VMware Communities : All Content - Lab Manager

Storage space used by VMs linked to a template

krdeepak75 — Fri, 20 Nov 2009 23:52:46 GMT

We find that the storage space for VMs linked to a template grows pretty fast when users do things like application installs, copy builds, source control updates etc. I understand that for such large writes we should avoid cloning.

Can we do following to limit the growth rate of the delta disk?

Suppose we create a template with base OS installed. Now a user creates a configuration off the base template and deploys the VM in Lab Manager. The user then adds a new virtual hard disk (this disk is not on the template), configures this disk (say to E: drive) and does all large write activity on E:. I understand that all of E: will be consumed from the shared storage pool all at once. But after that will the delta disks grow at the same rate as it would if E: was configured in the base template?

Is there a way to shutdown Lab Manager gracefully?

SgRddY — Fri, 20 Nov 2009 17:07:25 GMT

Thanks for the information, it if very useful. I know that we can use API to deploy and undeploy VM's from Lab Manager maybe we could adapt it to do this at the server level. Anyone?

Windows 2008 Unattend.xml File Location

samualcc — Thu, 19 Nov 2009 21:38:19 GMT

I know that VMware prefers that the pre and post config functionality be used to customize templates. However I find modifiying the sysprep.inf and unattend.xml files to be much more powerful and flexible when deploying templates. Changing the unattend info for Windows 2003, XP, etc is easy. However I cannot figure out where Lab Manager is pulling the unattend.xml file for Windows 2008 setup.

Now on my lab manager server, I have found the original unattend.xml file. It is located in e:\Program Files\VMware\VMware vCenter Lab Manager\Tools\CustomizeGuest\Windows\Sysprep. However what I cannot figure out is how to load that unattend.xml into Lab Manager once I modify it. I know about the Build Sysprep package option under settings-->guest customization. It work GREAT for 2003, etc but not for 2008.

Any help on this would be great. For those interested in how to properly modify the 2003, XP sysprep file I can post that as well.
Thanks.

How to import VMs from vCenter to LM4?

shrenik — Thu, 19 Nov 2009 19:33:38 GMT

Notice that when you import the VMs into Lab Manager, you do so from the VM Templates page and the imported VM becomes a template. Lab Manager does not guest customize the imported VM, i.e. the template is an exact clone of the vCenter VM. However, configuration VMs that you create out of this template will be guest customized (IP config, SID changes, etc.).

Sounds like you created configuration VMs that got guest customized and something related to the SID update caused them to no longer be able to access the domain. If you always deploy these configurations in fenced mode, then you may disable guest customization on the template - go to the template's properties page and check off guest customization. Doing that will disable guest customization of the configuration VMs that are created from this template.

VMs are removed from the domain

AWo — Thu, 19 Nov 2009 09:44:26 GMT

hansbertrams schrieb:
Can anybody withLabmanager tell me how they use it?
Do you import machines from production to LM (fenced) and does it work without problems? Or do you customize the imported machine, with a new sid or something else?

Better open a new thread for this to get a response!

AWo
VCP / VMware vEXPERT 2009
[:o]===[o:]
=Would you like to have this posting as a ringtone on your cell phone?=
=Send "Posting" to 911 for only $999999,99!=

How to set a static IP everytime I create a linked clone of a VM?

Inquisitive3 — Thu, 19 Nov 2009 00:02:37 GMT

Thank you both for your help here.

LM4 adding DV switch & Host spanning to existing environment

bgarner — Wed, 18 Nov 2009 18:51:30 GMT

Ok last night I took our Lab Manager 4 cluster down and switched over to using the host spaning networking. Here is the process I used...

• In LM switch to Global Org

• In LM System>Settings>LDAP> Detach LDAP

• In LM Resources>hosts> undeploy all vm's on every host

• In VC remove nic's you use for guest vm's and your lab manager network from virtual switches freeing them up for use in a new vNetwork distributed switch. Take care not to completely disconnect any switches that have your LM or VC vm on them.

• In VC network inventory create a new vNetwork Distributed Switch to your datacenter object. Add the nics you freed up to the new dvswitch.

• In VC migrate you management vm's to the new dvswitch and then move the nic's they where using to the dvswitch. This is optional and really depends if your management vm's run in the same cluster as your lab manager vm's.

• In LM Manage>Resources>Host Spaning networks> click enable host spanning and then add your new dvswitch to the transport network list.

• In LM Manage>Resources>Hosts> enable host spaning on each host

• in LM Manage>Physical Networks>properties and change the binding to your new dvswitch

• In LM System> Organizations> edit each organization and change the default to allow configs to deploy across multiple hosts. This is also totally optional and depends on your sites needs..

• In LM do some testing as it should be working now...

• In LM System>Settings>LDAP> reconnect LDAP

We are now up and running with host spaning. I would keep your expectations low for this magically making HA/DRS/Vmotion work across all your hosts seamless. Our cluster has a few different but fairly closely related intel CPU's and it's still problematic. You can get into all kinds of trouble with saved states and revert points. One example would be a vm that has a revert point created on one host that gets moved to a host with a slightly different cpu type. This vm will fail to revert.

My other concern with this Is the service vm's that provide the host spanning transport network deployed into some resource pools that I didn't want them in. It's not clear how you get them to be created in a specified resource pool. Also you will need to disable host spaning on any host you want to put in maintenance mode.

Blake

Vmotion/HA/DRS and lab manager.

PezJunkie — Wed, 18 Nov 2009 16:12:52 GMT

Because of the way Lab Manager uses linked clones, storage vmotion won't work. You have to move entire trees together (There's a utility for that), but you can't do it while the machines are up and running.

