assigned to the VMs in each, and set the CPU and Mem shares based on those numbers.

CSC TechCom 2014 over – ctd.

The Tuesday and Wednesday were long days. A general session about the current business strategy started things off, then we were into the breakout sessions. Again, I can’t really go into details, but they varied in relevance to me. Some of them were very good though and really interesting to hear the gospel from those who are developing the strategies and direction.

Vendor sessions followed in the late afternoon, and over the 2 days I attended ones from Citrix, Microsoft and VMware. All were well delivered and interesting material.

The evenings I started off at the VMware stand in the Ignite zone (exhibitors stands). Food and drink was provided, some strange combinations were on offer, due to a loss in translation, the most notable example being fish and (potato) chips, rather than proper chips (what we call fries in the UK).

I was very lucky to be able to spend much of this time chatting to VCDX001 (John Arrasjid) about VMware certifications and primarily the VCDX process. He is a very knowledgeable chap and a great person to talk to.

The late evenings were spent at the VMware Hands On Labs area, where I was on hand to chat to people and assist with any queries on the labs.

A free bar followed the Thursday HOL, and a small number of tequila shots may have been consumed….

The Friday involved more general and breakout sessions, the talk on “Mapping” by Simon Wardley (CSC Leading Edge Forum) was outstanding, it was recorded (for viewing by CSC staff) but the recording unfortunately doesn’t do it justice.

Despite nearly an hours’ overrun, a few of us managed to visit a nearby shopping mall in the afternoon (Converse for $40 anyone? £45 in the UK!), then headed back to the airport for the long journey home.

If you’d asked me on the first day whether I thought it was worthwhile, I’d probably have said no, however the value of the sessions from Tuesday onwards, and the networking opportunities (how often do you get to have a face to face with colleagues on a different continent?) were immeasurable. Overall, it was a great experience, and inspiring.