2011-12-29

Internet Access is needed to Install vCLI 5.0

I came a cross a case this week that I thought would be worthwhile sharing.

A client needed to install the vCLI on a Linux machine, and this machine was behind a firewall that was blocking access to the internet.

image

The process was supposed to be very simple. Download the Tarball, copy it to the machine, untar and then install. But during the install this message popped up:

Do you accept? (yes/no) yes

Thank you.

ping: unknown host www.vmware.com
Network is unavailable, please configure the network first otherwise please
install the following modules manually for use by vSphere CLI:

Archive::Zip 1.20 or newer
Compress::Zlib 2.005 or newer
Compress::Raw::Zlib 2.017 or newer
version 0.78 or newer
IO::Compress::Base 2.005 or newer
IO::Compress::Zlib::Constants 2.005 or newer
Class::MethodMaker 2.10 or newer
HTML::Parser 3.60 or newer
UUID 0.03 or newer
Data::Dump 1.15 or newer
SOAP::Lite 0.710.08 or newer
URI 1.37 or newer
LWP 5.805 or newer
LWP::Protocol::https 5.805 or newer
VMware::VIRuntime 0.9 or newer
WSMan::StubOps 0.1 or newer

Ok I get it – the installation wanted access to the web that would download those modules. Due to my predicament with no internet on this machine – I downloaded the all the Perl modules on another machine and copied them over.

And by the way the easiest way to download the modules is by using CPAN, but in this case - they had to be downloaded and installed manually – which by the way this is the process:

  • Download the Module
  • Unpack the tarball
  • cd <package_name>
  • perl Makefile.PL
  • make
  • make test
  • make install

But even after installing the dependencies – vCLI still would not install – either it would ask for a proxy – or throw a message simlar to the one below.

SOAP::Lite 0.710.08 or newer
LWP 5.805 or newer
LWP::Protocol::https 5.805 or newer
VMware::VIRuntime 0.9 or newer
WSMan::StubOps 0.1 or newer

Thanks to William Lam – he pointed me to a workaround that allows for the installation to continue without internet access. 

The solution is to comment out a few lines in the install script – which will allow the installation to continue. I understand that there is already a feature request to change this behavior in the next version.

2011-12-21

Mercedes Benz and BYOD

I spent a week in meetings with our global IT group in the UK last month. It is always good to see the people face to face that you interact with on a regular basis and build those personal connections. It makes working together a lot easier and more productive.

During our time there, we had an outdoor activity at Mercedes Benz World. But first I have a confessions to make.

I have never driven a Mercedes Benz – because of the amount of taxes that we pay here in Israel on cars – a Mercedes is way too expensive – well all cars are expensive.. but a Mercedes even more so. So driving one was a great experience.

Mercedes Benz World was built on old race track – which was in turn converted into a airstrip during WWII and now is being partially used as a track again.

So we got to drive a Mercedes-Benz SLS AMG - Man what a car!! The sheer power of this machine is unbelievable. Going from 0-100 km/h in under 4 seconds – it feels like you are taking off in an airplane. What also amazed me was the fact that it could also go from 100km/h –> 0 in less than 2. This car can stop on a dime. We tried the skid tracks – and with all the safety features – and also with the safety features turned off as well. It was a lot of fun.

The second part was an off track course that we ran through with a …….. At one point we crossed a obstacle and the car was balancing on 2 wheels – yes only two – the front right and the back left and it bounced from one side of the obstacle to the other. Scary as hell but utterly amazing.

Benz1

Benz2

You are probably wondering by now – why is he writing about all this? What does this have to do with virtualization an modern IT? Well I had a light bulb moment.

Traditional IT is changing. We are faced with challenges that we have never faced before. Our users are used to having things at their fingertips. They need resources – they go to a cloud provider and order something on their credit card – circumventing IT completely. People want their corporate email on their personal devices, on their iPad, their tablet, their phone. They want to access the corporate network from anywhere in the world.

And not always can the IT department keep up. We are looking to have things controlled, secure, safe. Sometimes that is not fast enough for the evolving world. There are those in IT that are afraid of becoming irrelevant. We cannot always provide a cloud infrastructure that can compete with external providers in such a way that will be viable for our company – for a number of reasons.

  • Cost
  • Space
  • Process
  • Security

So what was the light bulb moment you may ask? Actually two.

  1. When seeing that car balancing on two wheels – you come to realize a car can do so much more than you think it can. So can we – we should just have faith in ourselves and we can accomplish so much more than we think can.
  2. Mercedes Benz introduced almost every safety feature in modern cars we know today ABS, ESP, airbags, Collision testing and Safety Zones – well a lot. The car can do unbelievable things, take corners that you would usually flip over multiple times in a regular car. Avoid objects in the middle of the road, because of the additional sensors that control the car.
    You can do crazy things in Mercedes, but the car – because of all the technology inside will keep you safe and stop you from killing yourself (and others as well). Because of this we trust the car – it will keep us from harm. Even if we make a mistake – it will correct it and keep us from harm.
    We always have the option of doing something stupid – like turning off the safety features – but then we are on our own.

We as IT professionals should be the same. We have to learn to give our users the flexibility and the leeway they need – but have the measures in place to keep them and the company - safe and secure. If they do something that is not safe for them – the tools should be in place to protect them – sometimes without them even knowing it.

They surf somewhere shady on the web – we have filters in place that block these kind of sites – to protect not only them – but everyone else. They get a suspicious attachment – we clean it out before they even get it.

We have learned to trust a Mercedes – because of its reputation and its credibility. For our users to trustimage IT – and not see us as an obstacle – as a nuisance – as not relevant – we have to build up such a reputation and maintain that credibility. Some people feel that if it were up to IT we would be riding around in a Ford Model T, but that is because we have a bigger responsibility to the entire company. Opening up all bittorrent sites to the whole company – just because that one user needs a file on bittorrent – perhaps is not the best way to do things

BYOD is here, and there is no stopping it – otherwise IT as we know it will no longer be relevant. We just have to adapt much faster than before.

