Think IPM

Wednesday, October 22, 2014

Hi Again! From Gotham to the Sunshine state.

Wow, it’s been 3 months since my last post.  That’s THE longest timeframe I’ve gone without posting anything in the last 6 years.  Yikes!  Here’s the deal for those curious:

imageI recently completed a complete relocation from NYC to Central Florida with my family.  Two kids, spouse and cat. :) If you have never uprooted your family and moved them a state or 2 away, [SPOILER ALERT] IT IS A HECK OF A LOT OF WORK.  A ton of planning, a bunch of research, a lot of coordination and more than a fair share of luck.  Timing real estate transactions, school enrollments (which DON’T always sync up when moving south/north due to weather differences) and of course balancing client engagements. 

Fortunately, I am employed by an awesome company that worked with me and helped enable this move south for me and my family.  The end result was a smooth physical transition with very little interruption to client projects and responsibilities.  The technology (both home and enterprise infrastructure) that we have available to us now allow for the kind of remote working environments which probably weren’t possible even a few years back.  I have been and am extremely lucky to be working for IPM at a time when these types of transitions are possible.

So why the 3 month lapse in posts?  For me, Blogging is a lot like exercising.   When you are on a roll and in a rhythm, it just works.  You hit the gym, feel great afterwards and know you accomplished something.  But stop for a while and it becomes increasingly hard to get back into that mindset of sitting down, typing and posting.  The longer you sit on the sidelines, the harder it is to get back into the game.  I’m sure this is typical for almost any type of habit driven behavior.

So this post (although not technical in nature) is my way of getting back into the routine of posting and blogging and breaking out of my writing dry spell. 

Thanks for reading!
CARLO

Click Here to Continue Reading >>

Wednesday, July 16, 2014

Some VDI Myths and a PDF!

A couple of weeks ago, I had the opportunity to help present at a Goldman Sachs’ hedge fund event with my CIO, Phil Alberta.  The event brought together about 70- or so CIO/CTOs from various hedge funds that Goldman does business with.  We ran a breakout workshop that spoke about and worked through challenges and some of the decisions involved in a VDI project/deployment.

image 

We ran 3 very interactive sessions and here were some the takeaways we got from the audience (via a show of hands) that I thought were interesting:

  • Almost ALL of the firms were already running a version of Citrix XenApp.
  • All except a single hand were running VMware vSphere as a hypervisor (The lone hand belonged to a Microsoft Hyper-V implementation).
  • It was about a 60/40 split between Citrix XenDesktop and VMware View implementations.

I was actually surprised to see so many View installs with the level of XenApp penetration already in the crowd.  I’ve normally thought that once in bed with Citrix XA, Netscalers, Web Interface etc… XenDesktop becomes the natural progression for most clients.

As a takeaway for the audience, Phil put together a high level list of some of the VDI myths he has come across while talking with clients and engineers alike.  His presentation (which I will try to get him to link via SlideShare) was built around confronting and knocking down a lot of these myths.

You can read the ‘Myth’ paper here.

Click Here to Continue Reading >>

Monday, July 14, 2014

Installing the Citrix Provisioning Services Console on XenApp

-- Great How to by Jacques Bensimon:

First, why would you want to do that?

Well, besides the obvious ability to manage your Provisioning environment directly from your XenApp servers without having to remote into your PVS servers, installing the console also installs the PVS MCLI.exe command line tool and the MCLIPSSnapin PowerShell snap-in that provides a number of PVS-related cmdlets.  This opens up the potential for all sorts of automation ideas, including a server assigning itself a different vDisk (under whatever circumstances you decide) and restarting itself to run that image.  Let your imagination soar!

Okay then, what’s the issue?

