I have a view 5.1 Installation that we just build and are about to go to production with. I am curious as I have 200 Linked clones with 1 spare that is powered on and every so often see random vm's within vCenters recent tasks going through and...

Create Virtual Machine Snapshot

Reconfigure Virtual Machine

Remove Snapshot

I am seeing this over and over with different linked clones as they are sitting off. anyone have any idea what is causing this? Initiated by TSMUSER.

Hi,

We have an issue where when a user logged in successfully to their VDI using thin client and try to login with same user credentials on another thinclient in a conference room without disconnecting the 1st one. User always get an error message "All available desktop sources for this desktop are currently busy. Please try connecting to this desktop again later, or contact your system administrator" on 2nd thinclient when they select their assigned pool. Then user clicks "close" button and can re-login without any issues.

So we did some troubleshooting and found when we remove the attached GPO the error is gone and user don't have to login twice. I tried to remove some settings that i suspected but no luck.

Any ideas?

Anyone have experince down grading view?  I assume i just uninstall and reinstall the older version.   This is do to third party support not any problems with View 5.2 itself.

Hi folks,

We are installing VMware View in our servers. We are planning to publish VDI pool. As existing physical desktops are in domain their profile and settings of user is synced with Active Directory. But user also save data in other drives like D: or E: which is not part of user profile. Is there any tool or strategy to migrate these data to users VDI or File Share folder redirection?

Please help me clarify this.

Thank you,

Hari.

Since upgrading to Horizon View 5.2, our logs in the View Administrator have been showing us an event of "Warn" serverity as follows:

Session for user S-1-5-21-467163239-218728984-193154552-16870 timed out.

However, these can be for any user, it's not just one user.

Additionally, some logs still show the username.

Additionally, some logs say "session for user null timed out"

Any idea what could be causing this? All of our pools are linked clones that refresh on logout.

I made no changes on the settings as the upgrade carries over the exact same settings and I verified that, I also even went back and uninstalled it all and went back to 5.1.0 and  its still getting the same error so now i'm back at 5.1.1 this error is on the security server, here is the whole read out

[2012-Aug-29 07:23:22.912390]: Connected to PCoIP server
[2012-Aug-29 07:23:23.542273]: SSL handshake completed with PCoIP client
[2012-Aug-29 07:23:23.606726]: Setup a PCoIP client to server UDP tunnel [3822433-1]: 10.99.35.118:4172
[2012-Aug-29 07:23:23.622351]: Setup a PCoIP server to client UDP tunnel [3092951605-1]: 0.0.0.0:4172
[2012-Aug-29 07:23:23.770789]: SSIG client connection error while reading SSIG APDU header: A request to send or receive data was disallowed because the socket had already been shut down in that direction with a previous shutdown call
[2012-Aug-29 07:23:24.078406]: Error reading SSIG header from PCoIP server: A request to send or receive data was disallowed because the socket had already been shut down in that direction with a previous shutdown call
[2012-Aug-29 07:23:24.124304]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:25.073523]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:25.891882]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:27.081335]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:28.091101]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:29.092078]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:30.083289]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:31.053015]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:32.091101]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:33.100867]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:34.100867]: Error reading UDP packet from PCoIP server: No connection could be made because the target machine actively refused it
[2012-Aug-29 07:23:35.076453]: header received

any ideas on how to resolve?

Hi,

One of our customer want to know if there is any way to detect through API followig errors of virtual desktop, and hence they can check the status and take any necessary actions to recover.

• Agent unreachable
• Unknown
• Maintenance mode
• Error
• Configuration error
• Provisioning error

Thanks

-Hao

Working from the documentation, the below is my working understanding of the Composer pool creation process (clearly I am aiming to use esxi servers that have local SSD drives, just for storing replicas and clones):

Have I left anything out? Have I mis-understood anything?

1. Also I am wondering: is it "ok" or even "good practice" to create a second snap-shot from a parent Vm (in my example above "Win7-Parent")?
2. Also - is there a decent way to calculate required storage? If my parent VM is say 35GB, then how big will the replicas get, and how big can I expect clones to get if I never recompose? (mostly trying to figure out how many clones + replica I can get on one hosts SSD drive) - or do I need to creat a new parent for each new pool that I want to create?
3. Also - am I right in thinking that the replica should be on the SSD drive (the same datastore where the clones will be) for performance reasons?
4. Also - does it make sense to use linked clones for persistent desktops? I assume that if file and profile re-direction are in place, then really it shouldn't matter whether the Vm ever gets "recomposed" or "refreshed", should it?
5. What is the difference between "recomposed" and "refreshed" and "rebalanced" anyway?

Thanx for any clarification you can help me with!

