VIOPS: Proven Practices for Deploying and Managing VMware : Popular Threads - All Communities

Migrated VCDB from SQL Server to Oracle?

schambers — Wed, 18 Mar 2009 17:25:28 GMT

Has anyone out there ever migrated their VCDB from SQL Server to Oracle?

It's a common question, but I've never seen an answer. I presume, like most people, that a few scripts are required to dump/load the database as well as configure things right.

Cheers

Steve

What other applications do we need proven practices for?

schambers — Tue, 02 Sep 2008 20:52:45 GMT

We are planning a proven practice for SAP.... what other applications?

VCB Backup solution

Dracul — Tue, 26 May 2009 14:44:50 GMT

Complete Backupsolution using VCBExport and Browse-Start over SAN

Author Thomas Hierl SCA Packaging Containerboard Deutschland GmbH

Szenario:

3 ESX-Server 3.5 connected to SAN with 3 VMFS-Volumes (Data1, Data2, Data3) and 2 RDM´s (DB-Server and Fileserver)

VCB Backup Proxy with c: d: e: k: (drive K is San 3,6TByte)

d:\VCB

k:\Backup -> Target for other Backups

k:\mnt -> Where VCB creates the Backups

k:\Save_NT01 -> Target of File Backup if the Fileserver

k:\Save_VM -> Target of the vmdk, vmx and vmfx Files Image Backup of VM-Ware Boxes

What does it

these Scrips are backing up all nessesary Data using vcbExport and browse-start over the SAN (not Network)

Start Points over scheduled Tasks

K:\Backup\backup_vmdk.bat -> Backing up all nessesary images to k:\Save_VM

K:\Backup\Start_BKUP.bat -> Backing up the Fileserver Data to k:\Save_NT01 and all other Backups to k:\Backup

K:\Backup\Clean_Log.bat -> Moving and cleaning the logfiles

Veritas Backup exec is saving every day the Target Directory to Tape

k:\Backup

k:\Save_NT01

k:\Save_VM (only Weekly)

Needet software

Robocopy

WinSCP

Putty

Usefull software

Eraser (cleans the unused space with '0000' perfect to compress the vmdk-files (180GB -> 62GB)

Please install on all Servers witch Drives contain much deleted Files

This current version of Eraser is maintained by Garrett Trant. (http://www.heidi.ie)

Sami Tolvanen, the original author of Eraser was a student in the university of technology in Finland.

You can learn more by visiting his home page at http://www.tolvanen.com/sami/

Installation and Operation

S0000625 = vcb Proxy and Backup Server (real HW with connection to SAN See VCB-Installation)

S0001176 = Fileserver (c: d: = vmdk, e: = RAW (Virtual used for the Backup Files script))

EC084479 = Virtual Center (c: d: = vmdk)

C:\WINDOWS\system32\drivers\etc\hosts add:

10.112.83.190 VMware ec084479.ecdomain.net #VMware VirtualCenter

this is needet for the SSL because the certificat name is VMWARE and not the real name of the Virtual Venter. Shit bug.

BackupDisk.vbs (thi is the script which is backing up the VMDK´s and starting the get_vmx.bat)

fill these Constants

Const VCB_INSTALL_PATH = "D:\VCB"

Const HOST = "VMware" '// Name of the Virtual Center (see Host-File)

Const USERNAME = "*********" '// Username of the Virtual Venter

Const PASSWORD = "*******" '// Password of the VC-User

Const ESX_PASSWORD = "*******" '// Password for the Rootaccount for the ESX-Server

Const BACKUPROOT = "K:
mnt"

get_VMX.bat (is creating the script and starting the script to copy the .VMX files to the Backup destination)

this line:

echo open root:%3@esx1.aschaffenburg.packaging.sca.se >> d:\vcb\PuTTY\script.tmp

You have to type in the DNS-Name of the ESX-Server which has a connection to all VMFD-Volumes (Data1-3 etc...)

User for the Copy is root and the Password is defined in BackupDisk.vbs

d:\vcb\config\config.JS

please insert username and password

How are you managing ESXi today?

schambers — Thu, 11 Sep 2008 16:14:36 GMT

With the new ESXi product, and the different way it is deployed,patched, monitored and managed, there's a bit of debate around about how to do all those things - sounds like a perfect opportunity for a proven practice!

If you want some fame and glory, and you've got ESXi running in your environment, even if just a POC, if you want to share the way you deployed it, how you patch it, how you monitor it, and how you configure it - there's definitely more than one way to do it, but there's a lot of hunger out there to see how everyone's doing it... who knows, if we see a few proven practices we might end up with a consistent pattern and HEY PRESTO! A best practice is formed from this very community... wouldn't that be great...

Who wants to start with a basic description of what they do? We can add some diagrams, why not even a video - we can embed video on here too, you know... and in documents...

The possibilities are endless...nearly...

5 VDI anti-patterns

schambers — Sun, 16 Aug 2009 21:15:12 GMT

VDI is fantastic and getting better, and the complexity lies in integrating the new technology with the old desktop practices.

Have you come across these five anti-patterns in VDI?

1. Liberal use of the Somebody Else’s Problem (SEP) field.

2. Indifferent business case

3. VDI changes the desktop lifecycle

4. Virtual Desktops are NOT Virtual Servers.

5. Nobody owns VDI.

Splunk

schambers — Tue, 25 Nov 2008 14:40:33 GMT

Anybody use Splunk with their VMware Infrastructure?

Announcing Splunk for VMware

VIOPS Capacity Management Webinar with Metron

schambers — Tue, 10 Feb 2009 17:36:43 GMT

Thank you for your interest in the VIOPS Capacity Management Webinar with Metron. This event has now been scheduled.

Wednesday, March 25, 2009 8:00 am PDT / 3:00pm GMT (Pacific Daylight Time is GMT -07:00, San Francisco) Register now
About the Webinar The target audience is intermediate experience/skill for VI, and we will cover the following topics: Key Metrics for VI3 Increasing Efficiency Intelligent Reporting Intelligent Trending / Forecasting Intelligent Modelling Your 7-step Capacity Management Plan
VIOPS Documents by Metron Proven Practice - Resource Pool Capacity Management with Metron Athene Proven Practice: Capacity Management Reporting for VMware Clusters with Metron Athene Proven Practice - Creating a VMware Capacity Management Dashboard with Metron Athene Proven Practice: Implementing ITIL v3 Capacity Management in a VMware environment Proven Practice: Using Athene to recycle VM’s
About Metron Metron is a privately owned limited company which was founded in 1986. Metron-Athene Inc is a wholly owned subsidiary of Metron technology Ltd. The company is Europe's foremost Capacity Planning and Systems Performance Management specialist. Metron's flagship product, Athene, provides fully integrated ITIL-compliant capacity management, automatic performance analysis and reporting for UNIX, Linux, Windows and Mainframe Servers . Find out more about Metron

Wednesday, March 25, 2009 8:00 am PDT / 3:00pm GMT

(Pacific Daylight Time is GMT -07:00, San Francisco)

About the Webinar

The target audience is intermediate experience/skill for VI, and we will cover the following topics:

Key Metrics for VI3
Increasing Efficiency
Intelligent Reporting
Intelligent Trending / Forecasting
Intelligent Modelling
Your 7-step Capacity Management Plan

VIOPS Documents by Metron

About Metron

Metron is a privately owned limited company which was founded in 1986. Metron-Athene Inc is a wholly owned subsidiary of Metron technology Ltd. The company is Europe's foremost Capacity Planning and Systems Performance Management specialist. Metron's flagship product, Athene, provides fully integrated ITIL-compliant capacity management, automatic performance analysis and reporting for UNIX, Linux, Windows and Mainframe Servers .

Find out more about Metron

SAP Sizing on VMware

deepchand — Sun, 25 Jan 2009 18:25:08 GMT

One of our Customer is planning to deploy a SAP in their environment. Our Hardware Vendor has proposed a following hardware configuration based on an inputs provided to them through a SAP vendor.

BI:

Production: DL580 G5 (QC)(E7340) with 2 Quadcore CPU´s (Xeon MP 2400/2x4M) and 32 GB Memory

Test: DL380 G5 (QC) ( E5420) with 2 Quadcore CPU´s (Xeon DP 2500/2x6M) and 16 GB Memory

QAS: DL380 G5 (QC) ( E5420) with 2 Quadcore CPU´s (Xeon DP 2500/2x6M) and 16 GB Memory

EP:

Production: DL380 G5 (QC) ( E5420) with 2 Quadcore CPU´s (Xeon DP 2500/2x6M) and 16 GB Memory

Test: DL380 G5 (QC) ( E5420) with 2 Quadcore CPU´s (Xeon DP 2500/2x6M) and 16 GB Memory

QAS: DL380 G5 (QC) ( E5420) with 2 Quadcore CPU´s (Xeon DP 2500/2x6M) and 16 GB Memory

ERP:

Production:

Databaseserver: DL580 G5 (QC)(E7340) with 4 Quadcore CPU´s (Xeon MP 2400/2x4M) and 64 GB Memory

Applicationserver: DL580 G5 (QC)(E7340) with 4 Quadcore CPU´s (Xeon MP 2400/2x4M) and 64 GB Memory

Test: DL380 G5 (QC) ( E5420) with 2 Quadcore CPU´s (Xeon DP 2500/2x6M) and 16 GB Memory

QAS: DL380 G5 (QC) ( E5420) with 2 Quadcore CPU´s (Xeon DP 2500/2x6M) and 16 GB Memory

PI:

There we have very low requirements. So it would be possible to install PI on the EP system.