I would like to take this opportunity to wish all of a Happy Chanukkah and a Happy holidays.

As always comments are more than welcome.

2011-12-19

vKernel vOPS 4.5 Launched

vKernel announced the release of version 4.5 of their vOperations Suite today.
What’s New:
  • New Automation Features - vOPS' automation capabilities are enhanced with auto-deletion of abandoned VM images, auto-merging of unused snapshots, an additional automated remediation for performance issues and auto-calculation of future resource requirements.
  • Automation Controls - vOPS 4.5 also adds the ability to more finally control some automation processes. For example, VMs can be grouped, and then these groups set with minimum and maximum resource amounts that automated right-sizing will respect.
  • vSphere 5 New Feature Support - vOPS 4 introduced compatability to vSphere 5. vOPS 4.5 takes this integration further with support for storage DRS and storage clusters, new HA methods and inclusion of data provided by the VASA storage interface.
vKernel will try to provide “trusted automation” – and with vOPS 4.5, data center staff will be able to automatically
  • Delete abandoned VMDK Files
  • Merge VM snapshots
  • Solve VM performance issues
  • Power-off Zombie VMs
  • Resize vCPU and VM memory resources based on vOPS’ recommendations
  • Update management dashboards in real-time in external systems
  • Calculate CPU, memory and storage requirements at any point in the future
Each one of these automation features will help avoid the need to hire in additional administrators as environments grow. For example, a VM administrator can manually delete approximately 60 abandoned VM images in one hour. With automated deletion of abandoned VM images, an unlimited amount of files can be deleted in seconds.

Other features available in vOPS 4.5 include:
  • Application type tags to denote which application is running inside a virtual machine
  • Storage DRS and storage cluster support (available in VMware vSphere 5)
  • Access to VMware VASA interface storage statistics available in vSphere 5
  • Capacity planning calculation adjustments for vSphere 5 HA methods
  • Improved support and visibility into Raw Device Mapping (RDM) storage objects
  • Additional management reports such as graphs for performance issue counts
  • Auto-set VM metric trend alarms that generate warnings when abnormalities are detected
Here is a short video on the new features.
What’s new in vOPS 4.5

2011-12-09

VMware AppSpeed to be Retired


This is the excerpt of a message I received yesterday.(the highlights and graphics are my own)

Dear Maish,

We want to provide you with an important update about the VMware® vCenter AppSpeed™ product. As customers continue to expand the use of virtualization and cloud resources, we are focusing on delivering management solutions that can support the flexibility that enterprises require. As a result of this focus, we have decided not to produce additional releases of vCenter AppSpeed. vCenter AppSpeed will be end of availability for new license purchases as of January 3, 2012, however it will continue to be supported through September 15, 2012, in line with our General Support Policy.

image

As part of VMware continued investment in our customers' journey to cloud computing, we have developed a more comprehensive application performance management solution, VMware® vFabric™ Application Performance Manager (APM). As a customer of vCenter AppSpeed, we would like to offer you a special opportunity to leverage the comprehensive capabilities of vFabric APM. Customers who are active on a Support and Subscription (SnS) plan from December 1, 2011 until the vFabric APM entitlement becomes available in early January 2012 may exchange their related licenses of vCenter AppSpeed and upgrade to vFabric APM at no additional license cost. This upgrade will take place on a one-to-one basis. Future purchases of SnS for vFabric APM will have a higher list price than SnS for AppSpeed because APM provides significantly more application performance management functionality and has a higher license cost than VMware vCenter AppSpeed, however, SnS costs will not rise prior to customer’s next SnS term. Learn more about VMware vFabric APM.

image

AppSpeed was a technology the VMware acquired back in 2008 from a B-hive Networks.

The product as you see from the mail above, is to be retired and incorporated into the bigger suite of VMware vFabric Product family.

2011-12-06

Restarting vCenter Services - with PowerShell

Has it ever happened that you need to restart a vCenter service? I guess that you have been there before. Once upon a time I wrote a post that mentioned that there are not enough tools available today for us to troubleshoot the vCenter service which usually ends in a restart of the vCenter service.

When you want to stop the vCenter service you will notice that there are several services that depend on the vpxd service so they also need to be stopped. Windows will prompt you for this, of course.

Dependencies Prompt

But Windows will not start these services again automatically when you start the vCenter service.

For vCenter 4.x - it is the VMware VirtualCenter Management Webservices (vctomcat) service

That was actually easy to get.

Get-Service -ComputerName vcenter.maishsk.local -Name vpxd | select -ExpandProperty DependentServices | ft -AutoSize
vCenter 4.x

vCenter 5.x there are two additional services.

vCenter 5.x

So that means when you restart a vCenter 5.0 Service then you have to restart another 3 services as well.

PowerShell again to the rescue - and the Restart-vCenterServices function

Function Restart-vCenterServices {
	$services= @() 
	$services += (Get-Service -Name vpxd).Name
	## Add the dependencies to the variable.
	(Get-Service -Name vpxd).DependentServices | ForEach-Object {
		$services += $_.Name
	}
	## First put the services in the correct order and then stop them 
	$services | Sort-Object | ForEach-Object {
		Write-Host Stopping $_
		## -Force was used because the services have dependencies - even though they are stopped
		Get-Service $_ | Stop-Service -Force
		sleep 5
	}
	sleep 5
	## We need to start the services in reverse order 
	$services | Sort-Object -Descending | ForEach-Object {
		Write-Host Starting $_
		Get-Service $_ | Start-Service
		sleep 5
	}
}

This function was written so that it would work for both versions of vCenter.