I have checked the services, and all of the necessary services are started. I try to open the webpage and I get the "unable to connect" message in Firefox. I have tried other browsers, all with the same result. Everything was working fine, until we recently moved our infrastructure from one location in the building to another. Once the View VMs (Connection, Composer, Transfer) were powered back on, the VMwareVDMDS instance would not start, so I backed up the LDIF files, removed and re-installed the ADAM without luck. Uninstalled Connection, which installed VMware VDMDS instance, and imported the most recent LDIF backup, and that's where I am now. I checked to see whether or not the server was listening on 80 or 443, and it doesn't appear that it is. Any suggestions on how to proceed?

FYI - I will not be able to attach any logs since I work in a secure area.

I checked the compatibility matrix and it says that 5.2 will run on vSphere/vCenter 5.0.   Any caveats?

To simplify my upgrade from 5.01 to 5.2,  could I install the 5.2 environment on the existing vSphere hosts,  remove the VMs from the existing 5.0 pools, upgrade their View Agents to 5.2,  add them to new 5.2 Connection server/pools,  and then have the Zero clients point to the new Connection server?

Also,  can I convert 5.01 Linked Clones into persistent desktops?

Thanks,

-MattG

Hi all,

We are currently fine tuning our new View environment before rollout to around 45 staff. I have the following specification:

3 x Hosts

HP ProLiant DL385 G7

24 x CPUs at 2.5GHz (AMD Opteron 6180SE)

100GB RAM

ESXi 5.1.0 /799733 -- VMware vSphere Essentials

HP Thinclient t510

[Total Resources of 287GB RAM and 3.64TB storage and 179GHz CPU]

I have:

viewSecurity server 5.2

vCenter server 5.1

View server 5.2

viewComposer 5.2

1 x Cluster

23 x Datastores (SAN into LUN accessed via 8/20q hp storageworks switch connected directly into the hosts)

I have created multiple master images of Windows 7 32-bit with 3GB RAM and Windows 7 64-bit with 4GB RAM ( some with 1 x vCPU and some with 2 x vCPU) through VCenter and done the following:

1. Optimised the VMs with the optimisation guide - http://www.vmware.com/files/pdf/VMware-View-OptimizationGuideWindows7-EN.pdf

2. This included running the commandswithnopersona.bat file

3. Reduced windows graphics to best performance and no pagefile on any partitions, i.e. turned off

4. Stopped all services as advised in the optimisation guide

5. Applied various Group Policy as advised by the optimisation guide in the OU of the Active Directory where the VM sits

6. Replaced the VM Audio (devtap) driver with the teradici one (latest version)

7. Disabled in the BIOS all COM ports and LPT1 ports

8. Removed Floppy Drive

9. Configured the VM with 128MB RAM and 3D enabled

10. Add this in the configuration of the VM: "mks.poll.headlessRates" as per KB note: http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2010359

11. Created the VMs with Thin Provision

12. Installed the Windows hotfix for the VMXNET3 adapter

In View 5.2 I have created Test pools of Linked-Clones with PCoIP as a default with full 512MB of vRAM and Automatic handling of 3D Rendering. These test pools have been pointed to the base snapshot of the optimised windows 7 machines.

The problems I have noticed (both on the HP thinclient t510 view client or my beefy Desktop PC accessing it via view client) are as follows:

1. Video is extremely choppy in Chrome and Firefox, smoother on IE (but still juttery)

2. Average CPU utilisation of VMs in 1 x vCPU or 2 x vCPU is always high, at least 60% upwards

3. If testing audio with online radio via any browser OR using something like Spotify AND scrolling up and down on a web page, the audio is significantly distorted and the CPU shoots up

4. The VM seems to behave particularly badly with regards to audio, applications seem calmer although a higher than average CPU utilisation is noticed

5. Connecting via the Security server into the connectin and onto the VM is smooth and even HTML access works very well

My thoughts at the moment are:

1. ESXi 5 hosts incorrectly configured with regards to the network cards (4 per host)

2. Networking switch issue, (not managed by me but easy to contact the service providers on-site)

3. PCoIP problems

4. Missing an important Windows 7 optimisation

5. Not updating the VMs with full windows updates before optmising

I have at this stage hit a brick wall and am concerned Im missing something really obvious here!

Athemiya

Hi,

I have a small pool with avout 30 Windows 7 64 bits desktops into it. On a reccuring basis, some users can't connect the the pool, even if desktops are availables. The pool is linked-clone with floating assignement. Each time the users log off, the desktop is refreshed.

The users receive some errors like the display protocol is not ready or no available desktops. But at this moment, many desktops are available.

If found in the log that the user try to reconnect always to the same desktop when the problem occur. The log only say that the user close the session by itself, but it's not the case.

The only way to fix the problem is to reset the desktop in questions and during the time that the desktop reboot, the user can log to another one.

Any ideas?

Hi all,

A very annoying and frustrating issue is arising when View 5.2 configures a new Pool of linked-clones from a pre-existing main image and snapshot.

When Vsphere 5.1 gets to the point of cloning the images and running through the tasks and usually after it has restarted the VM once or twice already, the event status logs the following:

Task:Reconfigure virtual machine.....VM name.....Invalid configuration for device '0'.