If the customer wants to have it separately you can use the same configuration as for EP.

Disc space:

BI: 6 TB (only Cube data - for ODS no information is available)

ERP: 14 TB

EP and PI: 400 GB

If we look into this sizingI feel that it is really tuff to sizing it on VMware infrastructure. looking forward for suggestions.

What should an application proven practice contain?

schambers — Wed, 10 Sep 2008 15:56:58 GMT

Following on from Re: What other applications do we need proven practices for?

As this site is focused on operational aspects of virtualization, what do you consider as essential content for an application proven practice? Here's a starter...

Virtual Infrastructure configuration - storage, network, memory, CPU... statements of how the underlying virtualization was configured, up to the virtual machine layer?
Sizing, capacity, performance considerations and examples.... I'm assuming something like "1 vCPU, 1GB RAM = 20,000 dbbench transactions" or something?

Are these good for a start? What else? Is someone out there sitting on a proven practice already.... :)

Oracle Financials, Linux, and VI

pclapp — Wed, 10 Sep 2008 16:35:14 GMT

I have a customer who is "exploring the possibility of using VMware in the Oracle Financials environments for Linux". Can someone point me to docs, practices, communities, etc that I can share?

Thank you!

R/ Porter

How to clear events in the virtual center database by using VI client?

VCplayer — Wed, 17 Sep 2008 04:08:54 GMT

Is anyone knows that is it possible to clear the events of virtual center by using VI client?

Otherwise, I think I should clear the database manually.

The size of database of my virtual center is 20G now,

I want to compact the database, how to?

Virtualisation and the Myth of "Standardization"

bmak — Wed, 03 Sep 2008 07:42:33 GMT

Greetings,

When we decided to adopt VI3 (about 2.5 Years ago) we did so because we believed the hype and liked the idea of being able to standardize around a hardware platform (in this case HP/AMD) that would be (somewhat) future proofed.

We purchased a number of HP Proliant DL385's for deployment across multiple sites and watched as our physical environment was recreated in the virtual and (the physical) was eventually decommissioned.

Over time our Hardware requirement grew as our virtual environment grew beyond acceptable performance limits based on the original hardware spec.

I started looking at adding additional hardware ESX servers in order to increase our capacity and discovered that our original DL 385's had gone end-of-life and the next "G" version was the only available option. We purchased and brought the new hardware online.

Very soon thereafter I discovered that we were unable to VMotion between the old servers and the new. There had also been a change in processor voltage that ESX did not like and would therefore not validate for VMotion (we were able to use Converter to move the virtual machine when offline).

This has now happened several times over the past 2.5 years and we are now in a situation where VMotion (a very strong driver in going with the ESX solution) will not talk between various ESX servers in our environment unless the processors match exactly.

When we took on ESX I was new to ESX but not to other VMware products and could not therefore plan ahead to avoid this situation. Also, there is a case for suggesting that we could have just purchased a batch of servers at the same time and thereby avoid any future problems with end-of-life incompatabilities. But we are a small company with a limited budget and my "power's that be" (I'm sure I am not alone in this) would have a problem with buying , for example, ten servers when we only need six.

I know that this has been discussed elsewhere and I know that there are very in-depth solutions for monkeying with ESX in order to get it to ignore processor differences (along with all the dire warnings about hell decending upon us if this goes wrong), but I wanted to at least post this here in the hope that my experience would act as a cautionary note to any considering ESX from a complete "newbie" perspective and in order to give an advanced "Head's Up" while you are planning any new ESX ebnvironment.

Standardization in the ESX environment is very much dependant on how much cash you have to spend.

BMAK

VIOPS Special Interest Groups (vSIG)?

schambers — Mon, 16 Feb 2009 18:45:02 GMT

Do you think we should have vSIGs that "manage" their own community on VIOPS? Vote now!

We currently have five communities - Strategy, Applications, Security, Management and Availability - which were chosen at the launch of VIOPS but now need some revision (e.g. Management is too high-level, and there are multiple topics in it)... so while we are planning a change, why not think bigger...

There is a growing demand from users to find expertise quickly and their view, expressed to us at VMware, is that forums are too hard to search/find information and that there are so many blogs out there they have similar problem finding information but more importantly, difficulty trusting that information.

On the other hand, there is a growing band of fantastic experts on virtualization that are looking for ways to share their expertise with other virtualization folks... so what's the answer?

Inside VMware we have Specialist Interest Groups (SIGs) who focus on things like Security, Availability etc.

Is it time to bring all of these together and have vSIGs on VIOPS?

I started a simple proposal document off and will be reaching out for the advice and guidance of the community.

Proposal: VIOPS Specialist Interest Groups (vSIGs)

Please let me know what you think!

Steve

Anybody using SMVI (Netapp SnapManager for Virtual Infrastructure)?

jlurye — Fri, 15 May 2009 19:06:22 GMT

Anybody using SMVI (Netapp SnapManager for Virtual Infrastructure)? We are using Netapp storage and Netbackup agents in the VM for backup. I'm looking for faster recovery times for Exchange VMs and other critical VMs. Is a backup proxy like VCB still necessary?

Thanks,

Jerry

Using NFS as a data store.

dchulhan — Wed, 03 Sep 2008 12:28:21 GMT

Can we have some discussion re:Using NFS as a datastore in terms of performance, best practice etc?

More VM's VS Less VM's

khunter — Fri, 31 Jul 2009 20:31:04 GMT

Having an on going arguement with App Devs re: putting their database driven websites into VM's. My questions/concerns are as follows.

Enviroment ESX Cluster 100 GHZ 200 Gigs Ram. 5 hosts. NFS Connectivity trunked 4gig link.

Our physical env suffered from major sprawl, and instead of just P2V'ing all the garbage into our fresh new cluster was considering redploying webapps and attempting to consolidate as much as possible.

One of our apps in the physical world consists of 3 boxes, One webserver, One SQL, and one Buisness Logic Layer box. Which then needs to be replicated x 3 for a full Test, Dev, and Production setup. So 9 physical servers for one app. Obviously the above multiplied over 60 apps = nightmare.

I have a dev arguing to continue to build the app as above with 3 VM's one each for Web, Sql, and Bll. My arguement is *WHY* bother with seperate vm's, why not put all 3 functions inside of 1 vm and throw more resources at said VM, since it all derives from the same pool. Aside from isolation (which is not my concern) is one method preferred vs the other ? Forget shares / DRS every VM has the same contention aspects. 3 vm's vs 1 vm ?

Let the fighting begin.

DR environment - Physical to Virtual

deepchand — Fri, 12 Sep 2008 08:18:53 GMT

Dear all,

We are working on a project where customer is looking forward to keep their primary datacenter on physical servers and DR site on a Virtual Infrastructure.

Few of requirements are:

- Real time replication of physical environment to virtual environment.

- Primary site they have HP EVA at primary site as well their are ready to go with HP Eva.

- They required full server replication including OS, Data etc to Virtual environment.

- Even they have few machines like Linux with Oracle & Windows NT

looking forward for a recommendation & Suggestions

Regards,

VSphere on DELL PowerEdge 1850R

BlackMamba — Thu, 12 Nov 2009 02:51:26 GMT

Guys,

I have 2 DELL PowerEdge 1850R that I would like to deploy in production. THE CPU is supportedly a 64 bit CPU and got VSphere running successfully.

However, I am unable to install any 64bit O/S. I am kinda stumped. If VSphere sees the CPU as 64bit, why the heck won't Microsoft O/S see the CPU as 64 bit. Is there a setting I am missing? Please advise.

Cheers

martin

Virtual Domain Controllers or Secondary DC

RNF0528 — Wed, 10 Sep 2008 13:24:50 GMT

Does anyone have a a virtual windows domain controller in a test or live envioroment? If not, does anyone have any thoughts on this?

RNF0528

Mirroring two ESX Servers with direct attached arrays

kanoute696 — Mon, 28 Sep 2009 14:26:08 GMT

Hi, am new to the forum so please forgive if I commit any faux pas

We have two HP DL585 servers (amd opteron) with Direct attached storage arrays (attached via scsi cable) We would like to create a VM environment using ESX 3.5 (this is whats available to us - packaged) We also have all the vmware tools available to us but no other budget to spend on NAS\SAN's

Is there a way to create a mirror between the two machines (and arrays so to speak) so that failover is possible if one goes down or something similar - I think HP storageworks storage mirroring does this but does VMware have anything build in to do the job?

we will be installing the ESX build on the local hard disks(mirrored) and the VM slices on the array (raid 5). Also the arrays are seen as another local disk on the server. Would Vmotion help or what could we put in place here?. Thanks

Alternatively, what is a good design to follow with the hardware we have? I'm finding it hard to get examples of VMware environments on the web. Thanks in advance

ESX Maintenance Standard Operating Procedures (SOPs)

schambers — Thu, 19 Mar 2009 17:39:36 GMT

Do you use SOPs for ESX maintenance? I'm currently compiling some for VIOPS taken from my own experience - I think I've spotted a pattern, you see...