HA/DRS works, but in ESX 3.5 any fenced config has to be kept together on the same host. (ESX 4/Lab Manager 4 can get around this if you configure distributed network switches.)

Why are you creating a new cluster? Does it not let you just enable HA/DRS on the existing cluster?

Best Practice for pointing LM to a new vCenter?

lrfraire — Wed, 18 Nov 2009 15:48:47 GMT

Any news

Lab manager website not working

kathirkk23 — Wed, 18 Nov 2009 09:56:01 GMT

Hi,

I have VC 2.5 with 5 esx 3.5 . I installed Lab manager recently and try to schedule the update but getting
"web site hosting the update signature and update packages cannot be reached" error

the web site for vmware update manager is not working even with direct internet.

https://www.vmware.com/PatchManagementSystem/patchmanagement

_http://pchmgmt-prod-wlc-vip.vmware.com:7004/patchMdSvc?WSDL

_https://xml.shavlik.com/unix/corefiles.xml
can any one help us to configure lab manager.

Thanks in advance
Regards
Kathir

Storage Quotas in LM 4.0

robert jensen — Wed, 18 Nov 2009 09:14:46 GMT

According to LM 4.0 release notes, its cable of handeling storage Quotes.

http://www.vmware.com/support/labmanager40/doc/releasenotes_labmanager40.html

I can however not find that option.

Can any of you point to where i find it ?

The only option i can find, is a VM quota.

What we are trying to do, is to have a couple of templates, on one LUN, taht other organisations can use.

Every organisation, have there own storage, and must not be able to create configurations, on out shared template LUN.

Storage qouta, would be a way to restrict organisations, from using more storage than they are allowed to, insted of giving them their own lun each.

If storage quotas not posible, can you then se any other way to acchive this ?

Regards

/Robert

How to back up lab manager and linked clones?

jjohnson0000 — Tue, 17 Nov 2009 20:31:04 GMT

I was wondering if anyone can explain to me how you go about backing up lab manager, as well as linked clones within lab manager?

My lab manager is running in a Windows 2003 VM. I know that you can't just simply back up the VM, because if you go to restore there will be issues with the database.

Also, most of the configurations in my lab manager are using "linked clones". How exactly do you go about backing all of those up? Since you have to undeploy the template you use to create the linked clone in, it does not show up in the ESX server. So if I were to back up the ESX server using VCB and such, it would back up the VM's that lab manager created, however it wouldn't back up the disk file (flat file) that the clones are looking to.

This creates an issue.

Even if I were able to backup the VM's, if I lost my lab manager server, it wouldn't do me any good because I wouldn't be able to restore the configurations the way they were before.

So how does VMware recommend you go about backing up Lab Manager?

Problem deploying after upgrading to Lab Manager 4.0

bgarner — Tue, 17 Nov 2009 19:19:12 GMT

I originally had some issues with all vm's in a specific resource pool not being able to be deployed. I ended up disabling and un preparing all hosts in the cluster. Then removing all resource pools in vc. Then re-adding the resource pools followed by preparing and enabling the hosts again.

Kinda painful process but it got us running smoothly again.

blake

Error when sending notification to "Everyone in Lab Manager"

sraisoni — Tue, 17 Nov 2009 00:07:11 GMT

Hi,

Try just clearing out the Email for system user this should resolve the issue for you.
Thanks,
-Siddharth

Missing authentication header.

cesarcesar — Mon, 16 Nov 2009 14:32:02 GMT

I also tried adding the code to your example and all i get is a white screen. The script fails at

$result = $client->__soapCall('GetCurrentOrganizationName', array());

I have attached an example perl script that works fine. Can you see in it why it works but the php version doesn't?

Thanks, Cesar

Track the number of Deployments over a period of time

IamTHEvilONE — Fri, 13 Nov 2009 13:48:31 GMT

To quote the movie SpaceBalls ... "now you see that evil will always triumph because good is dumb."
http://www.imdb.com/title/tt0094012/quotes

Regards,

EvilOne

Lab Manager on ESXi

Sanju84 — Fri, 13 Nov 2009 01:09:56 GMT

thanx that solved a few worries, how do i resolve the "valid NIC requirement" in setup prerequisites..

LM4.0 --> Where is the attach local CD/Floppy to VM??

bgarner — Thu, 12 Nov 2009 20:39:12 GMT

We have people install an iso mounter tool in their vm's. Users can connect to their local system via file sharing and then mount the .iso's. I only keep administration tools in the media share.

Lab Manager 3 reports - SQL Queries - looking for some help

paulrg — Thu, 12 Nov 2009 17:18:04 GMT

Does anyone know of a way (or have done this) to run period reports from SQL Server and dump them to a Web Server to display on a Web Page? .NET ASPX pages? Other?

Thanks
Paul

Windows 2008 R2 frezzing

jharris — Thu, 12 Nov 2009 16:47:39 GMT

I think this issue also occurs in Windows 7 64-bit as well. While excluding the SVGA driver from the VMware Tools install appars to be a fairly well known solution to some of these problems, I think there may be an additional issue perhaps related to the use of the E1000 NIC in VM's too as noted in the following post ESXi 3.5 + Windows 7 - Hangs.

I tried the suggestion of switching to vmxnet instead of E1000 yesterday, and so far it seems to have done the trick.

My configuration is vCenter 4.0, Lab Manager 4.0, ESX 3.5 U4.