Vsphere/vCenter then continually loops to reconfigure this device, but keeps on returning the above error and if left will run for 24 hours.

I have checked the following:

1. Made sure master VM and snapshot has no CD-Drive or Floppy present

2. Tested the snapshot and main image with other pools and linked-clone tasks results in no errors

I have checked the linked-clone pool desktop settings and there are no strange devices listed or anything that I can see that suggests the wrong thing.

Help!

Best,

Athemiya

Hello,

I’m involved on a View design and I came with something security/management wise with view. We have a requirement that we need to share the view servers with multiple “admins” but those admins cannot manipulate or see in any way other admins pools or base images (Delegation of Desktop Pools).

So I’m creating folders inside view and in vSphere folders and resource pools for them and assigning permissions to those directly for the admins.

If I login to vSphere or view with a scoped account it only sees that whatever folder they have permissions to … so far perfect…

The problem is when they are going to create a new desktop pool at the vcenter setting step of adding a pool, they can see and use everybody base images, everybody folders and everybody resource pools… which is what I’m trying to avoid…

I believe this is because the way view connects to vCenter using a specific account that can see everything on vCenter.

Is there a way to configure view administrator to let the inventory admin to only see whatever they have been scoped to on vCenter?

I’m missing something or is not possible?

Thanks

I'm trying to set up a 2 factor authentication to a virtual desktop. I have the backend Vmware stuff configured already. The problem that I'm having  seems to be on my physical machine and the way that the View client accesses the Gemalto cards. I'm able to access/configure the cards via the Gemalto web interface and thier diagnostic tools don't show any errors. However, when I use the view client, it fails immediately saying that "The view connection server connection failed. Smart Card or Certificate authentication is required."

I am experiencing this on a desktop and a laptop. The laptop has an internal card reader which gemalto's diagnostic test picked up (see below)

***** Test Card Connect : *****

Reader :  Broadcom Corp Contacted SmartCard 0

Connect to card : No Card Detected

Reader :  Gemplus USB Smart Card Reader 0

Connect to card :Card Present

ATR (length 9 bytes): 3B 16 96 41 73 74 72 69 64

I've also noticed that the either reader utilizes the gemalto .net driver when the card is inserted. I thought there may be a conflict between the two readers but I'm experiencing the same thing on the desktop which only uses the usb reader.

The usb reader that I'm using is hwp114112b

Has anyone else experienced this issue and how was it resolved?

Hi,

I want to know if anybody has experince VMware View deployment with HP MSA P2000 as the SAN Storage?

I have a deployment case with round 200 users wanted to use Windows 7 as guest OS. What I'm considering is

to use a HP MSA P2000 G3 FC with 20+ 15K SAS disks.

I searched on internet to find a whitepaper but could'n find any.

Anybody has experince or opinion about the choice of Storage for View deplyoment.

Thank you.

NetRock

hi guys.

i deploy a VDI solution as demo.

you suppose this configuration.

in one vdesktop that have windows 7 ,one of my user want have firefox for using internet and i dont want this vdesktop having internet.

whit regard this situation how i can give this program. i mean how i can give internet whit this program. you suppose something like remoteAPP in hyper-V.

finally i want run this program from one server that have valid IP.

and im sorry for my grammatically fault in this passage becaus i cant speak english well :D.

best regards.

I keep getting this message when trying to deploy a pool from a shapshot. any idea?

Currently running View 4.6 in production; was awaiting the school semester to complete before upgrading.  In the past 48 hours, a co-worker completed an unannounced upgrade of the current vCenter 4.1 to vCenter 5.1. Based on the compatibility matrix, we now appear substantially out of spec. Currently I'm seeing "vCenter at address https://vcenter.xx.xxx:443/sdk has invalid credentials" Event errors repeatedly and View Administrator Dashboard is showing the vCenter server down with "the service is not working properly". I've reset the vCenter Server Settings password in View Configuration > Servers and tried various flavors of domain\view-admin and view-admin@domain.edu account name with no joy. Current linked clones are functional but I need to resolve these errors. Unfortunately I'm not the vCenter admin. Suggestions on resolving the errors reported? Anything on the vCenter side to check for post-the-upgrade that might provide a resolution? Alternatively, with new features and functionality introduced with vCenter 5.1, are we simply at a place where the incompatibility between View v4.6 and vCenter v5.1 requires a more compressed upgrade timeline on the View side?

Thanks in advance.

Hi

I have a couple of View 4.5 connection brokers which randomly gives a "red failed arrow" on the system service Event Database. When I expand the Event Database on the Dashboard and click on the database server, the Event Database Details - Status contains no info. On a connection broker with no error the status of the Event Database says "No problem detected".

All events are being tracked to the database and I can't see practical problems except the red arrow on the Dashboard.

If I clear the event configuration and add the same database back, everything is ok for a while.

My databases are running on MS SQL 2008, and i can find no errors/warings on the database servers.

Anyone who has seen anything similar ?

/Gekken