Hope you enjoyed the ride…

2011-11-28

All of Us Can Benefit From TechFieldDay

Where would you go to hear more or find out about a new product or a new company? The Web? Blogs? the company’s website? Probably all of the above are true. But sometimes that information is
just… not… enough……..

I would like to share with you a resource that I find is of great value to me, and not well known (if you ask me).

I was looking to find out some more about Nutanix. They have a different look on the way you should virtualize your datacenter. Combine your storage and compute in one piece of hardware – provide a large amount of performance in one box, make it highly available and scalable – you have an Enterprise solution. I must admit – I am over-simplifying this a bit (a lot actually) but I encourage you to investigate this product more on your own.Nutanix

I went to their site to try and find our more about them. There are some whitepapers, some nice short videos, and resources, but I could not find the in-depth information that I was looking for.

How exactly does the product work? What is inside – what is the business strategy? So of course I could contact their sales department – and they would get back to me – then give me a sales pitch, and then back and forward until I would finally get in touch with someone technical that could answer my questions.

In comes TechfieldDay.

Stephen Foskett from GestaltIT has an amazing program where he brings together emerging companies some with new technologies (other established companies as well) and independent bloggers, to a two-day event full of presentations and discussions. These are highly beneficial both to the presenting companies and to the bloggers themselves. I had the privelege of participating in TechfieldDay #5 last year.

I remembered that Nutanix presented at a recently at Techfielday #8. Here is the link to the 2 hour presentation they gave.

I just sent an email to Stephen to thank him. I learned more about the company and their product from the two hour presentation that he recorded and posted – than I had learned from any of the the information on their site or anywhere else on the web. The reason being, the level of presentation delivered at TechfieldDay is not a marketing presentation (and if it is - the presenters will get their head chewed off). The questions asked are ones I would ask myself, technical, detailed and sometimes mind boggling. This is the stuff that interests me.

The fact that these sessions are recorded – and I can watch/download later - actually saves me and immense amount of time. True I spent 2 hours watching to a presentation (not in one stretch though) – but I am sure it will save me a lot more – getting the information I want about a company and their product and not the information their marketing would like me to hear.

I would advise you all - if you would like an in-depth explanation about a product - TechfieldDay is a great place to look. It is not only virtualization-centric, but also storage, backup, network – practically anything technology related can be found there.

2011-11-24

Set-VirtualSwitch and a Misunderstanding

I was doing some installations yesterday – and I wanted to add a second NIC to a vSwitch on a set of newly installed servers. Of course there were 10 of them, and I did not want to do this manually.

Easiest way to do it (besides installing them with a script correctly in the first place) was with PowerCLI.

From the Set-VirtualSwitch help:

--------------  Example 3 --------------

C:\PS>Get-VMHost *.128 | Get-VirtualSwitch | Select-Object -First 1 | Set-VirtualSwitch -Nic vmnic5


Add a physical network adapter named 'vmnic5' to the first switch of the host. Note that the 'vmnic5' adapter must not be assigned to
other virtual switches.

What I understand from that – you add an additional NIC to the vSwitch – it leaves the current assignment alone, it just adds another.

And this is how I started out.

Start

So I set out to add vmnic1 to the vSwitch

Get-VirtualSwitch -Name vSwitch0 | Set-VirtualSwitch -Nic vmnic1

The command completed successfully – but look what happened:

DCUI

Now this is weird part. In the DCUI – I had no NIC’s selected for the management network.

The case I had yesterday was that the transition on the switches was not fast enough – and I lost connection to the host. ILO was needed to fix it.

But back to this morning. I tried this on a workstation vESX and it produced to different results.

After running this command I was presented in the DCUI with the screen shot above but strangely enough I still had connection to the host.

ESXi Shell showed this:

Only vmnic1

Which was weird if you ask me.

I restarted the management Network through the DCUI and still had the same screen as before but I had now lost connection to the network

Lost Connection

I then added vmnic0 back into the vSwitch through the DCUI

Add vmnic0

I got my connection back..

Ping works

But strangely enough in the vSphere client I saw this.

vSphere Client

Which struck me as strange. I had this in the ESXi Shell

LightBulb!!

And now I finally understood what was happening.. The uplink was added to vSwitch0 – but not to the management network Portgroup, which personally I find strange.

What I actually should have done was this.

Get-VirtualSwitch -Name vSwitch0 | Set-VirtualSwitch -Nic vmnic0,vmnic1

This would add the second vmnic to the first – and would not have caused me all this trouble, and produced this:

Shell

vSphere Client

DCUI

So what can I (and perhaps you) learn from this whole episode?

  1. As always.. Test!! Test!! Test!!!
  2. The help could be more accurate (ahem… PowerCLI people…)
  3. Investigating a problem in depth – gives you a better understanding of what is going on.

Now if someone could possibly explain why the host behaved exactly the way it did, I would be grateful. Also why the default behavior of the PowerCLI command did what it did – would be also appreciated..

Hope you enjoyed the ride…

2011-11-17

Quest to Acquire VKernel – Say What?

That was an utter surprise to me. Seriously.

I don’t actually know really, what to think about this though. Quest and VKernel are direct competitors.

Quest acquired Vizioncore back in January 2008. It took 2 years until they discontinued the brand completely. So these words are something that still will need to be proven:

VKernel, a leading provider of capacity management products for virtualized data centers and cloud environments, will continue to operate as an independent subsidiary of Quest.

I find it hard to understand – why Quest would continue the product – vFoglight and VKernel are more or less the same, so to keep on development of two parallel products – does not make much business sense to me.

It seems the market is becoming more aggressive, and only the strong will survive.

This will be an interesting one to see how this rides out. And also to see if we will see any other acquisitions from other similar companies as well.