What is the "namespace" in SOAPHeader

cesarcesar — Thu, 12 Nov 2009 09:13:17 GMT

thanks

how does SSmove move data

mrweetman — Thu, 12 Nov 2009 07:44:09 GMT

Great!! Thanks

-Marius

Lab Manager RDP Breaking Fence?

jmvirtual — Thu, 12 Nov 2009 06:02:13 GMT

Hello All,

I've deployed in a Fenced only configuration 4 configurations each having one copy of the same Virtual Machine. The machines have SQL Server installed and MSDTC. I can work on the VM's fine in console mode all day. However, As soon as I RDP to a VM it breaks the MSDTC components. I get Ident Errors, COM errors, but overall DB connectivity works fines. Its just some functions in the virtual machine no longer work because of MSDTC being out of service. I worked on the issue for weeks I finally got thinking it my might have something to with having the console session open and RDP at the sametime. Maybe some how the user is still logged on and when I create teh RDP session I'm creating another instance...

When MSDTC breaks I have to delete the clone and import it from the template again. Which um takes sometime. Anyways I ended up discovering that it only happens when I RDP into the VM. If I hadn't recorded it I wouldn't believe it either. Its like somehow my RDP connection allows the machines to see each other which will of course break MSDTC. I've used Sys Prep though its not recommended, I've tried changing the GUID, I've tried linked vs full clones, in short everything that I can think of...

Deploy Multiple IP Adresses on a single vnic

dotbrianw — Wed, 11 Nov 2009 20:17:18 GMT

Awesome. I appreciate the suggestion. I've advised the team to figure out another way around it. While I believe that being able to assign more than one ip to a vnic in LM would be a very useful feature, I don't think it's something they can't live without. From what I understand they just want them for apache configs. Name based virtual hosts will work just as well and I won't have to track ip usuage in a seperate place.

Again, I appreciate the suggestion.

Labmanager 4.0, error or normal behaviour?

jezi — Wed, 11 Nov 2009 13:37:33 GMT

You are welcome. It doesn't seem like a lot of people are using LM 4 yet. (At least the issues I am having aren't anywhere on the support forums, where it directly applies to LM4) I am happy to add any info I can add to the collective... We have a fairly large implementation that I am swimming in daily if you have other questions like this.

Have a good one!

Multiple LM instances, single pool of users

jschlach — Tue, 10 Nov 2009 22:43:36 GMT

Well, firstly if anybody is doing this.

If so, I'd like to know how their experience has been, what were the pitfalls and what advice they were giving someone who was doing it for the first time. I'm working on spinning up a new storage system and create a new Lab Manager install with new ESX hosts, eventually using it with our existing system. The use of VMs is from our automated testing system, which would hide the details of our dual-lm system from end users. As long as they clicked on LiveLinks they might not even be aware of it. The idea is, we could also still be productive in lieu of a major system outage (a LM instance, our a storage system). We could upgrade half at a time while running on the other half.

Lab Manager, vCenter, and API's

stanbinev — Tue, 10 Nov 2009 20:07:13 GMT

1. Lab Manager has its own Web UI console.

2. Lab Manager exposes a SOAP API. You should search the Lab Manager forum for more info on the SOAP API.

Consoles will not display on IE7 version 7.0.5730.13

bveis — Mon, 09 Nov 2009 18:43:10 GMT

First off, the console plugin will not work on Chrome.

Re: the problem accessing the console in IE7, can you please describe what happens when the user clicks on the thumbnails? It's possible that you may simply need to uninstall/reinstall the plugin for IE7 (let me know if you need help doing this). My guess is maybe a recent update from MS messed something up. Are they using a 64bit IE7?

Change Network Label on running VM

THAGDF — Mon, 09 Nov 2009 12:31:58 GMT

I have a problem with changing the assigned Network to a virtual NIC when the virtual machine is running. In the vCenter it is possible to change the Network-Label of a vNIC in a running state. But with the Lab Manager i had to shutdown the machine to change that network label. The Lab Manager disables the Network-Label if the VM is running so I can't change it although the vCenter Server is able to realize that.

Is there a work-around or something like that to handle this feature with the Lab Manager?

VMXNET 3 and LabManager

jasonboche — Sat, 07 Nov 2009 18:03:28 GMT

Kudos and props to you MLaskowski012.

http://www.boche.net/blog/index.php/2009/11/07/lab-manager-4-installation-fails-with-vsphere-vmxnet-3-nic/

Jason Boche, vExpert
boche.net - VMware Virtualization Evangelist
VMware Communities User Moderator
Minneapolis Area VMware User Group Leader
vCalendar Author

Port is Busy: All pipe instances are busy on Lab Manager

jschlach — Fri, 06 Nov 2009 23:58:09 GMT

We just saw this error message come back from an API call. Did you ever open an SR or find out what the cause was?

LiveLinks and configuration name length

jschlach — Fri, 06 Nov 2009 18:04:29 GMT

Using LM 3.0.2.56 --

We have very long configuration name lengths, and when one with a 64 character name is captured to the library and LiveLink generated, we find that the LiveLink is undeployable because it violates the length limit. I have a user who somehow has two configs whose names start with "LiveLink-" and are 65 characters each. I'm not sure how those came to be until I investigate further.

Has anybody seen this behavior before?

Initialization Wizard > Prepare hosts problem

VoipDude — Thu, 05 Nov 2009 16:19:42 GMT

I'm having this exact same problem, but my version is ESX 3.5.0 Update 2, not ESXi. And, as far as I have read, 3.5.0 Update 1 is the minimum requirement.

Capture to library has been going for over 18 hours

sraisoni — Thu, 05 Nov 2009 00:57:20 GMT

Did restarting resolve the issue and you did not see any more occurrences of it?
Were you able to get any clue why it stuck?

Backing up Lab Manager using esXpress, Veeam, vRanger (non VCB)

landslove — Wed, 04 Nov 2009 20:27:20 GMT

Does anyone suggest an answer to Scott's question. I would like to know too. I don't have access to a SAN for LUN snapshot backup methods. I need another procedure.

