Thursday, October 27, 2011

VDCA410 Passed!

Today I have just received my results for my VCAP-DCA.
I opened my email

 I clicked to open that attached Exam Score Report


and I passed this time round!  I am not proud of the score.

I was pretty worried as after my examination, I discover I did several stupid things at my lab and one of which did not record down my results which means I have zero marks for it.  Instead of dumping the results to a file I actually missed out just one character and dump to screen and I proceed with the next task.

This is a very important lesson learned.

Needless say, though I did not score very well but a pass is a pass that was what was once told to me by a VCDX.I wish those who are doing the exam good luck and practice well.  Try to keep a cool mind and not be too anxious like myself.

Update
Receive my email notification of my certification and I am VCAPDCA-464.

Tuesday, October 25, 2011

View 5.0: Connection Experience

I did a setup of View 5.0 in my environment and have also use the Persona Management with folder redirection.  This demonstration was purely done on my whitebox and an ISP connection of uplink of up to 2Mbps.  The network speed isn't that great however we are able to see the performance.


Android View App to View Manager 5.0
Using the new View App version 1.2 which is no longer in Tech Preview, to test out the connection.  Pretty decent within my network.



View Client 5.0 to View Manager 5.0
Using my laptop with a View 5.0 client connecting to the View Manager to test out the connection within my network.



View Client 5.0 to Security Server 5.0
Connection was done through my 3G connection.  This is an extreme case not because I use 3G but my ISP uplink is only up to 2Mbps.  At the time of recording, I am also sharing with many people in my neighbourhood so this is not a good example.  However with this extreme measure, we can still see that View 5.0 is still able to deliver a decent quality desktop.  Do note that I have disable lossness using the PCoIP policy template on the PCoIP session variable.  Taking in a scenario whereby connecting from outside would not require a high image quality to an extend of see pixel to pixel.


Monday, October 24, 2011

VMware View 5.0: Android & iPad App 1.2

The android view app have been updated to 1.2.

I do not see any significant different from the Tech Preview version but I felt connection was faster. Maybe this was psychological.

Will post a review on the loading from a View 5.0 performance.


Friday, October 21, 2011

vSphere 5: Software iSCSI causes ESXi boot delay

Accordingly to KB2007108, this causes a delay in ESXi server boot up time.

This is due to the cause of each portal will try to connect to the target up to 9 times if it failed to connect, with more portal and more targets, this can result in a long delay.

Update: The patch for this issue has been release and a new ESXi 5.0 image is also available.  Please note when downloading you are provided with two different ESXi 5.0 images.  My recommendation is to use the latest though I do not know the reason behind why the previous is still available.

Wednesday, October 19, 2011

Openfiler remove/umap non existing volume

Some you of you might come across when using openfiler like myself of non existing volumes however it is still mapped to the target.

Or you wanted to remove those non existing volumes which keep showing in your volumes list.

First, to unmap the volumes from your target you have to edit the following file and restart your iSCSI target service.

iscsi_settings.xml in:  /opt/openfiler/etc/iscsi/targets

remove all ghost target, only from sub tag to tag (included), not then main tag.

Secondly, to remove the non existing volumes,

my volumes.xml from /opt/openfiler/etc/


The file should look like the below, remove   ... /> for each of your non existing LUN.




   
   
   


View 5.0: View PowerCLI

Recently from my previous post, I have upgraded my home lab to VMware View 5.0.

Everytime seems working.  For those who ain't aware, my ISP issue dynamic IP.  In such, I created a domain name from dyndns.org.

As to match my external IP to my View Security Server with my external dns name, I would need to use a script which was found from the source below.

I did some modification to suit my environment and corrected some mistake.  Below is my script.


However in View 5.0, it has to be launch using the View PowerCLI and Add-PSSnapin VMware.View.Broker is no long working anymore. (which it did in View 4.6).  This was documented in VMware KB 1034652.

A quick check with the View PowerCLI properties from Start Menu, I realize it is calling a powershell script from C:\Program Files\VMware\VMware View\Server\extras\PowerShell\add-snapin.ps1 so in such I added the change in my script as you can see.  This was documented

Now my script is running fine again and my log file is updated correctly.