2011-11-15

Welcome to Our New Sponsor – Embotics

I would like to give a shout out and a warm welcome to Embotics for joining on as a new sponsor for the blog.

What does Embotics do?embotics

Embotics provides easy-to-use, virtualization management and private cloud automation solutions with the fastest time-to-value in the industry

Welcome aboard.

2011-11-10

Many ways to skin a cat (or write a script)

I wanted to share with you another small example of how there are many ways to do things in PowerCLI. I wrote a post a while back about How to Speed Up Your PowerCLI Queries and there I came to the conclusion that not always is using Get-View faster

I came across a case that a customer had a snapshot that was taken (never mind the reason) on all the VM's and this snapshot is removed with a script thereafter.

They wanted to track the number of commitments left to complete.

So one way was

$a = get-vm | Get-Snapshot -Name "Thursday Backup" 
Write-Host ""
Write-Host "$(Get-Date -Format hh:mm) - There are still $($a.count) Snapshots left"
Write-Host ""

and that to complete this gave this result

[12:17:42 PM] ~> measure-command {$a = get-vm | Get-Snapshot –Name
"Thursday Backup" ; Write-Host "";
Write-Host "$(Get-Date -Format HH:mm) - There are still $($a.count) Snapshots left"; Write-Host ""}

12:21 - There are still x Snapshots left

Days              : 0
Hours             : 0
Minutes           : 3
Seconds           : 41
Milliseconds      : 410
Ticks             : 2214108194
TotalDays         : 0.00256262522453704
TotalHours        : 0.0615030053888889
TotalMinutes      : 3.69018032333333
TotalSeconds      : 221.4108194
TotalMilliseconds : 221410.8194

But there is a quicker way to do this.

Function Get-Snapsleft  {
$a = Get-View -ViewType Virtualmachine -Property Snapshot | % {
	$_.Snapshot | % {
		$($_.RootSnapshotList)
	}
}
$b = $a | ? {$_.Name -eq "Thursday Backup" }
Write-Host ""
Write-Host "$(Get-Date -Format HH:mm) - There are still $($b.count) Snapshots left"
Write-Host ""
 $a=$b=$null
}
And how long did this take?

[12:21:57 PM] ~> Measure-Command {Get-Snapsleft}

12:26 - There are still x Snapshots left

Days              : 0
Hours             : 0
Minutes           : 0
Seconds           : 2
Milliseconds      : 359
Ticks             : 23590940
TotalDays         : 2.73043287037037E-05
TotalHours        : 0.000655303888888889
TotalMinutes      : 0.0393182333333333
TotalSeconds      : 2.359094
TotalMilliseconds : 2359.094

That is one heck of a difference….

The technique to speed this up I took from mattboren over here..

Check your scripts, see if you can improve them – sometimes you might be surprised as to what you can achieve.

2011-10-10

Enabling VLAN tagging on Redhat Linux

I came across this one today, and am putting it here to document it ofr my own benefit. I needed a physical machine with one NIC to be able to have two different IP addresses on two different VLAN’s.

On Windows I am not sure if that is possible by default.

So how would you do it on Redhat Linux (taken from Howto: Configure Linux Virtual Local Area Network (VLAN))

This is the scenario. One network card with an IP of 192.168.10.1 (VLAN 10). I needed another interface on the same physical NIC with an IP address of 192.168.20.x (VLAN 20).

First you copy the network configuration

cp /etc/sysconfig/network-scripts/ifcfg-eth0 /etc/sysconfig/network-scripts/ifcfg-eth0.10

My original file looked like this:
DEVICE=eth0
ONBOOT=yes
BOOTPROTO=none
NETMASK=255.255.255.0
IPADDR=192.168.10.1
USERCTL=no
PEERDNS=yes
TYPE=Ethernet
IPV6INIT=no

I needed to add in the VLAN info and change the device name (add VLAN=yes)

DEVICE=eth0.10
ONBOOT=yes
VLAN=yes
BOOTPROTO=none
NETMASK=255.255.255.0
IPADDR=192.168.10.1
USERCTL=no
PEERDNS=yes
TYPE=Ethernet
IPV6INIT=no

I then copied the file again to my second interface

cp /etc/sysconfig/network-scripts/ifcfg-eth0.10 /etc/sysconfig/network-scripts/ifcfg-eth0.20

and changed the IP address and device name

DEVICE=eth0.20
ONBOOT=yes
VLAN=yes
BOOTPROTO=none
NETMASK=255.255.255.0
IPADDR=192.168.20.1
USERCTL=no
PEERDNS=yes
TYPE=Ethernet
IPV6INIT=no

I then removed the original IP address information from
/etc/sysconfig/network-scripts/ifcfg-eth0

DEVICE=eth0
ONBOOT=yes
BOOTPROTO=none
USERCTL=no
PEERDNS=yes
TYPE=Ethernet
IPV6INIT=no

And restart the network service

service network restart

Of course the configuration has to be done as well on the switch side as well to allow the trunk of both VLAN’s

Switch#(config)interface Gi3/41
Switch#(config-if)no switchport mode access
Switch#(config-if)switchport mode trunk
Switch#(config-if)switchport trunk allowed vlan 10,20

2011-09-18

Connecting a USB device to an ESXi 5.0 VM

This is something we have been asking for years and it was only available up until now in VMware View. We can finally connect a USB device to a VM running on ESXi. No additional licensing required!!!!!

From the What's New in vSphere 5.0

What's New

Could it be that simple? Well actually yes it is.

This is my host (in this case the free vSphere Hypervisor)

image

Here you have a VM Virtual Hardware version 8 (I was not able to test with Version 7 so if you would like to try and add a comment here - that would be great!!)

VM details

Add a USB Controller to the VM

Add Hardware

You can choose either USB 2.0 or USB 3.0

USB 2.0USB 3.0

And here is the VM Configuration after the addition.