Regards, Nick

Empty Virtual Switches leftover in vCenter

pkelly08 — Wed, 04 Nov 2009 17:09:38 GMT

Resolved. DRS was turned on and set to Manual. During power on the Virtual Router was moved to a different host after the switch was created leaving the old one behind.

Cannot ping GW from VM's in fenced mode

jschlach — Wed, 04 Nov 2009 16:49:26 GMT

i've just come across this, but strangely in just one of my configurations. looking into it, i saw that my VM had a cached arp entry for its default gateway ip address which didn't match that of the virtual router. my solution was to deploy the config, suspend the VR, delete all arp table entries in the guest VM, then capture back to library. when that's then checked out, it has to build a fresh arp cache and picks up the correct mac address of its gateway.

Delete Library Configuration?

Muzika — Wed, 04 Nov 2009 16:30:49 GMT

Lab Manager 4.0

I have a configuration in the library that is in use by other users. I updated the original template that I initialy captured into the library and would like to replace the configuration currently being used in the library with the updated one. Can I delete the original configuration in the library without impacting the users that have checked a copy of it out?

Being hesitant about this I captured the updated template to the library but get duplicate IP errors after I deploy even when fenced. I'm assuming it's because both configurations in the library were captured from the same template.

Just looking for a way around this... or best practices when updating a library configuration.

PHP Wrapper for interacting with LabManager SOAP API

stanbinev — Tue, 03 Nov 2009 16:59:12 GMT

ITGO-TTS is your organization

Main is your workspace name.

username/password for the user you use to login to LM

You should pass those through the API. When performing template operations you do not need to pass the workspace (though it wouldn't matter). Only working in workspaces you do.

LM thinks a host is in maintenance mode...but it's not

pkelly08 — Mon, 02 Nov 2009 20:39:54 GMT

Well, that was painfully easy. That worked. Thank you for the help.

"Login failed for user VMwareLMWeb" after a database restore

Buddy S. — Mon, 02 Nov 2009 15:41:02 GMT

I found this post looks close to yours except it is Lab Manager 3

http://communities.vmware.com/message/1060988;jsessionid=CC9B9AED74123F8243D8E3E8694E2772

Adding RHEL5 VM in a fenced configuration.

nilesh_joshi — Mon, 02 Nov 2009 12:45:49 GMT

Hi,

I am trying to add a RHEL5 VM in a fenced configuration.

I have deployed a configuration in fenced mode with allow 'in and out'. That configuraation has 4 windows VMs. Now I am trying to add RHEL5 VM to that configuration. But some how after adding the machine I can not access it from outside fence. VMware tools are installed and VM has been forecefully customized while deployment.

Am I missing some step?

-Nilesh

LM 4 Guest Customization Question

landslove — Mon, 02 Nov 2009 09:08:17 GMT

Check the log at:

c:\windows\temp\customize-guest.log

Check the date stamp to determine if it ran as when you expected.

Upgrade Lab manager 3 to Lab Manager 4

eyalvaknin — Sat, 31 Oct 2009 17:57:30 GMT

Thanks alote
I will let you know when I will do that procedure
Eyal

Lab Manager connecting to vCenter Server

ilyab — Fri, 30 Oct 2009 22:13:12 GMT

I'm guessing this happened because:

You upgraded to Lab Manager 3.0 (and pointed the installation to a particular VC when you did so).

You then moved the host to a completely different vCenter Server (4.0), but didn't update the connection to this new VC within Lab Manager (this is the +guess+).

-The Upgrade Wizard in Lab Manager 4.0 does not give you the ability to select a new Virtual Center installation, so it tried to establish the connection to the original VC setup (which may no longer exist).

If you bring back the same Virtual Center server (2.5) used for Lab Manager 3.0, upgrade it to vCenter Server (4.0) and try to run the Lab Manager Upgrade Wizard again (if you can roll back), or just try to start LM4, it might find the server and establish a connection and allow you to log in. Otherwise, you might have to deal with Support to work through the database.

Removing trees and garbage collection in LM4

CH-Dave — Thu, 29 Oct 2009 23:43:38 GMT

Update: Found the cuplrit. There was an EXPIRED vm residing in the datastore section of Lab Manager. Once I hoevered over it and selected "Context" I knew I had the right VM because the Directory was "3250".

Slow performance after SAN move

Scissor — Thu, 29 Oct 2009 23:10:52 GMT

Did you ever figure out the cause of your slowness?

Customization seems to be failing on Ubuntu Linux

brucefulton — Thu, 29 Oct 2009 20:37:55 GMT

We're only using Ubuntu 8.04 and XP and the moment, and the hit or miss applies to templates and configs of both. On machines that don't deploy with networking working, nothing can be pinged, even the default gateway on the local subnet, and there are no entries in the ARP cache.

Delete Media

pwyzorski-wyzguy — Thu, 29 Oct 2009 16:48:41 GMT

Awesome! Glad I could help out.

Don't forget if the answers help, award points

Slow login via web console 4.0

jezi — Thu, 29 Oct 2009 13:25:36 GMT

I am using LDAP as well as local accounts.

This ended up being a roles issue. Once I elevated permissions on the accounts I didn't have the issue.

Curiosity Question - Why does Lab Manager 4 not support Nexus

pwyzorski-wyzguy — Wed, 28 Oct 2009 22:27:14 GMT

I believe this is because with the Nexus the virtual network configuration is managed on the Nexus side instead of through the vSphere API. Lab Manager actually requires more direct control of the networking configuration through both the vSphere API and ESX host interface.