Also, I love the work of the IT Process Institute, Gene Kim (CTO Tripwire) and Kevin Behr (Assemblage Pointe) and others on bringing science to IT... ever heard of the Toyota Production System? Ever heard of an IT version? But I digress (if your interest is piqued - private message me!).

Back to the SOP - do you have one for ESX maintenance? My work so far has found five different triggers:

Swap - e.g. replace failed drive
Patch - e.g. security patch
Upgrade - e.g. minor release like 3.5 U2
Configure - e.g. lock down SSH so you can only access via bastion/jump box
Observation - e.g. to run a vmsupport script

So now I'm thinking that ESX Maintenance isn't an SOP, but a collection of SOPs - and the real SOPs are those five above.

But what does an SOP look like? The most basic example is a list of steps, but that's not enough for me. I want something that brings science to IT and allows for continuous improvement. I want the procedure to be followed rigorously, consistently - and where there is deviation it is because either (a) the process can be improved (improve the process), or (b) the person needs to be improved (improve the person - training).

So for my gold standard of SOP I've thought of the following attributes - and each SOP should fit on one piece of A4 paper

Trigger / Driver - what starts this whole thing, possibly a connection to a preceding SOP?
Input - what information is required to do this job?
RACI - for this procedure, who is Responsible (who does it), Accountable (who supervises?), Consulted (who helps), Informed (who is notified)
Resources - what tools and suchlike are needed for the actions (e.g. the list of changes to make to /etc/ssh/sshd_config
Actions - the unambiguous action to do, down to the most intricate detail - do not summarize, do not leave open to interpretation.
Timing - for each action, like the marks on the Toyota production line that shows in 32s you should be at step 3.
Output - A measurable outcome

So I'm desperately seeking SOPs and people who are interested in this stuff - I believe it's the key to increasing throughput and reliability of change, as do many others... do you?

New Graphics and Icons from VMware - anything missing?

schambers — Wed, 11 Feb 2009 20:35:08 GMT

Library VMware Icons and Diagrams

Anything missing from this good looking set?

Do you have any other graphics and icons you use - want to share that resource with the VIOPS community?

VIMA

schambers — Thu, 27 Nov 2008 00:04:06 GMT

Are you using it? How? We need practices on VIOPS around VIMA...

My colleague and epic blogger Duncan Epping has just written up some good stuff on VIMA... now we need a doc to capture some specific use cases / practices.

If you have a VIMA instance, let me know and we can collaborate.

There's a VMworld EMEA ticket up for grabs for Outstanding Contributors to VIOPS - could this be you?

Cheers

Steve

VSphere on DELL PowerEdge 1850R

BlackMamba — Thu, 12 Nov 2009 02:47:05 GMT

Guys,

I have 2 DELL PowerEdge 1850R that I would like to deploy in production. THE CPU is supportedly a 64 bit CPU and got VSphere running successfully.

Cheers

martin

Virtualization challenge #1

woroxon — Mon, 28 Jul 2008 17:50:07 GMT

After reading the practice of training i thought that a document that focus mostly in the vagarities of the new infrastructuresuch as mobility and flexibility.

case in point where i work we have implemented a Virtual center enviroment and one of the challenges that presented was that some IT's din't separated the VM from the host. That was especially dificult in a project were the VM need "float" between 2 especific servers that had the exact same infrastucture supporting them. Even after some explaining of the benefits of the configurations for the business continuity from both the great vmware sales pitch and our own vm implementer we get some question once in a while why the host need to be configured equal?.

What licensing options people buy for each piece of the puzzle when virtualizing.

schambers — Wed, 03 Sep 2008 13:15:21 GMT

From dculhan

what licensing options people buy for each piece of the puzzle when virtualizing. OS, App, DB, Middleware, Backup software, etc.?

Cluster in DMZ?

Harpstein — Thu, 29 Jan 2009 22:24:03 GMT

We're looking to create a complete VM cluster physically connected in a DMZ. Has anyone done this and has a proven practice for it? Even some things to watch out for or potention "gotcha's"?

Cheers,

Tim

Converter error

kanoute696 — Thu, 29 Oct 2009 16:42:28 GMT

Hi Kelly,

I've put this on a different thread but said I would ask you as you have been of such help in the past

Converter errors

Im running vmware converter standalone, ran on a W2k3 server (importing an online physical server,i.e itself ) . I am trying to P2V itself and send it to out to our ESX3.5 server\datastore (destination)

sorry I cant explain it any better than this.

Hi All, I'm getting this error

Unknown error returned by Vmware convertor Agent

and in the logs its saying

2009-10-29 15:18:11.119 'ClientConnection' 5028 info Making sure that UFAD interface has version vmware-converter-4.0.2

2009-10-29 15:18:11.135 'ClientConnection' 5028 info UFAD interface version is vmware-converter-4.0.2

2009-10-29 15:18:11.150 'P2V' 5028 info task,277 Task execution completed

2009-10-29 15:18:11.181 'P2V' 5028 info ufaSession,129 DoImport called

2009-10-29 15:18:11.181 'P2V' 4256 info task,275 Starting execution of a Task

2009-10-29 15:18:11.181 'P2V' 4256 info ufaTask,207 Successfully connected to VMImporter

2009-10-29 15:19:20.149 'P2V' 4256 error task,296 Task failed: P2VError UNKNOWN_METHOD_FAULT(sysimage.fault.ImageProcessingTaskFault)

Is there no way I can P2V it to itself and then copy it to the datastore

(when i try this, it just doesnt see a file the file even though its there)

What is the best P2V tool out there?..

Thanks for any help provided

Blades vs. traditional Rack servers

german — Mon, 08 Sep 2008 18:51:23 GMT

We are in the design stage of choosing hardware for our enviroment. I need some direction in regards to the footprint of server to use. We are evaluating the blade solution from HP (C3000) and the traditional rack server (DL380). Any input?

Experiences with Data Deduplication

rmalenchek — Wed, 03 Sep 2008 16:49:45 GMT

All,

I am currently reviewing data de-dupe solutions from EMC, Avamar and Data Domain. Has anyone here done any further research in this realm?

Any information / insight / advice would be greatly appreciated.

Thanks,

Russ

Smart Array E500 with msa 60 sata disks

OrcunUsturali — Wed, 03 Sep 2008 09:05:52 GMT

Hi to all out there ,

we are currently in a vmware project which is using completeley Hp products.What we have is

1-Hp dl 380 with local p500 controller with 3 sas disks configured Raid5

2-Hp msa 60 with 6*750 sata disk drives

3-The msa 60 is connected to the Dl 380 with Hp smart array e500 card.

We can successfully install wmare 3.5 to the server locally,Then when we try to format the MSA 60 storage (we previously created the logical drive with raid5) it sees no target from the vi client console.

İ can see the storage adapter in VI client ,it says the model ( smart array e 500 ) ,but targets : 0

Then i try to see the documentation about the smart array e500 card ,it is currently supported by wmware according to the latest compability list.

Any help will be great ,thanks a lot

Is vmware suitable for my organisation?

shivanjali — Fri, 29 Aug 2008 16:28:22 GMT

Hi guys
I request some suggestion on pc system setup in my office

we currently have a setup of 5 pc's connected by a single server over a ethernet switch

but soon business is expanding and i need a better IT infrastructure
with around 12 system and couple of laptops all will be connected ,

How can VMWARE be usefull to my organisation , and which software,

initially i was going to buy 3 servers, but after coming accross vmware i think i dont need that

i need some sort of security as well, so that my employees have only .limited access to the system,

1-I run Autocad on 3 systems

2-is Thin Client a good idea to implement
as i am looking for a secure sytem, where user access is limited to read only files

3-or a simple 12 desktop systems and network it the way it is ,

Suggestion would be helpfull

thanks

Best Practices for TS Windows 2008

jafernandez — Thu, 02 Apr 2009 18:18:08 GMT

Hello everyone:

I would like to get some information about virtualizing Terminal Sevices for Windows 2008. (Best practices, tricks and tips,…)

I’ve been looking for it, but I couldn’t find anything. There is some information about Citrix Presentation Server, but not TS.

Does any one have anything could be shared?

Thanks in advance.

Jesús

Has anyone figured out VCB 1.5 custom quiescing scripts?

dconvery — Fri, 08 Aug 2008 15:20:17 GMT

Hello -

I am trying to run custom quiescing scripts on a w2k3 VM. I am running ESX 35.U2 / VC 2.5U2 and VCB 1.5. According to the VM Backup Guide pg 52, the pre-freeze and post-thaw scripts are now located under C:\Program Files\VMware\VMware Tools\backupScripts.d.

For pre-freeze, it says "All scripts are invoked in ascending alphabetical order with freeze as the first argument." I created a pre-freeze-script.bat and a freeze_script_1.bat under that directory. Neither worked. Also, C:\windows\pre-freeze-script.bat does not work either...

Is there a file name convention for this? Any help would be appreciated...