VM Config

After adding the USB controller to the VM, these parameters are added to the VM's configuration (.vmx) file.

vmx1 

After the machine has come up it now has a USB controller in its device manager.

Device Manager

In the vSphere Client I can now attach USB devices that are connected to the my computer to the VM running on this ESX host.

connect USB

When I connect the USB device to the VM I get this warning - which is similar to the one you get when doing the same with Workstation

warning

Once connected to the VM - it shows up in the Device manager and in this case also as a local drive.

New Disk

New Disk2

And of course the details are updated in the .vmx itself

vmx2

This opens up a whole new world of connecting peripheral devices to the VM. I do wonder though what it will mean to all of the companies that have created solutions that were able to solve these issues until now with a USB-over-IP solution. (Digi)

It would be interesting to hear what kind of use you could make of this new feature in your environment. Feel free to drop a comment below.

2011-09-13

netstat for ESXi

The title of the post is actually misleading - on purpose - because there is no netstat for ESXi. The reason that I bring this up today is because of a Twitter conversation from today regarding SSH access and VMkernel interfaces. I was looking to see which ports were open and what interfaces were listening.
But that is a different post.

What is netstat? according to Wikipedia:

netstat (network statistics) is a command-line tool that displays network connections (both incoming and outgoing), routing tables, and a number of network interface statistics. It is available on Unix, Unix-like, and Windows NT-based operating systems.

Why would you use it? For one thing for example, to check if a host has an open connection on a certain port, if it is listening on a certain port - for troubleshooting purposes would be the proper answer.

So how do you get that information on ESXi?

Trying netstat on an ESXi host does not work - because that command is not there - see the screenshot below.

No netstat

Well that is not good - if the command is not in the busybox console then how would you go about getting that information? Well of course the clever people at VMware have already thought about this and have exposed all this information through esxcli. William Lam wrote a great set of posts on esxcli
esxcli Part1 - What is esxcli?, esxcli Part2 - Automating esxcli using vMA and esxcli Part3 - Automating esxcli using PowerShell

This is how you would go about getting the information from esxcli. (Be aware the command differ according to the different versions - 4.x is not the same as 5.x)

esxcli network ip connection list

esxcli1

That is fine and dandy - but to get that info you need to either:

  1. have access to the DCUI (and have it enabled of course)
    or
  2. access remotely with SSH (and also have it enabled of course)

But what if you do not want to enable neither of the above - that means you have to do it remotelyand for that you have two options, vCLI or PowerCLI.

The vCLI way

esxcli --server esx1.maishsk.local network ip connection list

vcli1

But me being more of PowerCLI guy I would do it like this.

The PowerCLI way

$esxcli = get-esxcli -vmhost esx1.maishsk.local

$esxcli.network.ip.connection.list() | ft

PowerCLI1

Output is almost identical - just that in the case of PowerCLI the values are returned as a set of objects - a  VMware.VimAutomation.ViCore.Impl.V1.EsxCli.EsxCliObjectImpl object to be precise. Once these presented as objects I can start to mold and dice my results to my liking.

For example - I would like to check if there is any connections open on port 80 (http) - with vCli - this is not so simple - because you are working essentially in a DOS window - so filtering is not the easiest with findstr. Using the console or SSH is easier - a simple grep will work as you can see below.

esxcli network ip connection list | grep :80

esxcli2

With PowerCLI

$esxcli.network.ip.connection.list() | where { $_.LocalAddress -like "*:80" } | ft

PowerCLI2

I hope you can see that the options this way are pretty much endless - like filtering all connections to show only those from a specific IP, or a complete subnet.

So that is how you netstat on ESXi….

2011-09-08

VCP 5 Beta Exam - a Pass - but it aint easy

I received an email today

September 7, 2011

Dear VMware Beta Candidate

Congratulations!  You have passed the new VMware Certified Professional on vSphere 5 (VCP5)

certification exam. Thank you for your participation in the beta exam.  Your input and participation were invaluable to this process.

We will be adding this certification to your transcript within the next three weeks.  You will receive an email notification with additional instructions once your education transcript has been updated.  Physical certificates will be sent after your shipping address has been confirmed.  Please contact certification@vmware.com if you do not receive your notification.  Please include your Candidate ID from your exam score report.

Thank you for participating in the VMware Certified Professional on vSphere 5 (VCP5) beta exam and congratulations once again on becoming one of the first VCP5s!

I think this is the last time I will take a Beta exam. It is a lengthy and not trivial process.

Let me run you through the flow of a Beta Exam.

  1. Beta is announced.
  2. Hope you know someone - that will get you in on the Beta of that product.
  3. Play with the product on your lab.
  4. Close enough to GA of the product - depending on the case a beta Exam will be come available.
  5. Pray there is an available slot in the tight timeframe that the beta is open - usually 7-10 days.
  6. Book the Exam - almost always it costs money (less than a regular exam)
  7. Look over the blueprint - and rely on your experience and knowledge of the product, because there is no-one who can help you out here. You are on your own.
  8. And this is the worst - sit for about 4 hours - going over 100-200 questions, some of them are so whacky - some are intentionally wrong and some plain stupid. This is the worst part of the whole process. Sitting for that amount of time is not an tedious and exhausting experience.
  9. And then wait - 6-8 weeks after that - to see if you passed or not. Usually people that have taken the live exam after launch get their results before you do.

I do feel that participating in a Beta Exam is something that is beneficial to the candidate in a way - you are actually saving money. But this should also be perceived as a service to the community - because you are acting as a "guinea pig" for VMware to help refine the exam when it goes live and offered to the masses.

I said after this one - I will not do any more beta exams, it takes too much energy - but getting a pass mark - it always gives the extra boost. Let's just say - I will decide next time it comes up what I will do.

image

Congratulations to all those who passed the Exam!!