Don't forget if the answers help, award points

Error in Lab Manager 4

jkrolak — Wed, 28 Oct 2009 17:41:43 GMT

Here's the text of the email I was sent:

Problem:
User has a configuration: "Sample
configuration" with bucket_id = '4'
Typically in a new installation it is ID of workspace
Main in Default org.
Since the user deleted that org and workspace, the bucket
ID was lost but the config still existed.
To fix the problem:
Run the following script
update servergroup set bucket_id = '2' where sg_id = '4'
This script will update the bucket for that configuration
to be Global/Private.
The error will be gone.

I had to install SQL Express Management Studio and run that line as a query in it. Make sure to select the database before running the query.

4.0 ESX Password Change in LM

jezi — Wed, 28 Oct 2009 02:42:51 GMT

Error was - Folder.registerVm task on RegisterVM_Task failed: The specified key, name, or identifier already exists.

Workaround - The only way I found to resolve this issue was to unregister and then reregister each VM. Your passwords get out of sync which caused LM to throw errors and for me half of my environment was unavailable in LM. The field to update the ESX password was grayed out no matter what I did. I ended up with dup host and. What a mess... My resource pool was pulled apart while some VMs were deployed and some weren't. I did run a repair on LM prior to the re-registering the VMs and it didn't change anything.

Same Error for me...Folder.registerVm task on RegisterVM_Task failed: The specified key, name, or identifier already exists.

I should have mentioned that I was able to re-register by redeploying in Lab Manager. This was for all the VMs that were deployed when they lost connection. In some cases I did have to browse the data store, find the vmx and add to inventory (vCenter). Then remove from inventory. Then deploy in Lab Manager...

vcenter 4.0 and lab manager 3.0?

_morpheus_ — Tue, 27 Oct 2009 18:09:09 GMT

LM 3.0 doesn't support vcenter 4.0

how to change specific user deployed VMS quota Lab manager V3??

clevi — Mon, 26 Oct 2009 14:57:31 GMT

found the problem.
when trying to change user quota of specific user it is prohibited to be on the global zone.
the problem is that i'm ADMIN! so of-course im on the global zone... and it was impossible to change it.
by chance i've moved to default zone and than i was able to change user quotas.

the idea behind that is to use global zone in roder to change the quota for all users in organizations.
in order to change for specifc user we need to be on the OU he belongs to.

Cannot deploy workspaces, cannot capture libraries

jezi — Sat, 24 Oct 2009 03:22:09 GMT

Jess

VM Deployment

_morpheus_ — Fri, 23 Oct 2009 16:01:50 GMT

The VM will get one internal IP, one external IP, and the virtual router will also get one internal and one external IP, so 4 IPs. The formula is 2n+2 where n is the number of NICs being fenced (or the number of VMs being fenced if each VM only has one NIC) and the +2 is for the virtual router.

Domain Member

pwyzorski-wyzguy — Fri, 23 Oct 2009 01:02:38 GMT

Windows AD does not work across a NAT.

Don't forget if the answers help, award points

The configuration was automatically undeployed

shrenik — Fri, 23 Oct 2009 00:53:44 GMT

What version of LM is this? Are you seeing this error in LM's UI/logs or in VC?

Lab Manager requirements

vmproteau — Thu, 22 Oct 2009 18:18:23 GMT

Thanks for the information. I do have a seperate internal Lab environment with isolated VCenter VM, I may use that.

Let battle commence - Lab Manager on VMs

efulton — Wed, 21 Oct 2009 20:09:00 GMT

Both VC and LM are running a MS SQL database, and both are attempting to gain control of ports 80 and 443, so they won't coexist within the same guest. Similarly, neither will live on an AD server because of the same sorts of resource contention.

However, the AD controller is optional, you can get most of the same functionality creating local users within lab manager.

Otherwise you're pretty close to a minimal setup.

Two esx servers would be minimally recommended, though you can get away with one. For demo purposes, the raw cpu power of the boxes doesn't matter all that much, though you will want to get as much memory into them as possible.

The easiest and lowest cost way to get shared storage is to have a physical linux box sharing it's disks via NFS.

multiple drives in template not appearing in deployed configuration

jvanqua — Wed, 21 Oct 2009 01:45:47 GMT

Ok, I forgot that you have to initailize and formet the drives after the VM starts up. Luckly, I could save the configuration as a template and not have to worry about it again until I make a new template.

Joe

Lab Manager deploy approaches

jvanqua — Tue, 20 Oct 2009 20:05:09 GMT

Let me echo the more RAM and CPU statement from before. We are using our LabManager for a test and staging environment and I am very impressed. The hardest part has been to sit down and define the templates as much as possible in advance. Then get the most common templates created and start creating the configurations. I give my leads access to create configurations and deploy machines ( I limit them to 2 or 4) and then I have a group who can create templates.

Hope this helps.

Unable to power on configuration

kajh — Tue, 20 Oct 2009 08:22:48 GMT

DiviAlcatel: Thanks for the tip. Solved the exact same problem with my ESX/Lab Manager setup after upgrading it to v4.

cheers

Lab Manager Server p or v???

efulton — Mon, 19 Oct 2009 19:02:52 GMT

I'd suggest V myself.

1) LM vmdk can be on the SAN, and thus backed up atomically with the datastores via san duplication mechanisms (very important--the lm database the datastore cannot be out of sync)

2) Initial hardware purchasing -- get all the same and save yourself some effort instead of doing a second round of research for the win2k3 boxen.

3) leverage DRS and HA for the LM server itself

Similarly, the VC server can go in a VM as well.

The only reason I'd think of P would be if I had some spare server "lying around" that's just limited enough to be a poor choice for ESX, but just powerful enough that it'd be a waste to scrap it....