Dave

Citrix On VMware

kboutrup — Thu, 16 Oct 2008 07:44:58 GMT

This thread is about running Citrix Presentation Server on VMware!

Running Citrix Presentation Server on (CPS) or Citrix XenApp (CXA) on VMWare ESX 3.x can definitely be done, and with great results to!

One important thing to be aware of is that there is no right solution or configuration that fits everyone. It all depends on your specific configurations (Host/VM), your resources (fx. Resource Pools), the app's you run in a CPS/CXA VM and lastly the behavior of you users (peek times, app usages etc.).

I have attached a document that may help you in the right direction. It's is not a "fact sheet" (!) but it can give you an idea of what other people have done, configuration wise, in optimizing their CPS/CXA environment when running on VMware ESX 3 and 3.5.

Cheers,

/Kenneth

vSRM VLAN configuration?

vikash_roy — Wed, 01 Apr 2009 10:14:15 GMT

I am planning to start POC for vSRM using ESX3.5 U2. I have done demo and it went through but could not show all the functionality. Currently for POC purpose I need to design VLAN.

My question what kind of approach should I take for VLAN. Shall I use "streached VLAN's" as few/NetApp says or shall I re-ip those machine?

If someone can point me to understand from pros/cons as well as "How To" that would be great help.

Thanks,

Vikash

VDI Justification

bckirsch — Thu, 26 Feb 2009 13:30:34 GMT

Good Morning,

I am a little new to the VIOPS so hopefully this won't be a repeat to anyone. Currently I am an IT Architect at a very large Children's Hospital located in Wisconsin. One of my current projects is writing a business case for VDI. Now I came from a fortune 500 company before this so writing white papers is not foreign to me but I am struggling with this one a bit.

Server virtualization is pretty easy to justify due to the high cost of servers but on a desktop side the hardware costs of a desktop are so cheap and the server side hardware and software licensing for VMware View are such a high initial capital cost it's a very tough sell in today's economy. Currently I am focusing on the power savings of the thin clients, however that savings is in a operational budget and not a IT budget. Of course it saves the entire organization money but saving money for another group is a hard sell sometimes.

If I focus too much on the labor savings to re-provision desktops I get the desktop teams scared they are going to lose their jobs (and I have stressed to them that they will get busier since there will be more thin clients because they are cheaper to put in).

Any thoughts, insight or suggestions would be helpful...additionally I would be happy to share the documents contents with the VIOPS site if it can benefit anyone else.

VM Test plan Req

bala_1182002 — Sun, 07 Dec 2008 08:39:20 GMT

Hi ,

I required Basic VMware Test plan cases . Please refer some important VMware test cases

Regards

Bala

Roll up, roll up! VI3.Blueprint project needs you!

schambers — Sat, 07 Feb 2009 20:25:24 GMT

Hello all, I have just started a project to write up a VI3 blueprint - building a VI3 solution in 180 days.

Check out the VI3.Blueprint Project
Go straight to the VI3.Blueprint Directory

I am seeking reviewers and co-authors to help over the next two months. This is due to complete at end of March. What's required:

Reviewer - a commitment to read and feedback on specific documents, as many or as few as you can fit into your schedule. Also set some of the quality criteria for each document. I have some rewards in mind if you want to contact me and find out.

Co-author - as it says on the tin,a commitment to write and see-through-to-publishing, as many or as few as you can fit into your schedule. Again I have some rewards in mind, so contact me to find out.

It goes without saying that not only will this be interesting work, it will also be high profile as your name gets known inside VMware and to everyone outside via the portal. We get about 70,000 hits a month so far and it is rising exponentially at the moment as we approach content critical mass. This is without promotion via the main VMTN channel which has over 600,000 users so you can imagine the pageviews this stuff will get.

If you are interested in either role, please send me a private message or email. If you have questions about the above, feel free to post them in this thread because if you're thinking it then someone else probably is too.

Cheers
Steve

Message was edited by: Steve Chambers, to point at new Directory + update hits (from 40,000 to 70,000!)

New site layout and imagery

Rodos — Fri, 20 Feb 2009 22:04:05 GMT

Congratulations to the VI:OPS team and the web people for getting the site updated.

The new layout looks great and I like the coloured background on the documents (more like VMTN).

Great to see this space of VMware land moving forward more every day!

Metrics for determining worthiness for virtualization

schambers — Wed, 03 Sep 2008 13:12:59 GMT

Steve -> Branched this from the "What apps do we need proven practices for"

What I'm looking for is not so much for a specific app, but rather a series of metrics I can run against physical servers so I can determine their worthiness for virtualization. We've run accross a number of apps that don't heavily use CPU, RAM, Disk, or Network that run notably slower in a VM vs physical servers. We then read about how context switching can kill virtual performance, but I can't find any metrics to evaluate this counter against. What performance monitors does everyone else look at, and what thresholds does everyone use to make a decision on?

Being able to reasonably predict the success of an application before we actually do it is important now that we have the low hanging fruit done. All it takes is one bad experience to sour an application owner or project manager on VMs and they will run home to what is known: Physical Hardware.

Idea for Proven Practice: Symantec eVault

Harpstein — Wed, 10 Sep 2008 16:49:13 GMT

So we're gearing up to implement Symantec's eVault solution. It is supposed to be coming online sometime before the end of the year. We currently have our Exchange 2003 running on our VI, so that is where our eVault is going to run as well. Once we have out MASSES of email archived, we are planning for the Exchange 2007 upgrade, also on our VI. I can create proven practices for these both, if there is enough interest, as we are implementing them. Thoughts?

Thanks,

Tim

HP Blades, Virtual Connect, 10Gb, Flex-10 and vSphere

julianwood — Fri, 21 Aug 2009 11:48:21 GMT

Is anyone working on a reference document for HP Blades and ESX particularly using HP Virtual Connect

We are working out how best to deploy HP Blades with Virtual Connect and 10GbE and would be great to get some opinions and hopefully get some common guidelines together for people to follow

Here are some of our requirements:

We use NetApp NFS for all storage.

We require multiple VLANs for our production networking.

These are some of the things I'ev been considering.

Should we just use a single 10GbE uplink per Flex-10 module.
Should we not stack chassis together.
Should we have shared uplinks or just dedicated uplinks?
Should we just pass through VLANs to ESX or actually set up the VLANs in VConnect?
Should we get the ESX hosts to do the failover or get VConnect?
Should we carve up our FlexNics or just using the single 10GbE?
Should we separate LAN / NAS physically or with VLANs or not at all?
Should we separate out Service Console and VMotion?
Should we use portgroups to direct traffic over particular uplinks but allow failover?

What Blades should we use?
The 490c can take more memory but doesn't have hot swappable disks. Does it matter? You cannot currently mirror the internal solid state disks.
The 460C has less memory but HSS disks, so is that not a single point of failure.
8GB Dimms are very expensive so we're thinking of going 4GB dimms and getting more blades which works out cheaper.
Do we run ESX or ESXi installable or ESXi Embedded?
Anyone running ESXi off the internal SD cards on a 490C?

Would love to see some reference architecture or at least get everyone's thinking for how to set up vSphere with Blades.

Are you interested in ITIL topics...?

schambers — Fri, 05 Sep 2008 15:38:30 GMT

Any ITIL people out there? I'm wondering how many folks we get on the site who are focused on ITIL v3, or specific topics like Capacity Management?

If you're not ITIL focused - is there someone in your org that is? Would they benefit from some proven practices that show how things change with virtualization? Why not send them a link to VIOPS and get them involved?

We already started a couple of these.. Proven Practice: Virtualization Configuration Management

We could add a lot more... but only if folks are interested...

Tweep?

schambers — Thu, 12 Feb 2009 10:43:15 GMT

If you use Twitter, then put your ID on here and I'll add you to the VIOPS Twitlist, and update your profile because it now has a new "@Twitter" field - check mine out!

How do you rollback ESX patches?

schambers — Tue, 18 Nov 2008 18:07:52 GMT

I guess the first part of the question is how do you apply the patches... do you use VUM? Or do you patch an image and then roll out a complete image...? Or do you execute remote commands to get/install the patch? Other ways?

Migrate Infrastructure client vm without vmotion

frank.mundt — Wed, 15 Oct 2008 08:23:07 GMT

I Have a vmware Infrastructure client that is a vm. I need to patch our ESX hosts witch the newest updates, so I need to migrate the vmware Infrastructure client to another ESX host. The problem is that I do not have vmotion. My quetion is: how do i migrate my vmware Infrastructure client vm without vmotion??

Hope you can help me.

Celerra refresh - will it change storage strategies?

schambers — Sun, 01 Feb 2009 10:56:59 GMT

http://www.theregister.co.uk/2009/01/23/emc_refreshing_celerra/

I'm hoping someone like Chad or Bernie might be able to help here - I was reading the thread above and got interested in file-level de-dupe versus block-level de-dupe... to do that, the Celerra needs to be familiar with the filesystem (e.g. VMFS) - so for file-level de-dupe to work with ESX then the Celerra must be able to understand VMFS2/3....

Any comment Chad/Bernie?

Cheers

Steve

Do you run VC in a VM?

schambers — Thu, 04 Sep 2008 15:46:26 GMT

I started a poll on this question: http://viops.vmware.com/home/poll.jspa?poll=1014

So far, it's evenly decided between Yay and Nay!

