Yesterday I passed the VCAP5 Datacenter Design exam and want to share my thoughts and study ressources with you in this post.
This was by the way my first VCAP exam and of course I knew this would be a real challenge and I didn´t got disappointed
If you read some other exam experience posts before you know that time is mostly a problem and many people don´t get all 100 questions finished in time. I finished the last question with ~15 minutes left of the available 4,5 hours (as non-native speaker). Properly you also know that the exam changed some months ago and you now don´t have the possibility to go back & forward between the questions anymore or to flag a question for review later. So forget all the tips about doing the Visio-style questions on the end etc. because it´s not possible any more, you need to answer all questions straightforward.
Before your exam starts you get the info how many Visio-style design questions you will get within the 100 questions. In my case I got 6 but I also read that some people got just 5, so it seems this vary from exam to exam. I think you should keep in mind how many of the design questions you already completed. In my case I made a note down to the whiteboard after each one. When I was at question 60 I got a little bit nervous, because I just had 1,5 hours left but then I realized that I already completed 5 of my 6 design questions.
Some of the multiple-choice questions are pretty long and many of the info’s you got provided you don´t need at all to answer the question. I always read the question and the answers first before taking a look on the scenario described on the beginning of these questions. So I knew already for which information I need to look for.
If I came to a question where I didn´t have a clue at all I read it maximum twice and then went over to the next one. Don´t spend too much time on thinking about it!
When I came close to the last question I was pretty happy about my time management and like mentioned above I also had some minutes left.
Beside of the questions I think the most difficult thing is to stay 4,5 hours full concentrated during you always fell to hurry up. After the exam I felt totally wiped out. So make sure to get enough sleep the day before to be fit for the exam. In my case I needed to travel 250km to the Pearson VUE center so I decided to take a hotel room to avoid traveling 3-4 hrs + 4,5 hrs exam in a row , which was afterwards the right decision.
In my opinion the exam shouldn´t be so long. I think only half or at least 3/4 of the amount of questions should be more than enough to prove your design skills.
Here you find a list of all resources I used to study for this exam:
-Book: VMware vSphere Design (watch out for the new on, which just got released)
-Book: Managing and Optimizing VMware vSphere Deployments
-Offical Course Material: VMware vSphere 5.0: Design Workshop
-Video Training: Designing VMware Infrastructure
-Video Training: vBrownbag Sessions
-Blog posts: DCD study guide out of many ressources (great work by Karim Elatov), Josh Odgers – Architectural Decisions & Gregg Robertson´s Mock Design
-Other: Of course I also read some of the PDFs, which are mentioned in the blueprint. If you have limited time to study I would recommend focusing more on the other material.
Last but not least you should make sure to watch the official DCD UI-Demo to get used with the Visio-style design tool.
Good luck to everyone taking the exam!
Eric is doing a great job for the community to put so much effort in this voting every year and his vLauchpad is a great resource for many people.
The fact that my blog got also listed the first time this year let me thinking about, which of my posts helped people properly the most.
I think the view-statistics of the individual posts are a quite good measurement:
1.1 – Implement and Manage Complex Storage Solutions
Skills and Abilities
-Determine use cases for and configure VMware DirectPath I/O
-Determine requirements for and configure NPIV
-Determine appropriate RAID level for various Virtual Machine workloads
-Apply VMware storage best practices
-Understand use cases for Raw Device Mapping
-Configure vCenter Server storage filters
-Understand and apply VMFS re-signaturing
-Understand and apply LUN masking using PSA-related commands
-Analyze I/O workloads to determine storage performance requirements
-Identify and tag SSD devices
-Administer hardware acceleration for VAAI
-Configure and administer profile-based storage
-Prepare storage for maintenance
-Upgrade VMware storage infrastructure
-PCI and PCIe devices supported
-Max. six devices can be connected to a VM
-Not supported: Snapshots/vMotion/FT/HA/DRS/Suspend & Resume/Hot add
-Intel VT-d / AMD IOMMU needs to be enabled in host BIOS
-VM needs to be HW version >= 7
-Adding PCI device to VM creates memory reservation
-vSphere client: Green icon = device active & enabled / Orange icon = change made & reboot required
-Saves CPU cycles on workloads with very high packet rates
-VM can use hardware features (TCP Offload Engine, SSL offload), which are not supported by vSphere
To configure pass-through devices on an ESX host:
- Select an ESX host from the Inventory of VMware vSphere Client.
