Server 2012 RDS - User Profile Disks - Errors during Logoff

I have set up a test Server 2012 RDS collection (Single Server for now) and implemented User Profile disks.

I have two problems.

First: My generic test user can connect and does successfully use the user profile disk as expected. However, at logoff, the system event log contains these errors:


The error (NTFS 137) is: The default transaction resource manager on volume C:\Users\ts3.test encountered a non-retryable error and could not start.  The data contains the error code.

The warning (NTFS 50) that concerns me is:

It appears that the user profile disk is being "disabled" or "disconnected" before the profile data is completely written at logoff. What can I do to troubleshoot this?

Second:

Update: A post from Mike Connor on the following page: -LINK- solved the problem described below. 

My administrative user always logs on now with a temporary profile. At the beginning, the UPD was working and mounting. That stopped working. In attempting to troubleshoot, I logged the admin user off and deleted the UPD disk file from the share. I remember it working again after generating a new UPD disk file in the share. Soon, it quit working again. I deleted the UPD disk file again from the share and ever since, it has never regenerated a new UPD and always logs on with a temporary pr

December 19th, 2012 7:01pm

What is hosting the UPD share?  It is another Windows Server 2012 or something else?
Free Windows Admin Tool Kit Click here and download it now
December 19th, 2012 9:47pm

Thanks for replying!

The share is hosted on a domain joined (same domain as collection) 2008 R2 Enterprise SP1 - 64 bit.

The RDS server has "full control" security rights on the shared folder. "Everyone" and the RDS server have "full control" on the share.

December 19th, 2012 10:13pm

What is hosting the UPD share?  It is another Windows Server 2012 or somet
Free Windows Admin Tool Kit Click here and download it now
December 20th, 2012 12:56am

There have been reports of issues when the UPDs are stored on SMB shares that don't support SMB 3 or non-Microsoft SMB servers.  A quick way to check is to deploy your UPDs on a SMB 3 share and see if you still have the same problem.

I think the issue that Chris is having though is more due to a network/disk driver problem as the event ID 50 seems to indicate. 

December 20th, 2012 1:26am

Thanks @dgeddes. I have no problem with deploying UPD on a windows share. Do you happen to know if UPD can be stored on a NFS share?

Chris, sorry for jumping into your thread.

Free Windows Admin Tool Kit Click here and download it now
December 20th, 2012 2:35am

No, it needs to be SMB and preferably SMB 3. 
December 23rd, 2012 8:53pm

Thanks for replying again, Don.

So you are suggesting I test storing the UPDs on a Server 2012 share? Done.

I verified the template UPD was generated and the User UPD was generated. It still appears to be disconnecting the UPD before the system is finished writing all the profile data. Anything else I should look at?

No change:

Error messages.

-

Free Windows Admin Tool Kit Click here and download it now
December 27th, 2012 6:31pm

I'm guessing you don't have anything else for me to try then?

Thanks for trying.

January 3rd, 2013 6:58pm

Additionally, I have completely disabled UPDs for the collection. Both my "basic" and "admin" users log on and off now with no errors in the event log.

User profile disks seem like they would be far superior to folder redirection, as folder redirection misses some files within the profile, which are either discarded after logoff or have to be copied in at logon and out at logoff.

We are excited about the potential time savings the UPDs would seem to provide, as well as the lower complexity involved in configuration.

However, with 25-30 users per RDS server logging in and out sometimes several times a day, we would not like seeing that many errors in the event log. Also, we are not certain that corruption of user profile data is not occurring when these errors are evident each and every time a user logs off.

If you have other troubleshooting steps I can take, I would be glad to try them. I dread the call to MS support and the time it will take to work with them.

Thank you.

Free Windows Admin Tool Kit Click here and download it now
January 3rd, 2013 11:51pm

In my opinion, you really need to look at this from a networking/file system standpoint and I'm not really an expert in that area.  I know you don't want to do it, but a network trace and working with Microsoft Support to review it is probably what I would do next.
January 4th, 2013 3:33am

I opened a Support Incident with MS support. Let's see where this goes.

Thanks, Don.