2011-09-07

VMworld Labs - Addendum

As always the VMworld labs are always a hit. Everyone wants to try out the new technology, see the new stuff, and get a hands on feeling with the latest and greatest. To add to my HoL post from VMworld, the Fast Pass was gone within a very short time, so my apologies about that one.

Borrowing from Duncan Epping's post VMworld Labs 2011.

2011 Labs Prelude

The labs were (as they always are) impressive, two monitors at each seat, with a thin client. A great user experience, albeit a bit sluggish at times. The aquarium was nice touch and a good visual aid as to what was was being deployed during the show. By the way - the racing car dashboard from last year - beats the pants off the aquarium. Just sayin..

One of the sessions I attended was LAS4000 VMworld Labs Hardware Architecture. Here I learned several things some I found surprising, and others not. The not first.

  1. The entire infrastructure was running in the cloud. Why was this not surprising - because - Cloud is what VMware is selling, it is their vision. It is what they are betting their chips on. So they have to (and rightfully should) present solutions using this technology.

I was surprised to hear..

  1. The entire Lab Environment  was running on NFS. Not FC, not FCoE, only NFS. That is change from previous years. If my memory serves e well, last year is was mostly FC and a small amount of NFS storage.
  2. EMC was not the only player for the underlying spindles. Nexenta (a much smaller company) was used to power a good amount of the storage used for the labs. I was originally led to believe that it was only Nexenta storage, but that turned out to not be entirely accurate.

A few closing comments regarding the organization.

  1. Extending the labs to provide options to test also the partner integration is a great addition - just it should be kept to a minimum (like it was). 
  2. Waiting in line for 45 minutes to get a lab seat is not good. Not for those who want to take the labs. That time spent in line is wasted - it could have been spent on the floor, or in a session. There has to be a better way of doing this. Perhaps pre-register for labs (like sessions), but also allow for walk-by's as well. Those who have pre-registered will get quicker access. That way those who really want to do the labs, and close off time on their schedule and are there on time - do not have to spend an hour in line.
  3. The real world scenarios - were perfect!!! Whoever had that idea - brilliant!! Well done!
  4. Looking at the number above from Duncan's post I noticed something which troubled me. On his post VMworld Labs wrapup 2011 – Las Vegas2011 Wrapup 
    I noticed that the number of Labs was less than last year and the number of VM's was only slightly higher, despite the fact that there were more hours, an increase of 2% despite the increase of 13% more lab hours. Again a record was broken, but not by much.

    labs1

But what really troubles me is the huge difference in goal vs actual. 148,000 vs. 225,000, that is a really big miss. It would be nice to hear what actually was the reason for not achieving the pre-defined goal. Was it performance? Was it architecture? Was it logistics? Was it over-zealous goals?

I do want to stress again. The labs are amazing - the technology behind it is amazing. The option of trying something hands-on is wonderful - and for some it is the main reason to come to VMworld. If the post-mortem analysis could be shared - I am sure that this is something we would all benefit from. This is not to criticize any of the the VMware Integration Team in any way. So if someone could share more details on why the big differences, I would be grateful.

2011-08-31

VMworld HoL (Hands on Labs)

As always they are brilliant!! The lines are long - at some time during the days it calms down but it is well worth the wait
I took one lab yesterday HOL24 with the Horizon App Manager. It looks like a really great product.
Some info from yesterdays VMware mail.Take note of the last sentence.
Hands-on Labs
Even though the Solutions Exchange closes tonight at 5:30pm, you can stop by and take a Lab until 10:00pm – or- stop by tomorrow morning at 7:00am.
So far 4,900 labs have been with 54,000 VMs deployed, and any attendee who has completed more than 4 labs can now get a Fast Pass which will get you to the front of the line.

2011-08-29

The ESXi Quiz Show - #VSP1956

This is going to be fun!! Why you may ask?

  1. because it has never been done before
  2. it is a great idea
  3. more than 400 people have already registered for this session.

This was Duncan Epping's brainchild.Bring a group of vExpert's together - and make an entertaining session out of it all. VMworld is all about knowledge, well ok also about technology a bit, and beer a bit as well, but what good is knowledge if you cannot share it with others?

There will be 3 Teams.

vExpert Team – The Raging vBulls
  • Chad “Warrior-Monk” Sakac
  • Jason “vTerminator” Boche
  • Maish “vBeliever” Saidel-Keesing
  • Eric “Link Master” Siebert
vExpert Team – vPredators
  • Vaughn “Sgt NFS” Stewart
  • Tom “VDI Warrior” Howarth
  • Mike “Axel” Foley
  • Scott “VMGuru” Herold
VMware Team – vRaminators
  • Frank “distributed” Denneman
  • John “VCDX 001″ Arrasjid
  • Kit “VC Ops” Colbert
  • Massimo “Hybrid Cloud” Re Ferre

I like all the "nicknames" we all have attached to the our names (and I will not tell you what the original name Duncan thought of for me [movie star])

The vPredators will compete against the Raging vBiulls - and the winner of that round will go up against the vRaminators.

It is going to be a blast, buzzers and all all, John Troyer, Duncan Epping, and a special mystery judge will be present as well.

If you have not already signed up - do so!!

2011-08-28

VMworld 2011 Day 0

So my day started off not so well. Irene - you must have heard about that gal somewhere, that small little storm that was blowing up on the East coast? Seriously though, I hope that it does not cause too much damage and everyone stays safe and healthy.

My original itinerary was to fly at 00:05 on Saturday night from TLV to JFK, a stop-over of 2.5 hours and from there to Las Vegas.