With current versions of ESX (3.5) I think it's safer to run VC in a VM, but you wouldn't have caught me doing that in version 2.0 back in 2004...

Is anyone out there running a management cluster? Perhaps consisting of two ESX Servers in a HA cluster and running virtual machines for:

VC + License server + update manager
VC DB
AD (inc. DNS etc)
Monitoring VM
Syslog VM

The VC server manages the two ESX Servers in it's own cluster, and also all of the "Application Clusters" out there, of which there were "islands" of between 6-12 hosts (each cluster was different, but the consistency within the cluster was tight - ie. same hw, same builds etc.)

It looked real simple and elegant to me, and with ESX 3.5, when the ESX server that is currently running VC dies, then so does VC but the other ESX server in the cluster will restart the VC VM automatically - it doesn't need VC for this.

Implementing Vmware Infrastructure 3.5

antony — Thu, 02 Oct 2008 09:03:25 GMT

Hello,

I need some advise of people with Vmware ESX deployment experience.

We would like to deploy Vmware Infrastructure 3.5 in our company by the end of this year.

I currently already have most of our enviroment virtualized, running Vmware server 1.0.4 on 2x Dell PE 2950 hosts with ubuntu-server as host OS.

We decided to make Vmware ESX 3.5 as a project for this year so I downloaded the eval license to play around with in our test lab.

After a few weeks of testing, I managed to setup a ESX lab envinronment consisting of 3 ESX hosts and 1 iSCSI san.

I created VM's, templates, mananged to config VMotion etc.

My question is, should I do the deployment of ESX in our production environment myself or is it still best to hire consultants for help?

How did you guys do it? With consultants or by yourself?

Also, what's exactly the pratical difference between ESXi and ESX? Does ESXi have all the features such as vmotion, HA, DRS, etc?

Thanks for your help!

Antony

what licensing options people buy for each piece of the puzzle when virtualizing

dchulhan — Wed, 03 Sep 2008 12:25:54 GMT

In Addition, I would like to know what licensing options people buy for each piece of the puzzle when virtualizing. OS, App, DB, Middleware, Backup software, etc.

SAN Migration and VMware

difenza — Fri, 05 Dec 2008 16:03:25 GMT

Hi,

next week i'll support storage guys of one our customer in the migration between 2 emc san, on the actual storage there are vmfs datastores and rdm.

I'll be there only to monitor the vmware infrastructure and I'll activate only if something goes wrong.

Migration will be accomplished using emc san copy. (personally I would have preferred to see both the storages and manually migrate the vms from a datastore to another but the customer has chosen a different way)

I've read this blog post http://gurudelleccelsopicco.wordpress.com/2008/07/01/san-migration-and-vmware-is sues/ regarding volume resignature, this will always occur?

I've also read a tsx ppt (Most Common Storage Issues & How to Troubleshoot them) talking about these problems.

Based on your experience is there any particular measure I can take to prevent these kind of problems?

Thanks in advance

NFS as a datastore for ESX server

amarnadh — Thu, 12 Feb 2009 17:22:57 GMT

Hi All,

Iam new to this forum. I am Amarnadh.

Well after seeing my subject in this forum you might have know that this was already discussed by all of you very long back. But in my case I want to have a clear idea about this.

In my work place we have 3 esx servers and a storage ofcourse we ran out of the storage as it was around 3.2TB. We have installed all SAP moudles on that. I had a request last day to day to create a new VM instance for a SAPSRM module where it has a hardware requirement of 4GB of memory and 300GB of Hard disk space. Well I did not have any space left over on my iSCSI EMC storage box. I made a scenario and presented to my manager that was I have a physical DELL 2950 server which has windows 2003 R2 and which is having a space around 500GB. I though of making this drive share through NFS service(in windows2003R2) which comes by default in Windows 2003 R2.

Please let me know can I share a windows drive through windows enabled NFS service and add this to my one of the esx server as a datastore? If this possible please let me know the steps like how to do it ? and what are the things to be configured on both windows server 2003 R2(physical server) and on my esx server.

Awaiting for response!

Regards,

Amarandh

Whitepaper ideas

daniel_uk — Sat, 28 Jun 2008 14:21:21 GMT

Steve,

Some material ideas which I would like to throw into the ring to see if you feel it has mileage;

Building a business case and ensuring solid ROI predicition - This goes into more indirect benefits such as remote site server consolidation and DR benefits
Tips for Implementing a successfull PoC - Highlights how to ensure you start on the right foot with your board and grab as much attention to detail as possible without blinding them with science for the first phase.
Virtual Governance - Based on setting policies and procedures for Virtualisation in an IT workplace
Service Oriented Infrastructure and Vmware - Explains details on how VMware enables "On demand" computing based on workload demand
DRS up the stack - Predictions on how the Beehive acquisition could possibly change VMwares reputation as just a Virtualisation platform company

Let me know if any of the above could be of use for VIOPS.

Kind regards

Dan

Red Hat Linux on ESX - current & best practices

dchulhan — Wed, 03 Sep 2008 12:33:43 GMT

How much of the community is involved in deploying RHEL on ESX and what are the experiences and best practices so far. What tips and techniques are there for creating templates, customizing, deployment, etc.

Deploy ESX 3.5 Update 3 via Kickstart Script

ron999 — Mon, 23 Mar 2009 16:16:50 GMT

Hello all together,

i hope anyone can help me. I have created a Kickstart Script but the whole %post section doesn't work. I have now idea why. I have searched through the www but found nothing that helps me in my problem.

Thanks a lot for any help.

Regards

ron999

ESX Boot Partitions

jbruelas — Wed, 15 Apr 2009 20:46:18 GMT

Little question (hope it is not silly)

what would be the recommended layout for ESX Boot Partitions?

(for local storage?)

(for boot from SAN?)

kind regards

Jose B Ruelas

P.S. if it is something "official " , I mean, validated by VMware, that would be nice

Life Cyle Manager

Harpstein — Wed, 03 Sep 2008 20:39:05 GMT

As expected, virtualization has caught on here like wildfire and now we are beginning to suffer the repercussions. Our System analysts have caught on to how quick it is to deploy a server and ask for them on a consitant basis. This is starting to cause a bad case of server sprawl where there are servers not even being used any more taking up valuable resources. LCM seemed to be the answer to this, so I went about getting it setup with a demo license but was never really able to put it to effective use. Has anyone tried it out and been successful with it? I really would like to get some sort of automated or semi-automated deployment solution in place to ease the burden of VM management.

Signature blocks in profile

virgilwashere — Wed, 10 Sep 2008 14:31:02 GMT

There's no way that I can see to save a signature in my profile. This would be nice.

Virgil

Virgil
Endless Consulting
Brisbane VMware User Group: http://tinyurl.com/VMUGBRIS

VDI - End User Hardware - What are people using?

svogel — Fri, 24 Oct 2008 18:44:31 GMT

Hi- Looking to expand my clients look at virtualization into the desktop world. What are VM pros using at the end user hard ware level to connect to the VMs in the data center. Right now all of our testing has been done with older workstations using RDP. We would like to consider Wyse or nComputing terminals but have no experience with those and VMs.

Thanks for any ideas or best practices at the client access layer.

Steve

vSphere in a box - Thoughts / suggestions.

darren@vmote.net — Wed, 20 May 2009 08:25:00 GMT

Scenario

With the GA of vSphere tomorrow (21/05/09) I'll be looking to install the latest and greatest and have a blank sheet of paper to make a hardware shopping list. I'd like to collect opinion of what community members are up to in terms of running vSphere in a 'sand box' environment without access to server, blade & SAN infrastructure.

If all goes according to plan I'd like to capture this process and publish it within the Strategy section in VIOPS referencing input from yourselves and (hopefully) with collaboration define a solid baseline document. I've long seen many threads and discussions regarding this facility for ESX3 for customer demonstrations, training etc... So let's start one for vSphere here.

Assumption

This is not an attempt to build a full working environment capable of hosting a myriad VMs, just 1 or 2 to demonstrate some / all vSphere features.

Questions / queries

How are you running your Beta / NFR small test environments?

Hardware

Purpose built and 'tuned' base build PCs running vSphere natively?
Purpose built and 'tuned' base build PCs running vSphere in VMware Workstation / VMware Server?
Mid-range laptops using VMware Workstation / VMware Server?
MacBook Pros using Fusion?
What are the specifications key to making it usable rather than just getting it installed?

CPUs, cores, RAM, disk, CD/DVD, Bus type, Chipsets
Model numbers, revisions, specifications

Have you found two machines are needed to really leverage the enviornment?

vSphere specifics

If you're running vSphere in a sand box environment what features have you found to be successful and what are troublesome?
What are the specific Virtual Hardware build revisions / VM tweaks needed to get started?
Are there any specific install or component configurations needed during the install or immediately after to get vSphere running?

Anything else?

Host Operating System on PC / Laptop / Mac

Hotfixes
Updates
Components (i.e. .NET)

Considerations for disk space?
Install media type

ISO
CD/DVD
ISO mounted using a software utility

Network connectivity specifically so this environment may work?

I think this enough to get the conversation started, now it's over to you.