The PVS console (MMC snap-in) wants to run under the .NET Framework v4.0, and takes steps via a .config file and a couple of Registry entries to force the use of that Framework both by the Microsoft Management Console mmc.exe (when it must load a .NET-based snap-in) and to some extent by other .NET assemblies installed on the machine, with unpredictable results.  At the very least, it will cause an ugly error message when starting the XenApp console (I believe AppCenter is its name this week, at least in XA 6.5) because one of its components (the Single Sign-on piece) is designed to run under the .NET Framework v2.0, but that’s probably the least of the issues it could potentially cause with other consoles and apps.

So, is there a solution?

I’m insulted you’d even ask that question! J  Here’s what you can do on your XenApp image:

1.       Install the (x64) PVS console.

2.      Copy mmc.exe and the (newly added) mmc.exe.config from %SystemRoot%\System32 to the installed PVS console folder.

3.      Also copy "en-us\mmc.exe.mui" from System32 to the same folder (in a new "en-us" folder).

4.     *Delete* mmc.exe.config from System32!

5.      *Delete* the (newly added) "OnlyUseLatestCLR" entries from "HKLM\SOFTWARE\Microsoft\.NETFramework" *and* from the same Wow6432Node subkey!

6.      Change the PVS console shortcut to explicitly specify the use of the "mmc.exe" copy in the installed PVS console folder.

Then what?

1.       To configure MCLI.exe for remote script execution (after installation of the PVS Console), run:

MCLI.exe run setupconnection -p server=pvs_server_name port=54321

This will create the following Registry entries:

[HKEY_LOCAL_MACHINE\SOFTWARE\Citrix\ProvisioningServices\Mcli]

"Server"="pvs_server_name"

"Port"="54321"

"User"=""

2.      To register the PVS PowerShell snap-in, run:

%SystemRoot%\Microsoft.NET\Framework64\v2.0.50727\installutil.exe "%ProgramFiles%\Citrix\Provisioning Services Console\McliPSSnapIn.dll"

To set up a connection to your PVS farm within PowerShell (to remotely execute all subsequent PVS cmdlets against the specified server), use the MCLI-Run cmdlet as follows:

MCLI-Run setupconnection -p server= pvs_server_name,port=54321

That’s it!  (Note that all of the above applies equally to any non-PVS machine on which you wish to install the PVS console and execute remote PVS commands via either MCLI.exe or the equivalent PowerShell snap-in, not just to XenApp).

Later,
Jacques

Click Here to Continue Reading >>

Friday, June 13, 2014

First Look: Citrix Provisioning Service Cache size

Although its been out for a while, I know there are still a lot of you that haven’t had a chance to check out Citrix Provisioning Server 7.1 yet so here’s a peek at a neat new feature you can look forward to.

ScreenClip(2)

The new Provisioning Services agent now gives you details on the cache being used within a read only session.  How to properly size the cache of a provisioned machine has always been something that comes up when talking to clients so I think being able to run a machine for a while under typical usage and check on the cache being used is a neat little enhancement to the agent.  You could always go to the file system and sniff it out yourself but since the information is so useful, why not put it front and center for display.   Eventually, I’d love to see this information make it up to the central console.

Monitoring cache sizes is useful for determining whether you under or oversized the local hard drives (especially in virtual environments) and also to help determine reboot schedules (to clear the cache).

There’s plenty more enhancements in 7.1 but I liked this one. :)

Click Here to Continue Reading >>

Thursday, June 12, 2014

Safe to upgrade to vSphere 5.5 Update 1 with NFS

After a couple of months of ignoring Update Manager’s rollup patch to vSphere 5.5 Update 1, it is now safe to click install. :)   (Although you might want to wait a week or two for others to test it before deploying into production)

Back in April, a vSphere bug was discovered in Update 1 that would cause an APD (All Paths Down) situation for some NFS datastores.  It was advised by VMware to not upgrade to this latest version if you had NFS datastores. 

They have finally come out with a patch that addresses the ADP issue.  You can read more about it on Duncan’s Blog below.

http://www.yellow-bricks.com/2014/06/11/vsphere-5-5-u1-patch-released-nfs-apd-problem/

Click Here to Continue Reading >>