I was notified that the flight to JFK had been cancelled. Bah!!! I started to look at the alternatives that were available from Delta, and on their site they said that nothing was available un till Tuesday. An then I felt really bad. I have been looking forward to coming to this event for almost a year, and because of a higher power - despite all my planning and work - this would not happen.

I would like here to publicly commend the staff from Ivy Worldwide (they are partially sponsoring my trip) who went above and beyond to get me another flight. Within an hour - they had Delta on the phone. and had me booked on an alternate route to the US.

This meant that I was going to arrive in Las Vegas 5 hours later - but hey - that's life. I flew from TLV to Amsterdam, short stop-over, from Amsterdam to Minneapolis, another short stop-over, and then from there to Las Vegas. A long trip, not enough sleep, but what we won't do to make our dreams come true.

I am currently writing this in the middle of the second leg of the journey, and still have another 7 hours of travel, but I am looking forward to arriving later in the day, Joining for the rest of the vExpert program update and maybe getting some sleep somewhere in the next week :)

That was day 0. I am almost there!!

2011-08-26

VMworld 2011 US Program Guide

Program Guide

The Program Guide is now available.

Get it here

(No breadcrumbs that I could find regarding what will be happening next year.. Sorry!)

VMworld 2011 Mobile Apps

This is a repost of the original blog posted on the VMworld site

Take VMworld with You on the Go

The Mobile Web App

iOS Application for the App Store

Make sure you have VMworld with you while at the conference with the official VMworld 2011  mobile application. It is a show program, breakout session schedule builder and Solutions Exchange guide for attendees of the VMworld conference. The application is synced to email stations throughout the event and is compatible with iPhone, BlackBerry, and all other smartphones.
With the mobile application, you can:

  • View sessions by day and track
  • Schedule Breakout Sessions
  • View Hands-on Labs
  • Browse exhibitor listings by name
  • Find your way around the expo
  • Create a personal agenda containing the Breakout Sessions you pre-registered for
  • Complete session surveys
  • Follow the event’s Twitter stream
  • Receive the latest conference news

Visit www.vmworld.com/mobile on your phone to download and install the official VMworld mobile application.

2011-08-24

I Want More Bandwidth - but KISS - #BRC2K11

This will be the 3rd and final post for the Bloggers Reality Contest that I am participating in before VMworld 2011.
The topic we will be dealing with today is converged networking.

Why is it necessary?

A few years ago, when I was starting out with virtualization, I started out with rack mount servers. From the start I knew that I would be using Network attached storage - and that the minimum amount of network cards I would need for my for these ESX servers was 6 1Gb NICs, two for management and vMotion, two for network traffic for the virtual machines and two more for iSCSI / NFS. It became apparent very quickly that this does not scale, for a number of reasons.
  • The connection of each ESX host to two redundant switches, become a cumbersome process, which takes up a considerable amount of time both of the Networking team and the Server Team as well. Connecting the ports to the correct switches, making sure that the VLANs are set for each network port and so on.
  • It became evident that using 6 ports for each ESX server would leave no free ports for the rest of the servers in that rack. Each patch panel has 16 ports by default. 2 ESX servers per rack - eat up almost all of the ports immediately, which means either running more that 16/24 ports to each rack, or limiting myself to how many ESX servers I can install in each rack.
  • In short, this is not an easy process
So how would you solve this? That is where 10Gb Ethernet comes in. Instead of running those six 1Gb NICs I mentioned above - run two 10Gb NICs - that will give you all the throughput you had before, and then some, of course redundancy included.
Unfortunately though, this does solve all of your problems. What happens if you also need connectivity to a Fiber channel array? That means more cables coming out of your servers - more port being used (be they ethernet or SAN fabric).
And all of the above of course is relevant also for the Storage stack as well.

What are the solutions out there?

If someone were to ask me who are the two major players in the converged networking game today, I would instinctively say HP and Cisco. The solutions they provide are similar in some ways, but are very different in others.
I would like to stress this is my own understanding of both of these solutions - I could be mistaken in some of the details, and as always would be happy to hear your feedback if there are any errors in my description.
Cisco control the network stack. I think that this is pretty much agreed upon by almost all. They have also gained a large market share in the past two years for the converged systems market. The continued UCS growth is something that HP cannot afford give up because it is taking good percentage of their market share.
HP have Virtual Connect and their Flex-10 technology for HP Bladesystem that will allow you to converge the both your Ethernet and Storage traffic over the same network card,
Cisco have their CNA cards that will allow you to pass Ethernet and FCoE over the same network card.
HP keep all traffic internal to the Chassis internal - thus internal VM / vMotion traffic stay within the chassis, as opposed to UCS which sends all the traffic up the top of the rack, regardless.
It is extremely difficult to explain in such a short post which one has more benefit, which one is better and who has the better solution. The answer to that question is actually very simple. The vendor that has a solution better suited to your needs - is the best solution.

Summing UP

This is the last post of a series of three that I (and several other bloggers) have written as part of the Bloggers Reality contest.
We touched on topics that were related to HP's solutions and products that we were exposed to. Some of us are very familiar with their products, some of us did know anything about them at all. Despite that, we all had some great articles published on the each of these subjects. We all learned new things, and most of all (I think I can speak for all of the contestants) we all had a good time!
I would appreciate your comments on this post, what you thought about this blogger's contest, and what you would have liked to see more of, and of course also less of.
Please remember this is a contest and your vote for this post is needed (and your comments as well).

One last note - KISS - Keep it simple stupid (for all of you who did not know what that was)

2011-08-18

How many vMotions?? - The PowerCLI Way

As an addition to Luc Deken's and Jonathan Medd's great VIProperty Module, I wanted to add one property that I found useful.

Ever wanted to know how many vMotions have been made in your clusters?

Using the GUI is fine, but not for more than one cluster. You can see the option on the summary tab of your cluster

image

And of course the PowerCLI way.