Add-PSSnapin VMware.VimAutomation.Core
#Add-PSSnapin VMware.View.Broker
& 'C:\Program Files\VMware\VMware View\Server\extras\PowerShell\add-snapin.ps1'
# Name of the Security Server
$SecurityServer = "sg-ss"

# For logging creating a timestamp
$TimeStamp = Get-Date -format yyyy-MM-dd-H-mm

# Filling $CheckedIP with the external IP address, using whatismyip.com checkip.dyndns.com service
$wc = New-Object net.WebClient
$CheckedIP = $wc.downloadstring("http://www.whatismyip.org/")

$CheckedIP = $wc.downloadstring("http://checkip.dyndns.com") -replace "[^\d\.]"

# Now check the current ExternalPCoIPURL entry [0] is connection server [1] is securityserver
$CurrentSettings = Get-ConnectionBroker
$CurrentIP = $CurrentSettings[1].externalPCoIPURL


# Check if $CurrentIP starts with the IP address from $CheckedIP
# Used StartsWith because $CurrentIP has port address at the end
$Result = $CurrentIP.StartsWith($CheckedIP)

# Are IP address the same?
If ($Result)
{
     # Yes, both IP addresses are the same, do nothing, only write a log entry
     $row = $TimeStamp + "," + $CheckedIP + "," + $CurrentIP + ",nochange"
}
else
{
    # External IP is not equal to IP set in externalPCoIPURL
    # Changing the externalPCoIPURL
    Update-ConnectionBroker -broker_id $SecurityServer -externalPCoIPURL $CheckedIP

    # Check if it was succesful
    $NewSettings = Get-ConnectionBroker
    $row = $TimeStamp + "," + $CheckedIP + "," + $CurrentIP + "," + $NewSettings.externalPCoIPURL + ",changed"
}
$row | Out-File -FilePath "c:\check-ip.log" -Append




Reference


Update 24th Jan 2013
www.whatismyip.com now have more than just text and the container will store more than just IP which breaks the code and not updating the security server.  Have amended the code above with http://checkip.dyndns.org.  Tested and running fine now on View 5.1.2.

Friday, October 14, 2011

VCAP-DCA Experience

Today I make an attempt to retake my VCAP-DCA.  Yes, the first time I failed by 13 marks as I wasn't prepared enough and as this exam state it is related to Data Center Administrator.  Its more to administrator than a implementer exam (I work as a consultant so daily administration is a problem for me).

Beside studying the blueprint, practice is very important.  Luckily for me, I have a whitebox setup with ESXi 4.0 and I was able to test out vMA, VDS, etc.  This flashcard is also very useful.

Base on my first experience, I realize I am weak in vicfg commands and as well as performance charting and some hardening/security questions.

You will be tested base on ESX/ESXi 4.0 and vMA 4.0.  If you realize that vMA 4.0 and vMA 4.1 there are some changes and commands for e.g. vifptarget does not exist in vMA 4.0.

Unlike my first attempt, this time it was very smooth for me and I have my questions answered just in time.  Previously I was rushing the last 10 over questions as I was troubleshooting some mistakes I made.

A good advice, if you can't do the question, then do not bother to think and scratch your head as you will waste a lot of time and end up not enough for the rest of the tasks.

Also once you skip the question, it will be tough for you to go back later:
  1. Either you do not have the time.
  2. You cannot check the task you have to click back button several times and it take times to load as well.
You do not have to take note of the hostname, credential, etc as this will be provided at every task question.

Also there will be documents for your to refer to however, every time you refer to will take you time.  The machine response is quite slow so make full use of your time.  I did make my reference to the document once or twice.  By the way the resolution of the screen is about 1024x768.  You will need quite a bit of scrolling left right up and down.

Remember to clear your mind and don't be too excited over it, I made a mistake myself during one of my task and I know I will not have a single point for it even though I know the answer!  And yes not getting the results does not means you will have no points awarded, the steps you have taken to achieve it will be awarded base on the blueprint.

Overall this give me a great insight, and it was also the first attempt that I have prior experience from that I managed to pass my VCP 510.  As I mentioned in my VCP 510 posting, it is like VCAP-DCA in multiple choice.

Now the next 10 days is my crucial days to know if I managed to pass.  Meanwhile I will prepare for my VCDX application for the last panel in Germany next year.  This is more of an experience to go through the process and whether or not I succeed it's not important.  Important is the process where I managed to learn and pick up the knowledge and skills.

Feel free to leave me any comments regarding VCAP-DCA or VCAP-DCD and I will do my best to answer them.  But due to the non-disclosure agreement, I cannot provide you the questions.

Thursday, October 6, 2011

vSphere Design: Storage IOPS

A friend of mine asked me this question:
"Any idea roughly how much IOPS should we estimate for a VM?
My estimate is about 400 IOPS per VM – taking physical server for comparison.
SAS disk is about 200 IOPS and typically, a server would have 2xHDD for OS, so total IOPS is about 400 IOPS.