- In the Configuration tab, click Advanced Settings. The Pass-through Configuration page lists all available pass-through devices.
- Click Edit.
- Select the devices and click OK.
- When the devices are selected, they are marked with an orange icon. Reboot for the change to take effect. After rebooting, the devices are marked with a green icon and are enabled.
Note: The configuration changes are saved in the /etc/vmware/esx.conf file.
To configure a PCI device on a virtual machine:
- From the Inventory in vSphere Client, right-click the virtual machine and choose Edit Settings.
- Click the Hardware tab.
- Click Add.
- Choose the PCI Device.
- Click Next.
From time to time it can happen that you have a session below the monitor section in HP DP which stuck and you are unable to abort it.
If you select Abort you will see this request in the log but the session still doesn’t stop.
In case it´s the only running session you can easily go to the task manager and kill the related bsm.exe process. This gets a little bit harder if you have multiple running sessions, because you will not see a relationship between the bsm.exe processes and the sessions.
The way to find out which bsm.exe relates to which session is the use of two PowerShell commands.
You get the process IDs and the start time of the sessions. Now you just need to take a look at the HP-DP monitor and check the start time of the session you want to kill.
Afterwards you can kill the bsm.exe via the task manager, where you will find the process ID in the column PID or you can simply use the PowerShell command:
The killed session disappears from the GUI after you restart your client.
By fact I find myself often thinking about the correct CLI command on Celerra & VNX systems, I decided to do this post for my personal notes. May it also helps someone else.
The commands are sorted by subject and I will expand this post by time to time.
By checking log files for a customer of some HP DataProtector (V6.20) backups, I noticed for multiple jobs the following error message:
[Critical] From: BMA-NDMP@xxxxxx.xxxx.xxxxxxx.com “xxxx_VTLRZ102_D01″ Time: 10.01.2013 08:12:18
[90:63] By: UMA@xxxxxx.xxxx.xxxxxxx.com@scsi4:0:0:0
Cannot load exchanger medium (Target drive is busy.)
All these jobs tried to use the same drive, which was busy for some unknown reason. Even a restart of a single backup job ended up with this error, if nothing else was running on the cell server.
I found the following workaround to get rid of this:
Below Devices & Media I located the affected device and selected Properties via right-click. Below the tab Control you find Busy drive handling (see screenshot).
I changed the default value temporary from Abort to Eject medium.
A restart of a job using this drive did unload the tape which was stuck. The session finished successfully. After changing the value back to Abort the job still finished without any problem.
It was impossible to bring the two Celerras and the two CLARiiONs into one domain.
If you logged into with one of the CLARiiONs SP IP-addresses into Unisphere you could see all 4 systems, but both Celerras showed “Not logged in” as status.
The other way around I couldn´t even login with Unisphere on the control stations with the user “nasadmin” and the scope “global”. And for sure if I selected the scope local, I didn´t see any of the systems, expect the Celerra i connected to.
Resetting the nasadmin password via CLI with the command..
nas_storage -modify id=1 -security
…didn´t bring any success.
After involving the EMC support and some hours later we found a working solution to get rid of this error / bug:
-Connect via putty to both control stations
-Browse to folder /nas/http/domain
-Take a backup of all files inside this folder
-Delete all files inside this folder (you may need to login as root)
-curl -kv “https://AAA.AAA.AAA.AAA/cgi-bin/set_incomingmaster?master=BBB.BBB.BBB.BBB “
(AAA = IP address control station / BBB = IP address domain master)
- /nas/sbin/naviseccli -h BBB.BBB.BBB.BBB domain -add AAA.AAA.AAA.AAA
(AAA = IP address control station / BBB = IP address domain master)
When executing the command in the last step the files you deleted on the beginning should be recreated. If you now login to the system (CLARiiON or Celerra) you should now see all systems in one domain and you can “jump” to each system from a single Unisphere GUI.
Please note that you may need to relogin into Unisphere, if you had opened a session during this process.