Thanks,

Darren.

Message was edited by: Darren Woollard Amended the title as I didn't expect the use of an '&' to be annotated.

new design looks great

virgilwashere — Thu, 26 Feb 2009 09:20:25 GMT

Steve,

Like the new web design.

Capacity Management

SUNRAY — Wed, 08 Jul 2009 15:57:14 GMT

Hello,

My mandate is to go further in and seek questions that the technical team might have not think of. We have a person that will be working on the high level of this section.

I wanted to know if you have any information or documentations, on Capacity Management VMware infrastructure but on the technical side of it. Things that we need to look at globally. Physical Host, Vm's, SAN's, Lun, virtual Switchs, virtual Nic, physical host Nic, Clusters ,client demands, etc...

Questions I need to know and graphic or other good things to ask my technical support and gather their information to develop the Capacity Management Process.

I have found allot of things on the "Web" but I'm not sure of the information is want I need.

How and where do I start?

Avatars

Rodos — Fri, 20 Feb 2009 22:21:16 GMT

Is it possible to get more Avatars loaded? What about people loading their own.

I notice that the VMware logo Avatar is available for anyone to select, any way of restricting it, probably no big deal though.

Rodos

http://rodos.haywood.org

Need for "Proven Scripts"

dconvery — Thu, 11 Sep 2008 23:31:30 GMT

I just wanted to beat Steve to the punch :o)

How about "Proven Scripts" that are checked by someone other than the poster before being published? VCB scripts, Kickstart scripts, DRS?HA setup...Maybe somenice automated ESXi configuration scripts.

Steve mentioned that maybe a separate scripting section will be created.

WELL?????

Dave

Importing VMDK file created on ESX2.5 to ESX3i

myemmi — Fri, 20 Nov 2009 14:37:05 GMT

Hi Everyone,

First time in this community. I was just wondering if anyone tried or knows if we can import VMDK files created using ESX2.5 enterprise version into ESX3i (free version)?

Thanks You very much.

Mourad

Trusted SSL certificates - request for peer review

GuyChapman — Wed, 19 Nov 2008 16:27:01 GMT

As I guess most of us will know, the default in VMware is to use self-certified certificates for the web interface. Best practice would be to replace these with trusted certificates either from your Active Directory certificate store, a valid certificate from a trusted vendor, or whatever other corporate certification authority you use.

Fixing the hosts is pretty straightforward, there is a great guide on the forums, but here are the steps (which assume that you have installed and configured OpenSSL)

This is my process, please peer review and comment.

Prerequisites

You will require a copy of OpenSSL, the open source SSL key generator. This can be downloaded from http://www.openssl.org/ and the current version at time of writing is 0.9.8h. This is not the only tool available but it is the one on which VMware's documentation is based and it is supported by VMware.

The software must be installed and configured per the instructions on the website. It's not hard, and the documentation below also includes some help on this.

Background reading advised:

VMware documentation on SSL certificate generation: http://www.vmware.com/pdf/vi_vcserver_certificates.pdf
A thread in the VMware knowledge base: http://communities.vmware.com/thread/88053?tstart=0

Back up your certificates before you start. And back up your VCDB as well, just in case.

Creating the certificate request (CSR)

The CSR is created on the machine on which you installed OpenSSL. Note that if you use a Microsoft CA the key names and specifications differ between Microsoft's CA and OpenSSL. This is covered as necessary in the procedure.

Procedure:

Open a command window and change to the OpenSSL binary directory (default c:\OpenSSL\bin)
Check the settings in your local openssl config file. This is named openssl.cnf by default, you should copy it to openssl.cfg because Windows interprets .cnf as a SpeedDial shortcut; this is not fatal (you can still edit using notepad <path>\openssl.cnf) but it is a gotcha. It will save time if you add the basics to this file, such as the country code, email address and so on.
Create a new rui.key file as follows:
> openssl genrsa 1024 > rui.key
Generate a new certificate signing request at rui.csr as follows:
> openssl req -new -key rui.key > rui.csr -config openssl.cfg

The certificate request is now complete. You can open rui.csr in Notepad and copy the contents to the clipboard for the next stage.

If you want to place the files in another directory you can include the path (e.g. ../internal/rui.key) but you must use the Unix style forward slash not the Microsoft style backslash as the folder delimiter.

Microsoft specific information

If you use a Microsoft CA then you will probably need the following, whihc is taken from my MS CA:

Choose "Submit a certificate request by using a base-64-encoded CMC or PKCS #10 file, or submit a renewal request by using a base-64-encoded PKCS #7 file"
The certificate type is "Web Server (Exportable)"
When you download the certificate, make sure you choose the "base64 encoded" option (which is not the default)
I usually download botht he certificate and the certificate chain

Generating the rui.pfx file

Save the certificates generated above, to the machine with OpenSSL, to the same folder as the rui.key and rui.csr files. Change the certificate name to rui.crt. Encode the certificate file as follows:

1. From the OpenSSL binary directory, run
openssl pkcs12 -export -in rui.crt -inkey rui.key -name <common name of server> -out rui.pfx
Note that you can use a relative path to use files in a different directory, as previously.

Installing the certificates

The files may now be copied to the server (VC or VM host). The default directories are (assuming the drive letter matches your ocnfig):

For VirtualCenter: C:\Documents and Settings\All Users\Application Data\VMware\VMware VirtualCenter\SSL\
For ESX: etc/vmware/ssl/ - ensure you transfer in ASCII mode to convert Windows CR-LF terminators to Linux terminators, or the certificate will not work properly and you will be unable to attach to VC.
For VMware Server: C:\Documents and Settings\All Users\Application Data\VMware\VMware Server\SSL\

Remaining issues

Here's the biggie: when you replace the certificate on the VC server, you lose the virtual server connections. Experience to date indicates that if you reattach them, the VMs are fine, but I have only done this in the lab so have yet to run a structured test to check that all resource pools and folders remain intact with permissions unaffected.

Best practice for handling security incidents

schambers — Thu, 26 Jun 2008 16:31:32 GMT

Does anyone have a canned procedure for handling incidents?

I'm thinking of a list of triggers, some verification steps, standard responses - some kind of nice fluid incident handling procedure that my Ops and Security folks can use with the VMware staff?

Thanks

Steve

Multiple LAN subnets in one host

german — Mon, 08 Sep 2008 15:36:02 GMT

I am not sure if I am in the right group, but I consider this a startegy question in order to proceed in the implementation. We are new at ESX but have taken the training. We are implementing a solution in our datacenter to host clients networks via VMs. The issue comes in dividing the incoming traffic (Such as RDP and SMTP) to specific Nics inside the host. Keep in mind that there are many VMs with each one with at least one NIC. Any guidence will be appreciated.

Thanks,

German

How long did it take you to complete the automated vm deployment project?

schambers — Thu, 12 Jun 2008 14:48:25 GMT

Pete, referencing your proven practice (Automating VM Deployments) at Nationwide with Enigmatec, how long did it take you?

Thanks!
Steve

Support from Microsoft

macman — Fri, 05 Sep 2008 21:18:54 GMT

Several weeks ago, we had MS in house to do a AD Health Assessment. Very informative and worth every penny. During our discussions with the tech onsite, he mentioned Microsoft's policy on supporting OS's running in a VMWare environment. Here is the KB article that explains the MS position.

http://support.microsoft.com/kb/897615/

Of course upper management here got spooked on the FUD and started to seriously question our deployment of ESX. Has anyone ever had MS deny support because the OS was running on VM?

I would appreciate any feedback!

Proven practice needed: integrating CLARiiON iSCSI with ESX

GuyChapman — Wed, 19 Nov 2008 16:52:32 GMT

We have CLARiiON CX3 series SANs and all are comboi (i.e. include both iSCSI and fibre). Windows to iSCSI works fine but for labs and small scale tests we want to attach ESX hosts to the SAN as well. EMC's documentation is poor and PowerLink is not the world's easiest to navigate support site. There appears to be a missing step, the one which registers the ESX host with the storage processor. If anyone has a known working step by step to achieve this, I'd be grateful if they could share it.

Self-Provisioning Portal

mvaughn — Tue, 07 Jul 2009 23:53:00 GMT

Has anyone deployed a portal or web applicaiton to allow users to self-provision VMs within your environment? I really like Amazon's EC2 interface, and would like to investigate options to add a similar interface internally for our VMware environment.

network implementation best practice

pfuhli — Mon, 28 Jul 2008 15:28:55 GMT

Hi,

I can't find any detailed white paper where the network implementation of an ESX farm using beacon probing was described.

DOC-1031 goes not deep enough for the architectural things.

We are nearly at the point to have a solution but there are popping out new issues every week we discuss the scenarios. At this time we found out that 8 network ports per ESX host might not be enough for security and redundancy issues. I can't believe that.

Is there any document in deep detail handling the implementation e.g. of beacon probing one of you might suggest?