New-VIProperty -Name NumberVmotions -ObjectType Cluster -Value {
	param($cluster)
	$cluster.ExtensionData.Summary.NumVmotions
 } -force

And now you have a new parameter called NumVMotions (and yes that the way VMware defined the property)

[18:11:39 PM] ~> Get-Cluster RD | select Name, NumberOfHosts, NumCPU, NumberOfVMs, NumberVmotions


Name           : RD
NumberOfHosts  : 6
NumCPU         : 52
NumberOfVMs    : 384
NumberVmotions : 2170

And for all your clusters..

Get-Cluster | select Name, NumberVmotions | ft -AutoSize

Great Module!!!!

2011-08-14

Orchestration Will Rule Them All #BRC2K11

On Wednesday night I participated on session 2 of the Blogger Reality Contest. From the rankings that were published on Wednesday - I am actually not doing so well (so your continued and strengthened support would be appreciated) - but hey this is not about the competition - it is all about the participation. It is great to interact with a whole new set of people, and I am sure that the relationships built during these three weeks will last. But let's get back to the technology part of it.

Wednesday's topic was Converged Systems. I will not bother you all again with the converged part, we covered that all last week, and the three topics covered were:

  • HP VirtualSystem
  • HP CloudSystem
  • HP AppSystems

The one I would like to cover here in more detail is HP CloudSystem. A while back I wrote a post about The Datacenter - in a Few Years From Now where I presented this graphic below.  Hypervisor for Hypervisors  The relevant part I think is the one here below

And what does this have to with cloud though? The way I see cloud vision is having your pool of resources from which you can provision your applications and vApps to your end users. At the moment this is limited to a cloud for each Hypervisor. I cannot have a VMware vCloud that has Hypervisors that are not ESX. I cannot have a Microsoft Cloud with Hypervisors that are not Hyper-V and so on.. Yes, there are some products that will allow you to manage the multiple clouds under one "umbrella" and place the VM in your Hyper-V cluster or ESX Cluster or RHV Cluster according to different criteria. But today they still need to be separate clusters. Utilizing such an idea will make my infrastructure completely vendor agnostic - and I (the customer) could pick and choose whichever hypervisor suits me.

VMware has their cloud product - vCloud Director. Microsoft has their Private Cloud, Redhat has their CloudForms, Citrix has their Project Olympus. The purpose of this post is, not to get into who has the best cloud offering, nor who has the more mature product. The reality of today is that most infrastructures are running mainly on one platform, and one platform only. Will that change in the future? I think so - in my post Why Should you Care about Veeam Support for Hyper-V, I started to talk about the market share that other hypervisors are gaining and will continue to do so in the future.

I spent a hour on a conference call last week with a colleague who had built their own private cloud solution. I was interested in hearing the details behind the platform and how it was built, and I was quite surprised to actually hear that the whole infrastructure was built on VMware - but they had decided on a 3rd party solution for the orchestration of the the cloud and not to use vCloud Director. The reason being mainly because of the lack of integration the system actually has into current business process. The amount of time it actually takes to integrate into their current CMDB, and a number of other reasons as well. Looking back on the conversation I actually am starting to see that the above graphic that I made should actually be changed. A layer that translates the operations from kind of hypervisor to the other will not actually be the correct solution or the right way to go.

The way to actually go would be to create a system that would allow you to manage Hypervisors/Clouds from multiple vendors, all under one management portal. This way you as the Admin 1would still not care if the machine is running on a VMware / Hyper-V / Xen / Redhat cluster - for you the machine will be managed the same way regardless. The Orchestration software will be that to deploy the virtual machine according to your pre-defined criteria of cost / SLA etc. and present a unified interface to the end-user so they will be able to access their machine - again completely ambivalent to the what the underlying  hypervisor is.

HP CloudSystem HP CloudSystemcould be such a platform. Providing a layer of automation that is external to the the actual hypervisor - could actually be the answer. Treating Server / Storage / Network / Application Pools as exactly that - just pools - and to the end-business user who does not care what platform they run on - but rather that they work, this abstraction should work just fine.

You could even take this one level further by adding an external public cloud to your pool of services. Something like Cloudbursting.

Cloudbursting is an application hosting model which combines existing corporate infrastructure with new, cloud-based infrastructure to create a powerful, highly scalable application hosting environment.

To explain this in my terms, you use the public cloud as an extension of your compute resources to extend and scale as needed for your application.

The HP CloudSystem core is built on the HP BladeSystem architecture and includes the Matrix Operating Environment that enables rapid provisioning of complex infrastructure services and adjustment of those services to meet changing business demands. It also includes HP Cloud Service Automation software to manage the entire cloud lifecycle. Cloud Service Automation provisions applications, manages and monitors the cloud, and provides a single service view across the cloud and traditional IT.

HP CloudSystem’s core capabilities can be extended with

Cloud Maps: tools and best practices that enable CloudSystem to quickly and easily provision various kinds of common application environments from major vendors such as VMware, Oracle, SAP, and Microsoft. Cloud Maps can substantially reduce the time and effort needed to deploy CloudSystem services.

  • HP storage
  • HP networking
  • HP Software

To sum up this post - The future of the cloud (and your datacenter) will be in the automation layer and will look something like this. It will all lie in the orchestration layer.

Orchestration Layer

As this is a contest your vote does count. Please add you vote in box below. Thanks. Your comments of course are always welcome.

(Small Print) HP is not the only one to offer such software or orchestration platforms. There are several other vendors and other 3rd party companies that provide the same or similar technologies. HP is one of the sponsors of this event, which mean they are partially footing the bill for my trip to VMworld this month. I am highly grateful for the opportunity and honor but I am under no obligation to write anything positive (or negative for that matter) about any of the presented solutions. The opinions and thoughts above are my own so take them as they are.