Free Windows Admin Tool Kit Click here and download it now
January 10th, 2013 1:59am

Hi Chris,

Could you please let me know if you have any feedback from MSFT

I am having the Exact same errors after each user logoff/UPD volume dismount:

The Error (NTFS 137)

The warning (NTFS 50)

The warning (NTFS 140)

No Raid involved , or GPO's set

My next step will be getting a Procmon trace to check what happens on the logoff time

I will Also try to enable few of the RDS related GPOs mentioned on this link below :

http://microsoftplatform.blogspot.ie/2012/09/overview-of-all-new-windows-2012-gpos.html

I.E:

*Turn Off UDP On Client

*Suspend user Sign-in to complete app registration

*Select network detection on the server

Many Thanks


January 11th, 2013 3:26pm

djamsite1 -

I have been out of pocket and today is the first day I've had available to work on this ticket. I will post whatever I learn from MS back to this thread.

If you come up with a solution, please post it here as well. 

Thank you.

Free Windows Admin Tool Kit Click here and download it now
January 22nd, 2013 7:58pm

Hi

I have the same errors in my setup. I am running the environment on a VMware ESXi platform 5.1

This is my system eventlog

January 25th, 2013 2:14pm

  The vibe I'm getting from MS is that this is a "common" issue. They are working with one or two clients and trying to come up with a solution or workaround. They told me they would keep me posted on results. The last time I spoke to them was a week ago, Wednesday (Jan 23, 2013).
Free Windows Admin Tool Kit Click here and download it now
January 30th, 2013 5:22pm

Hi Chris,

Thank you for the updates,

I have been in Touch with MSFT and the issue seems to be a known as you said probably a BUG ...

There is no solution so far it might take some time

I will Share a lengthy info on what i have done so far with no luck

I have tried everything possible to find the cause,

These can be eliminated as causing this error, BUT YOU COULD CORRECT ME IF I AM WRONG !!!

=================================

*Configuration issue (tried on different labs same result)

*Network or drivers ( Does not make a difference, took network traces and there is not packet loss or issue with SMB/RDP/TPKT protocols)

*Disabled all the NIC features (ipv6-Chimney,TaskOffload,Jumboframes,VMQ ..)

*Disabled SMB 3//2/1 through the registry and PowerShell (few new option through the SMB module)

*Disabled few system services that are called during the RDS connection (spoolsvc , VSS ...)

*Used the minimum options under RDP connection (disabling the local resources, Experience to the lowest...)

*GPO (Tried all RDS ones including the UserProfiles policies)

*RDSPNF.exe (new RDS service for 2k12 ) seems to take longer time to act before the logoff

=====================

The GOOD NEWS: there is no corruption of the profiles i have tried numerous times using scripts (logoff/logon)

The only thing to avoid the errors is Detaching the UPD for the specific user from the RDS host before the logoff.

However it might corrupt the profile :(

The conclusion is that the user profile closes faster than the VHD of the UPD is detached or the other away around

which might bring back some bad memories (UPHCLEAN) windows 2k3/XP issues

UPHCLEAN

===========================

Workaround is to create a PowerShell/Batch  to detach the VHD before the VHD detach,

But i don't think that is a good solution it might corrupt the users profiles and make thing worse

Chris please let me know if you get any updates i will also

Regards

February 1st, 2013 8:45pm

Hi Chris and djamsite1,

I am having exactly the same errors as you mentioned before. Did anyone of you already get a useful feedback from MSFT ?

Regards

Free Windows Admin Tool Kit Click here and download it now
February 11th, 2013 4:06pm

FoG_Sigi -

  I have not yet received  any new information. MS collected more logs from us last week, but they have not reported any progress on the issue. I will post back to the thread once I have information that could help.

- Chris

February 13th, 2013 10:05pm

Hi Chris, djamsite1 and FoG_Sigi,

I have the same problems on my Server 2012 Session-hosts. I hope MS have a KB in the future.

Regards

Karsten.

Free Windows Admin Tool Kit Click here and download it now
February 14th, 2013 9:05pm

 

Hi Guys,

Just wanted to update you on the case,

Still no feedback yet from MSFT it might take longer than expected, its definitely a BUG :(

I believe that we need to wait till SP1 ...