The thing I am thinking of is a kind of a standardized implementation scenario where ESX hosts are connected to two different physical switches and where you will rely on security constraints of separating the traffic of COS, VMotion, etc, productionVMs in different VLANs.

regards,

daniel

Removing Storage (LUN's) in ESX 3.5

Randy1963 — Wed, 03 Sep 2008 16:31:05 GMT

I am currently tasked with moving our vmware Datastores from an IBM 8100 and HP E5K SAN to our new EMC CX3-80 SAN, the LUN's are all fiber connected. All of the documentation I have found regarding Storage is how to Add storage. I need some good information on removing storage from ESX 3.5 using Virtual Infrastructure client 2.5.0. These are some of the issues that come to mind. I do plan on removing the LUN's after hours during downtime and thought I could find good information here regarding vmware ESX 3.5.

1). After migrating our vm's to the new LUNS and when removing the old LUNS from ESX 3.5 using Virtual Infrastructure client 2.5.0 are any vmware hosts taken offline or is it disruptive to the vmware host when a rescan is executed from the Storage pane of VI and the LUN's are removed?

2). Do the vmware Hosts need to be rebooted after removing LUNS or is a Rescan all that is needed?

3). Are there any recommended clean up steps that can be recommended or any known issues that have been found when testing the removal of LUNS from ESX 3.5 while the hosts and guests are running?

4). Is there a recommended process (e.g, remove the LUN's from ESX 3.5, then remove the zones from the fabric, then reclaim the storage from the SAN) relating to vmware ESX 3.5

Thank you in advance for any information that you can provide regarding these questions.

Randy

Site Recovery Manager

Harpstein — Fri, 05 Sep 2008 19:17:29 GMT

Ok...so one of our biggest hot topics here is BC/DR. We have just began construction on a secondary facility approximately 10 miles from our primary facility. This new facility has been designed with a "server room" to house our BC/DR solution. This solution is, as yet, undefined. We are strongly leaning to a VM infrastructure and possibly utilizing Site Recovery Manager. This is probably the foremost topic that I need to bring answers about back from VMWorld. Has anyone out there developed or implemented any type of solution using this product? Are there any other solutions for BC/DR that anyone has deployed for their VI?

Reliable VM Uptime

dcap — Fri, 06 Feb 2009 17:56:46 GMT

Is there any best practices on monitoring VM uptimes? From what I've read, uptime in vCenter isn't super reliable (http://www.boche.net/blog/index.php/2009/01/11/uptime-lost-during-vmotion/) and we are trying to figure out some best practices around how to montior uptime effectively.

BMC and BSM

daniel_uk — Thu, 27 Nov 2008 20:55:28 GMT

Hi All,

My organisation is embarking on implementing some of the core BMC product set, within the range that they offer includes Virtualisation management and automation tools in the Bladelogic for configuration need and Automation of tasks with Run Book automation etc.

Looking at how powerfull the technology is when integrating with the various VMware API's is anyone using this in there organisation to really cut process and reduce opex? For example I have seen that T-systems have taken the new vOrchestrator to automate a complete SAP environment, has anyone seen or done this?

Other usecase areas I have seen is I am also interested to know if anyone is leveraging such technology to automate increase of available virtual resource when an SLA is not being met based on hardware.

It seems that with vOrchestrator, CapacityIQ etc VMware is really moving towards building the complete business service managed datacenter after looking at VMworld 2008 material, i'd be interested to know how anyone has either sold or is trying to pitch this in your current organisations. ALso has anyone had issues when defining the important elixir to orchestration of business processes in order to automate such tasks?

Hopefully someone can contribute

Daniel

Windows 2008 Server in VM

Harpstein — Tue, 06 Jan 2009 21:41:09 GMT

We are going to start deploying our DC's and RDC's as well as Exchnage 2007 in our VDI. I have heard that 2008 doesn't work well with memory ballooning and over subscribing. Anyone have any best practices for installing 2008 server or anything to be wary of?

Thanks,

Tim

VI:OPS FAQ on the Homepage

fwegner — Thu, 31 Jul 2008 06:43:43 GMT

I have trouble understanding the exact difference between this VI:OPS site and our existing sources of information like community forums,technical papers, whitepapers, product documentation, email distribution lists, and various wikis. The VI:OPS FAQ document helps somehow, but it is hidden in the Pilot Project section. The FAQ should be visible on the home page. Actually, the mission statement would be sufficient. It could have a direct link to the FAQ.

The FAQ currently does not explain the added value of the VI:OPS portal sufficiently IMHO. Maybe an explicit example of a best practice would help. We should also mention the relationship to all the other sources of information we have. This VI:OPS site must not be isolated. Here is an example of I think of:

Emails/Forum postings >process-and-review> VI:OPS Best Practice >process-and-review> VMware Documentation

The review bit is importent. I assume all practices here will be reviewed (by VMware PSO and/or VMware engineering and/or VMware Support). There should be a flag indicating who has reviwed a piece of advice. Otherwise you end up with the same trouble which you have with all the information in the community forums: You never know if it will work, or if it will even harm your systems.

Emails/Forum postings are unstructured and random. And finding valuable pieces of information is a matter of luck sometimes. This must not happen at VI:OPS, because otherwise there is no point in having this site.So I assume every best practice is labelled as such and also reviewed. There must be a moderator who approves a best practice. Also, meta data is very important. I.e. which VMware products and product versions does this apply to, which third party software, which hardware (if applicable), etc.

Also, I guess we only need the VI:OPS site because the official VMware documentation does not (and most likely cannot) cover all use cases of the product usage. But some practices may be good to include into our official docs.

I repeat my most important point: Do not treat VI:OPS content like a community forum. If you do that I do not see any added value, because I can have this in the official VMware communities. (General comments like this post are well suited for a forum post, but not best practices I think.)

My 2c.

Proven Practice suggestion : Preparing for a Physical to Virtual migration (P2V) and succeeding - a guide from real world experience.

darren@vmote.net — Thu, 24 Jul 2008 12:47:14 GMT

(Paper in progress and evolving)

I propose publishing a paper covering the key factors in making a P2V project run as smooth as possible and complete in a timely manner. The reason? I consider there is a gap in SMB confidence in starting a consolidation exercise (in whatever guise) to a Virtual Infrastructure.

The document is not intended to be a step by step guide but to contain the following topics : Business Considerations, Business Requirements, Estate Review, Target Environment, During Migration, Communication and P2V Procedures & Tools.

This publication will be interspersed with real world examples of how a general text book approach does not always work with pointers on how to remove oneself from the most common misconceptions.

Two high pressure and successful migrations (a company merger and a Disaster Recovery / Business Continuity gap) will be the basis for the experience I will draw from.

Comments and thoughts are appreciated for :

The...

(1) Intended audience.

(2) The structure.

(3) The output style.

(4) Opinion for a demand for this paper.

Regards,

Darren.

How do you do VI3 Log Management?

schambers — Mon, 30 Mar 2009 20:06:10 GMT

It's a question I often hear: there are multiple components that need monitoring in VI3, and they all have multiple logs - how am I supposed to monitor them all (should I?) and what am I looking for?

There is a growing consensus that VIOPS could be the place to answer these questions - do you have more questions about Log Management? Perhaps you have some answers, such as How to centralize your VI3 logs?

I think there are some high-level topics here:

1. How do you manage logs, meaning which logs are "important" to you - and why - and how do you care, feed, water and harvest them (log rotation and centralization).

2. How do you monitor and analyze the logs - what messages do you look for, and how do you look for them?

If this topic interests you, then post why and what you want to get or want to give. Then we can get a collaborative effort together to help the community.

Bring it on!

Potential Proven Practice: Chosing the Rack Server Route

Harpstein — Wed, 10 Sep 2008 16:35:58 GMT

As requested by Steve, I was wondering if those out there would like to see a proven practice on what our primary cluster solution looks like from a hardware perspective. The document would include issues and concerns that were addressed by choosing rack mount servers as well as the SAN and network considerations. It would also describe the actual hardware that was chosen.

Input?

Thanks,

Tim

Ideas or feedback?

schambers — Tue, 29 Jul 2008 10:54:24 GMT

Got an idea to improve VI:OPS? Got some feedback?

We're all ears!

VMWare View Server and the VDM Client

netrangerx — Mon, 15 Dec 2008 14:14:12 GMT

I have setup a testbed environment using ESX Server 3.5, Vcenter 2, and Vmware View. I can connect to virtual machines using the VMWare View client either installed or through the VMWare View web interface. However, we have 2 IGEL PC to ThinClient cards that we are testing and they have the VDM Client installed on them. From what I can gather the VDM client is an older version than the View client. The IGEL card is Linux based. I am able to connect to the View server usding the VDM client, and even see the Pool I want to connect to. But when I try to connect to the pool and chek out a machine I get the following message...

The virtual machine unexpectedly disconnected

I have tried just about every possible network configuration to try and get this to work.

My question is should the older VMD client be able to connect to VM's through VMWare View? if so, what could possibly cause this message? Thanks.

Paul

Virtual competency center

daniel_uk — Wed, 03 Sep 2008 09:21:44 GMT

Hi all,

Is anyone able to provide realworld engineer to ESX host management ratios? I beleive its about 10 ESX hosts to every engineer at present but wondered if anyone can provide full examples?

Also any names of corporations would be good so I can use in a business case, it would be great as name dropping always makes people feel more at ease :) PM if you need to keep this discreet.