However, this number can be quite big especially if we have many VMs.
I’m now struggling with provisioning more capacity than needed for VMware environment to meet the IOPS requirement. This actually drive up the storage cost."


My reply as was follows after a short thinking:
Correction. Typically server with RAID 1 does not have 2 x IOPS. It has 1 x IOPS. Its not writing to two HDDs but its mirror to one by the raid controller.

Two things to note. Yes take the physical server IOPS using the worse scenario; 1. for database server, 2. for a normal app server.

Size two types of datastores, 1. with higher IOPS cater for database; 2. for normal app server. If you can use SSD that would be good definitely. vSphere 5 allows you for datastore cluster which is another added advantage.

Where n is no. disk for RAID, the calculation to get the IOPS required:


RAID 10 has the best IOPS for and can be use for database VMs.
read (n x IOPS)
write ((n x IOPS) /2 )

RAID 5 suitable more for normal VM which use more read and doesn't use a lot of Write.
read ( (n-1) x IOPS)
write ( (n x IOPS) /4)


Then ask the customer for each server what would be the percentage of read and write.  Knowing the percentage will allow you to size the storage more accurately unless you are able to provide the worst scenario and base everything on 100% write. 

IOPS Reference for HDD
15k SAS 180 IOPS
10k SAS 130 IOPS
7,2k SAS 80 IOPS
5.4k SAS 50 IOPS
SSD (SLC) 6k-30k IOPS
SSD (MLC) 1k-2k IOPS


To calculate the actual HDD IOPS here is a calculation reference from here.
Avg Latency = l in seconds
Average Seek time = R/W = S in seconds
Actual IOPS = 1/(l+S)

This is base on a rough sizing example.  There are definitely tools provided by storage vendors to size this better.  Some storage vendors might even have technology that can improve IOPS e.g. EMC fast cache, etc.

Update: Added portion from http://www.yellow-bricks.com/2009/12/23/iops/

(TOTAL IOps × % READ)+ ((TOTAL IOps × % WRITE) ×RAID Penalty)

So for RAID-5 and for instance a VM which produces 1000 IOps and has 40% reads and 60% writes:
(1000 x 0.4) + ((1000 x 0.6) x 4) = 400 + 2400 = 2800 IO’s
The 1000 IOps this VM produces actually results in 2800 IO’s on the backend of the array, this makes you think doesn’t it?


Update 26th Dec 2013
Added the calculation to find the IOPS for a HDD.

Tuesday, October 4, 2011

vSphere 5: VAAI Thin Provisioning Block Reclaim/UNMAP Issue


In the view of the new feature in vSphere 5 on VAAI, the reclaim of space from thin provision datastore whereby a file is deleted from datastore or when a storage vmotion is performed or even a snapshot consolidation. 
Performance issue may arise and create error when the above are performed.

Details can be found in this blog:

Knowledge base:

Sunday, October 2, 2011

ESXi 5.0 White-box


Two new white-boxes purchased to add to my current whitebox. Finally I can have a cluster!

Am going to try EVC for my Core2Quad box with my new i7 box.

Here are the specifications of my new whitebox if you are interested in building one. Note the chipset of the mainboard which is the crucial part whether ESXi can be installed on it. So far I discover P67 chipset seems to be working for all models.  So I bought the Deluxe series as the Evo series was out of supply.

I bought my Intel Dual Port from dealextreme.  It took very long to shipped and its not cheap as they are server NICs.

Components Parts
CPU Intel i7-2600 3.4GHz/8MB/LGA1155
Board Asus P8P67 Deluxe-B3
Memory 4 x Kingston 4GB DDR3 10600/1333 CL9
HDD Western Digital 1TB 64MB Caviar Black SATA 3
NIC Intel Pro 1000/PT Dual Port PCI-E X4
Video Asus EN210 512MB DDR2 HDMI (L/Profile)
DVD Samsung SH-S243S 24x Internal DVD Writer
Chasis Coolermaster CM Elite 431 Plus
PSU CoolerMaster GX550W Single Rail (5Yrs)
FAN CMaster BladeMaster 120mm R4-BMBS-20PK Fan
Flash Drive SanDisk Cruzer Blade CZ50 (8GB)























Assumed Support from Third Party Solutions

While I was doing some presentation slides for one workshop, I happen to look for 3rd party virtual switches support on vSphere. This is t...