LabManager 3: Error: The numPort value: 8608 in spec exceeded maxPorts 8192

ArrowSIVAC — Mon, 19 Oct 2009 14:36:53 GMT

I have poked around a bit but do not see where in the dvSwitch I can modify the default port group count settings from the default 512.

I also would like to clarify that all that would be effected with this change is the logical number of host "NICS" that the port group could allow to connect to it. It would seem logical that a count of 256 would be plenty considering that most "PortGroups" represent a VLAN segment which usually has a single range of IPs (254 usable if class C masked) plus that large of a broadcast domain is pushing the higher limits of best practices anyway.

Programmatically create a configuration from templates?

ACKH — Fri, 16 Oct 2009 15:15:07 GMT

Thanks for your answer. It actually looks like this is possible using the internal API. I haven't tried it yet though.

Many Thanks!

Wait until computer is booted

ACKH — Fri, 16 Oct 2009 15:13:04 GMT

Well, the problem here is that the machine status is set to "On" which is basically correct but it is not fully booted yet. I did not find a way to get that information. In the VMware server (the freeware one) there was a very useful "VMWaitForToolsInGuest" method which blocked until the VMware tools in the VMware image were started.

I checked the internal API but could not find something similiar...

Using SOAP I seem to be limited to two actions at any one point

AHutton — Fri, 16 Oct 2009 12:49:06 GMT

I have an sutomation system using the SOAP api, and it seems like no matter how many commands I issue only 2 or 3 of them are every being acted on at any one time and all the others queue up. Is this a hard limit, is this a setting I have set somewhere? Any helpw ould be appreciated.

The relationship between Lab Manager and VMware Center

maru3 — Fri, 16 Oct 2009 00:34:11 GMT

Thank you for your helpful answer

All of a sudden I can't obtain or set an IP

_morpheus_ — Wed, 14 Oct 2009 21:24:06 GMT

Yes you have to reboot the ESX server in order to change the port limit. Sorry to deliver the bad news. Make sure you set it really high so you only have to do this once.

Create workspace in Lab Manager API

shrenik — Wed, 14 Oct 2009 18:36:28 GMT

CHM files are blocked and marked as potentially malicious when you download them from the internet zone. You could right click and go to the properties of the CHM file, find a button called "Unblock", hit it and that should fix your problem.

multiple snapshot of guest OS

rbkinsey — Wed, 14 Oct 2009 15:42:32 GMT

Unfortunately lab manager does not. It only supports 1 snapshot. You might
want to consider taking your XP image, creating a template of it and then
adding in your different scenarios as a configuration. This will enable you to go back to your XP
image and then you baseline configurations.
For example:

XP Base Install(template-published )

Template 1 with App A (saved as a configuration)

Save in the Library and then have your users copy to workspace to use

Template 1 with App B (saved as a configuration) -----Save in the Library and then have your users copy to workspace to use

Template 1 with App A + App B (saved as a configuration)-----Save in the Library and then have your users
copy to workspace to use

Etc…

I
do not believe you can revert back to the original snapshot if you already
deleted it. Need more info on that question
and maybe you should open up another Post.

Rob

Thin Provisioning in Lab Manager

scottkey2003 — Wed, 14 Oct 2009 11:30:00 GMT

I have created templates within Lab Manager and VCenter that are thin provisioned but the machines that are created from these templates are all thick. REading other forums says that the vmdk's have to be edited. Is this the case? Is there a plan to have this step automated in a future release?
Scott

Help deploying unfenced configs from library to new unfenced config

rbkinsey — Tue, 13 Oct 2009 23:03:58 GMT

I am having the same issues. I am trying to deploy a workstation in an unfenced environment and having the same challenges. Any new news from VMWare?

UnFenced Costomization-Scripting challenges

rbkinsey — Tue, 13 Oct 2009 22:27:44 GMT

One way I plan on utilizing lab manager is to use it in an Un-Fenced environment for application UAT testing. For this to happen I need to modify the Customization Script. I was curious on how others were modifying this to change system name, OU location, time zone, personal settings, Domain, etc... If anyone can point me in the right direction it would be greatly appreciated.

Are others trying to do this as well?

What do other businesses use this product for?

Thanks, Rob

Lab Manager v3 licensing question

tascheHyaene — Tue, 13 Oct 2009 19:12:40 GMT

Much obliged.

managing LM resources by limit users and suspend?

sraisoni — Tue, 13 Oct 2009 18:42:30 GMT

To answer your first question; there is no way to limit the users that can log on into a system at any point.
However if you are using LM 4.0, you can create workspaces, assign users and resources to it thereby limiting the resources being used by the users in that workspace.

Change vcenter server

shrenik — Tue, 13 Oct 2009 18:22:49 GMT

Here's one way to do it: Backup the DB of the old vCenter server ... create a new machine and restore the DB backup on it. Install vCenter server 4.0 on the new machine and during installation point to the restored DB instance. Then point LM to the new vCenter server.

Lab Manager keeps a list of objects it sees in vCenter based on the object IDs. Doing what is described above ensures that the objects do not change as LM sees them in vCenter, otherwise things can go bad. Take DB backups of both LM and vCenter where necessary, in case you need to rollback anything.

Static Fenced External IP

vanslyk — Mon, 12 Oct 2009 23:02:07 GMT

Hello Steve,

Just checking-in again on current status of this feature since the last update was in January. Have you identified a specific release yet?

Thanks,

Kirk Van Slyke

Console View shows as black or disconnected on lab manager sessions run from a different domain?

shrenik — Mon, 12 Oct 2009 19:23:33 GMT

Try this: http://kb.vmware.com/kb/10233

There's a bunch of other threads on the blank console. Check them out, you can search for them.

deployment problem!

shrenik — Mon, 12 Oct 2009 19:15:39 GMT

Well technically, the 8 host limit is not a Lab Manager limitation but a VMFS file system limitation. But yes, Lab Manager honors that limitation.
Fencing is one of the common reasons to run into this issue, but not the only one.