Also in fully outsourced IT environments does anyone currently see a trend towards the virtualisation operational teams being managed and sourced by internally rather than adopting outsourced managed service offerings? From observations of various solutions offered by outsourcers, when looking at a fully outsourced infrastructure cost including any virtual deployments it seems much better sense to insource your Vmware team and outsource other non value add areas of your infrastructure.

Introduction to Virtualization Security: A Primer for 3 year olds

howardcat — Mon, 28 Jul 2008 21:46:29 GMT

Congratulations to VI:OPS!

Hello to all VIOPs professionals out there...

This looks like a really exciting forum to open up and discuss some very important topics. It is great to see real recognition of the new challenges and issues we face today within virtual infrastructure.

Since we are all kinda getting to know each other, and coming up to speed on various issues, etc..., I thought it might be helpful to post a link to a short (4min 38 sec) tutorial. Whether you are new to virtualization, or "an old pro", I am sure you will find this informative and entertaining.

Feel free to use it as a training tool with your customers, as we all try to educate ourselves about the relevant issues.

"InformationWeek's John Foley speaks with Howard Fried, Executive Advisor for CatBird V-Security.
Fried talks about the CatBird V agent and gives an interesting demo of how the agent works."

http://www.youtube.com/watch?v=mBvjJFVel2k

On a more serious note, our CTO, Michael Berman, will be joining the conversation shortly, and is looking forward to sharing our lessons learned and best practice.
I am sure this is going to get very interesting, and we look forward to joining the conversation!

Best regards,

Howard

Howard Fried, howard@catbird.com
Executive Advisor
Catbird V-Security - catbird.com

Configuration Management of VM's and Hosts

dcap — Fri, 19 Sep 2008 14:53:10 GMT

I'd like to know how are folks doing configuration management for VM's today? Are you using traditional products like Bladelogic, looking at new vendors such as Fortisphere, or waiting to get more details on what VMware announced at the show this week?

Community Help wanted - How to Find ESX host's CPU Usage and Memory Usage

virtualbox — Mon, 13 Apr 2009 21:00:17 GMT

Often I have been asked this question that how do you check the ESX Host's CPU and Memory Usage. We can easily check this by clicking on the host on the Virtual Center and can see the CPU Usage. Also at the same screen I can see the Memory Usage as well.

For an example I assume I am having a box of 8 CPU each 3.2 GHz. So at the Summary screen I can see that CPU Usage as : 8303 MHz.

Now for memory I can see that I have a memory of 32GB and usage is 10GB.

So the real question here is I need to know these usage (CPU and Memory) from vm-support log files. I want to replay the log file and want to know what is the usage for these two resources.

Thanks in advance.

Rg,

Prasenjit S

debian support....it's time!

jhaaga — Tue, 19 May 2009 20:57:41 GMT

So what's the deal here? Ubuntu full support but no Debian support, dosen't this seem backwards? I've been waiting for full support since esx 2.5.

It is worth risking not getting offical support and going with a super stable linux destro, debian, or risking stability and going with ubuntu?

Software iSCSI initiator + SC + HA

difenza — Wed, 05 Nov 2008 17:28:33 GMT

Hi,

I'm implementing some esx infrastructure (some esx + virtual center + ha + vmotion + iscsi), in every of these i've this network configuration:

vswitch0 with 2 nic (active-active), configured to override the vswitch failover order (active-passive,passive-active) in the 2 portgroups (1 for sc and 1 vmkernel for vmotion, vlan tagging implemented to separate traffic and trunk configured on the cisco switches).

a second vswitch1 configured with 2 nic (active-passive) with 2 portgroups (1 vmkernel for sw-iscsi and 1 sc required by design) also here i've vlan tagging etc.

as you can see in the file /etc/opt/vmware/aam/FT_HOSTS i can find this:

10.10.10.6 server02
172.16.64.2 server02
172.16.64.3 server03
10.10.10.9 server03
10.10.10.5 server01
172.16.64.1 server01

where 172.16.x.x is the "lan" network and is the sc ip for every server, the vswif0, (also the def gw is in this network), and 10.10.10.x is the iscsi network (esx servers + san) and is the ip for the vmkernel portgroup on every server, the vswif1, (created for the iscsi sw _initiator) and also for the second sc (created to make the iscsi work).

what i can see is that the HA mechanism is trying to use every sc ip it can find (the "real one" and the "needed one" (created to make the iscsi work))

so there's a way to tell him to use only the first service console?

thanks

How are template customization specifications passed to a debian guest os?

d_chall — Mon, 08 Sep 2008 23:28:44 GMT

We are running VMWare Infrastructure Version 2.5.0 with VirtualCenter 2.5.0. I can create a debian 4.0 template no problem, but when I try to set the hostname, DNS servers and search domain in the customization specifications of the deploy template wizard they do not have any effect on the deployed vm. I have managed to install open-vm-tools with some success, though I get a notice that it has not been configured correctly. Thus far all attempts to reconfigure the vm-tools has failed and made the system unstable. If I can somehow access the values entered in the customization specs wizard from the guest os, I can easily write a script to put the settings how I want, but so far I can't tell how these values are passed to the guest os or stored. Please help!

virtualized Active Directory domain controllers - dealing with poor disk I/O

jreiter — Tue, 28 Oct 2008 22:21:45 GMT

My company has been going through a large virtualization project for many of our servers. It's been going well, but we've been running into some pretty bad problems regarding disk I/O performance on virtualized AD domain controllers. I was wondering how others have handled these disk I/O problems, what sort of hardware you're using, and if you're seeing these problems at all.

We are using HP blade chassis to host the ESX servers. These ESX hosts are mounting volumes on a NetApp SAN via ethernet. All virtual machine vmdk files are on the SAN, allowing for Vmotion. What we've run into is that when you promote a Windows Server to a domain controller, it automatically disables disk write caching in order to decrease the likelihood of AD corruption in the event of a system crash. Disabling write caching in Windows causes such an excess of disk I/O that our network infrastructure can't handle it well, and the virtualized domain controllers virtually grind to a halt. We've seen simple reboots take upwards of 30 minutes, and running a chkdsk will almost hang the machine completely.

As a test, we migrated one of the domain controllers to the local storage on a blade (rather than it being mounted on a SAN volume), and ran the vm from that blade. At that point, the DC ran fine. Migrating it back onto a SAN volume caused all the disk I/O problems to come back. Similarly, demoting a domain controller back to regular server causes the problems to go away. I can only assume this is due to disk write caching being re-enabled during AD demotion.

We've upgraded our switches to much higher end units, and trunked multiple ports together on the SAN side and HP blade chassis side (to increase available bandwidth), but that didn't fix the problem. It definitely sped up our other VMs, but not the domain controllers.

Has anyone else run into these issues, and if so, how did you solve them? What sort of hardware are you using to connect your ESX hosts to your SAN? Or do you just run your VMs on locally mounted storage?

-joe

RSS feeds

Rodos — Sat, 21 Feb 2009 21:28:09 GMT

The RSS feed on the new homepage goes to http://vpp-dev-1.vmware.com/home/community/feeds/allcontent?community=1 is this right? Is vpp-dev-1 a temporary site?

Is there a feed that provides all comments and documents for all sections?

Many are like me and use RSS to track or keep up to date on sites, like VMTN.

Ta, Rodos

Peer review

GuyChapman — Wed, 19 Nov 2008 16:29:04 GMT

Let's get a section up for peer review, as part of the process of publsihing proven practices

Managing vSphere w/ SDKs / API best practices @ Technology Exchange - Developer Day VMworld 2009

heyitspablo — Wed, 29 Jul 2009 18:14:33 GMT

Folks, just letting people know about upcoming event that might be of interest to architects building solutions on the vSphere platform. Sessions range from Best Practices when using our APIs to vCloud API overview. Best of all you get a FREE vSphere Standard license for development and testing.

For session information, abstracts, videos, registration visit: http://communities.vmware.com/community/developer/techexchange

Would ITIL proven practices be useful on VIOPS?

schambers — Thu, 11 Sep 2008 15:04:00 GMT

http://viops.vmware.com/home/poll.jspa?poll=1015

Please vote, and pass comments on this poll. In VMware we have built up quite a bit of knowledge over the years on the impact of virtualization on ITIL processes - it's not everyone's cup of tea, but ITIL seems to be more ubiquitous these days, especially in large or governmental organizations.

The results so far show 93% say yes, whilst the other 7% think this is more applicable to someone else in their organization - who are those people, how do we get them on VIOPS?

Steve

Capacity Management Webinar - Feedback and Questions

schambers — Wed, 25 Mar 2009 16:25:44 GMT

Thanks to Metron for condensing a lot of capacity management expertise into just over an hour - great effort, guys!

If you attended, please provide your feedback so we can keep the discussion going:

1) What did you LOVE - any key slides, key pieces of information?

2) What did you LOATHE - what could be left out next time.

3) What questions do you have about Capacity Management that the Metron team can answer?

If you feedback, let us know your job role as well as your opinion.

emulating server 2003

raver2005 — Fri, 05 Jun 2009 16:04:47 GMT

i have a dimension 3000 running windows xp, i am usng vmware and running two windows 2003 servers as one team. however i only physically have one network card, is there a way to emulate haveing a second network card in the operating system server 2003