Also, Lab Manager v4 which supports cross-host fencing has already released a few months back. Check it out.

Planned support for Windows 7 / 2008 in LM 3.02?

IamTHEvilONE — Fri, 09 Oct 2009 20:09:52 GMT

You're right that we cannot comment on official standpoints.

Lab Manager will always rely on the ESX OS Support. Let alone, these are still experimental support in ESX 4.0 ... which means stuff still may not work as expected ... but it's there.

Regards,

EvilOne
VMware vExpert 2009

NOTE: If your problem or questions has been resolved, please mark this thread as answered and award points accordingly.

Best way to migrate to new datastores

Scissor — Fri, 09 Oct 2009 01:08:50 GMT

I have a question about the best way to move VMs to new datastores in Lab Manager.

Running LabManager 3.0.1.378

From the screenshot, you will see that I originally created two large LUNs (1 TB each) to use with Lab Manager (Disk1 and Disk2). One of my 1 TB LUNs currently has something like 140 Guests on it. I have since come to the conclusion that having so many machines share a single LUN is the cause of my IO contention/slowdowns, so I want to move to using smaller 200GB LUNs.

Of course, this setup is currently being used for QA testing, so I want to avoid having to power off any VMs if possible. What is the best way to migrate the VMs off my large 1 TB LUNs and onto smaller 200 GB LUNs with a minimum amount of end user downtime?

A few questions:

1. Can I simply toggle off the "VM Creation Enabled" option on Disk1 and Disk2 right now without breaking people who currently have their Guests deployed on them? I would toggle on the "VM Creation Enabled" option on my new smaller LUNs at the same time. That way any future deployed Guests would be created on my smaller disks, yet existing Guests already deployed on Disk1 and Disk2 should continue to work, right?

2. There are a few high IO Guests that I would like to move as soon as possible. Should I consider using the "Consolidate" function within LabManager if I want to get them moved sooner? Or would "undeploying" and "redeploying" the Guests after performing the changes in question 1 accomplish the same thing? Or would cloning them over to a non-LabManager ESX Host be the best idea?

3. If I want to eventually destroy my Disk1 and Disk2 (so I can reuse the space for smaller LUNs) then I am going to have to use SSMove.exe, right? I ran SSMove and poked around a bit . I saw that one of my source trees is almost 500 GB in size (see screenshot) Is there a way to break it up so it can be moved onto several 200 GB LUNs? Of course, using SSMove requires all Guests in the Tree to be undeployed first, and I would think that moving a 500 GB tree would take a significant amount of time.

Comments? Suggestions?

Thanks in Advance!

IP adress assignment behind fence

_morpheus_ — Thu, 08 Oct 2009 18:00:36 GMT

You can not use DHCP with a fenced configuration if you want the VMs inside the fence to be able to communicate outside the fence via the virtual router.
When you deploy a configuration for the first time, guest customization will run and configure the guest VMs with the internal IP. If this doesn't happen then check the VM properties to make sure guest customization checkbox is checked.
If your VM IP differs from the network you are deploying it on, i.e. your VM is 10.x and you are trying to deploy it onto a network with 192.x, then you need to use a network template and add that network template to your configuration as a virtual network. Then when fencing it, bridge physical to virtual.

LM 4.0 auto installs MS SQL 2005 Express - How to stop this

Lex_Frost — Thu, 08 Oct 2009 15:12:09 GMT

While preforming our first attempt to upgrade our Lab Manager 3.0.2 server to Lab Manager 4.0, it installed MS SQL 2005 express and moved our database to that instance of SQL. We already have our 3.0.2 database running on MS SQL 2005 Enterprise because it out grew the 4GB cap that MS SQL 2005 Express has. Durning the upgrade there was no way to tell it to use the MS SQL 2005 Enterprise instance, it just automatically installed and used the SQL Express software, which will not work for us. Is there a way to tell it to use the MS SQL 2005 Enterprise instance instead and not install MS SQL Express? We would really prefer not to have to waste the time installing SQL Enterprise again just to upgrade the Express instance that seems to be forced on us.

Sadly our upgrade failed anyways so we are able to do a fresh upgrade now. Any help would be greatly appreciated.

Lab Manager 4 as a CM tool?

stanj — Thu, 08 Oct 2009 14:49:05 GMT

Does anyone know if Lab Manager 4 can do source code Software Configuration Mgt?

We were using a CM tool called StarTeam for tracking code, updated, bugs, etc for our projects.

We are may be moving to Lab Manager for software development on VMs..

thanks

System Administrator cannot create new organization

ITRC_Architect — Thu, 08 Oct 2009 13:54:22 GMT

Bingo!

Thank you.

Fowler

Deploying template failes - Permission to perform this operation was denied.

nathanw — Thu, 08 Oct 2009 07:25:23 GMT

I have solved the issue myself!!!!! Do I get the points?

The Permission Set in Lab Manager v4.0 is somewhat different to v3.0 and obviously vCenter 3.x.

I applied the extra permission options as spefied in the Install Guide for Lab Man v4.0 here http://www.vmware.com/pdf/lm40_installation_guide.pdf

Resource pools...best practices for larger LM deployments...

jbrown2 — Wed, 07 Oct 2009 22:18:24 GMT

So I noticed the other day (while studying update my VCP) that the maximum supported number of child objects in a resource pool is 256 for ESX 3.5 (1024 for vSphere). How do you guys handle this for Lab Manager deployments when there are going to be more than 256 VM's in the environment?

Currently I have 4 Organizations in my Lab Manaber 3.0.2 environment and I've created a resource pool for each org in order to break them them up in case I need to separate out the hardware or prioritize/cap one Org in the future as the environment grows. It's been 6 months since we turned on our LM environment and we almost exclusively used fenced configurations in order to allow multiple copies of the configurations to be deployed at the same time. One of our Org's has really taken off and currently has 439 VM's in their resource pool with basically half of them being virtual routers.. I'm not terribly worried about it since we've been over 256 VM's for a couple of months in this Org and I plan to move this environment to vSphere in the next 4-6 weeks but as I'm adding additional capacity to the environment and migrating to LM 4.0 I'd like to take the chance to modify my resource pools so that they can scale for the next year or two until the next upgrade cycle.

Here are my current thoughts:

create # number of new resource groups within each Org's main resource group based on expected growth over the next 2 years

enable the Orgs to use the new resource pools and remove rights to the top level resource pool...maybe during my downtime to move to vSphere...

manage any limits/reservations for the various orgs based on the top level resource pool that I create for each Org (currently I haven't limited any org as my intention is to ensure the environment isn't resource constrained)

Thoughts? Examples of how you've done this your environment? We currently have 1TB of RAM behind this system and are planning an upgrade to a total of 2.5TB of RAM which I forcast will cover our growth through 2010.

Deployment and storage leases configured by Organization

bgarner — Wed, 07 Oct 2009 17:21:39 GMT

I have also requested this feature. Currently I'm logging in and setting the deployment leases to the longer durration manually for each VM that needs it.

Cannot export a template with Lab Manager v3

IamTHEvilONE — Tue, 06 Oct 2009 13:22:13 GMT

http://kb.vmware.com/kb/1002716/

There are some unofficial fixes if you call support .... which basically modifies some of the permissions that the patch implemented.

The most direct path is to uninstall the patch.

Regards,

EvilOne
VMware vExpert 2009

NOTE: If your problem or questions has been resolved, please mark this thread as answered and award points accordingly.

Quota per VM / Deploying

pretorianec — Tue, 06 Oct 2009 11:07:24 GMT

Hello,

I find in " What Is New in VMware vSphere™ 4:
Storage"
(http://209.85.129.132/search?q=cache:hKK6Rzo_vIIJ:www.vmware.com/files/pdf/Storage-with-VMware-vSphere.pdf+Storage-with-VMware-vSphere.pdf&cd=1&hl=ru&ct=clnk&gl=ua#5) net information

" Improved Storage Resources Control
As the scope of storage resources have increased significantly with large deployments of virtualization environments, so has
the need for greater automation and control of these resources. In the vSphere release, vCenter has been enhanced
with several new storage specific capabilities to help the virtual administrator manage these environments with a
higher degree of control. These enhancements provide administrators with proactive alerts and alarms to address issues
before they interrupt the availability of applications running on those resources. vCenter allows setting permissions and
quota limits on datastores, as well as per VM."

Can You explain me what does it mean - ... vCenter allows setting permissions and quota limits on datastores, as well as per VM.

Can I set quota limit (storage space) on particular VM with its snapshots ?

Thanks in advance.

Wrong host selected when connecting virtual networks to physical

schmidwi — Mon, 05 Oct 2009 19:40:46 GMT

Thanks for the quick response. . . This is actually how my systems are setup. The physical network (under resources) is bound to the correct ESX host, and it works perfectly well when I select the network and deploy the system. The problem happens when I connect to a network template and then bridge that to a physical network. The rules about physical network binding seem to be thrown out the door.

lab manager only connect to one Virtual Center at a time?

shrenik — Mon, 05 Oct 2009 19:02:10 GMT

LM 4 supports exactly one vCenter too.

Create a new Global Organization and Renaming current

shrenik — Mon, 05 Oct 2009 18:53:30 GMT

Not possible to change the Global organization.

However, it is possible to re-allocate the configurations to different organizations by going to a configuration's properties, and doing a "Change Owner". You'll have to do this for every configuration though.

Lab Manager 4 - Scenario and Usage

rbkinsey — Fri, 02 Oct 2009 19:58:33 GMT

Looks good. Only thing to consider is to leave your
'BaseX' template in your Library and have your requesters create their own
Linked clone by Cloning to their workspace. This will save you on management
and Chain Lengths. I find this strategy works best for Non-IT
users. GL

Roll Back to v3

IamTHEvilONE — Fri, 02 Oct 2009 00:11:58 GMT

The situation sounds pretty terminal. If you don't have a backup of the database prior to upgrading Lab Manager and don't have a snapshot to revert to .... that's pretty bunk.

More than likely, reinstall from scratch. that's about it, because I think the database changes too much to just attempt putting LM 3 back on there.

Regards,

EvilOne
VMware vExpert 2009

NOTE: If your problem or questions has been resolved, please mark this thread as answered and award points accordingly.

Full Clone Option Is Grayed Out (Template Deploy)

IamTHEvilONE — Thu, 01 Oct 2009 23:28:47 GMT

I'm glad to have been of service today, now you can rest easier. ^ __ ^

Regards,

EvilOne
VMware vExpert 2009

NOTE: If your problem or questions has been resolved, please mark this thread as answered and award points accordingly.

Errors installing Lab Manager 4.0

IamTHEvilONE — Thu, 01 Oct 2009 16:03:39 GMT

I don't see anything attached.

Please remember that with the update to vCenter 4, some permissions may have changed. Please ensure that you have correct permissions on the Lab Manager account.

Please post the permissions the LM Account has, and the screen capture of the error message.

Regards,

EvilOne
VMware vExpert 2009

NOTE: If your problem or questions has been resolved, please mark this thread as answered and award points accordingly.