Regards 

Djamsite1

February 20th, 2013 12:26pm

Hi Chris and djamsite1,

I only want to say thank you for keeping us up to date ;-)

Free Windows Admin Tool Kit Click here and download it now
February 20th, 2013 12:48pm

Hello all -

 I have been informed today that Microsoft is refusing to troubleshoot our issue because it is virtualized under vmWare ESXi. They require we reproduce the problem on a physical machine.

Anyway, We are evaluating whether or not this is important enough to us to go to these lengths to implement.

Good luck.

February 22nd, 2013 12:38am

Just thought I'd jump in and say that I'm having this exact same issue on 2x Server 2012 machines (belonging to separate customers), both acting as standalone RDS servers. The user profile disks is located in c:\userprofiles on the same server so there's no issue with network shares, drivers, SMB versions, it's as simple a setup as you could wish for.

The only slightly non-standard feature of these servers is that they are virtualised (both are inside a Server 2012 host, one with no other guests). There is enough space, good high-speed SAS disks, battery-backed RAID card on latest firmware etc. All kinds of tweaks were made to optimised Hyper-V performance according to various guides available too, so I can't think what else there is to blame.

It just so happens that I am having VSS issues on both these servers as well which are causing backup failures or crash-consistent backups at best, and the support guys at the backup software provider are pointing to these NTFS/WinLogon issues as the reason for the failures in both servers. (I have the same backup software running on several other hosts with no issues whatsoever).


Free Windows Admin Tool Kit Click here and download it now
February 23rd, 2013 5:19pm

Guess what, we're monitoring this post because have EXACTLY same problem on a single physical machine, the nearest thing to virtualisation we've got going on is that the affected machine is an RDS server. 

Eagerly waiting an answer MS....

February 25th, 2013 4:27am

Chris,

Dont let Microsoft cast blame on VMware so easily.  Its disappointing that MS plays that wild card so easily.  Obviously this issue is not unique to ESXi.  Microsoft supports their server products on VMware through the Server Virtualization Validation Program (SVVP). SVVP is available to all customers through their normal support relationship with VMware or Microsoft. Any level of support from either vendor (including per call support) is acceptable to access this option.

For more information see:  http://www.vmware.com/support/policies/ms_support_statement.html

Jason

Free Windows Admin Tool Kit Click here and download it now
February 27th, 2013 5:46pm

Hello all -

 I have been informed today that Microsoft is refusing to troubleshoot our issue because it is virtualized under vmWare ESXi. They require we reproduce the problem on a physical machine.

Anyway, We are evaluating whether or not this is important enough to us to go to these lengths to implement.

Good luck.

March 14th, 2013 3:53am

dgeddes -

  Thanks for the offer. I'm not exactly familiar with Forum protocols and functionality. For the life of me, I cannot figure out HOW to e-mail you. Is there something I'm missing? Is there somewhere I can click to PM you or send an e-mail to you without already knowing your e-mail address?

  Also, from the other direction, can YOU e-mail me without that information?

- Chris

Free Windows Admin Tool Kit Click here and download it now
March 15th, 2013 5:51pm

It is my forum name at microsoft .com.

March 16th, 2013 12:40am

I have the same error (ntfs 137) on my Windows  RDS servers. This two virtual machines are working on three nodes Hyper-V cluster. I tried to store user disks on Windows 2012 via network (on the same cluster), or directly on Cluster Shared Volume (SAN disk) connected to both RDS servers, or directly on system disk. Always the same error on logoff. It's definitly a BUG.

Free Windows Admin Tool Kit Click here and download it now
March 25th, 2013 11:39pm

Final word. 

No help from Microsoft. If anyone hears about any hotfixes or patches that will correct this, please post back on this thread for everyone.

Looks like they dropped the ball on this one.

April 4th, 2013 4:11pm

I'm looking into this for a different incident. If I develop a solution I'll be sure to share my findings here!
Free Windows Admin Tool Kit Click here and download it now
April 4th, 2013 6:58pm

So I am in the process of deploying an RDS farm. Should I enable the User Profile Disk or not? I did try it in the test setup before hand and it corrupted the Administrator profile so bad I had to completely trash it. I have now started from scratch. I have RDS installed on one server running Windows Server 2012 on VMware and waiting to add it to another physical server.

Has anyone had luck with solving this issue?

I am onsite ready to pull the trigger as I have 18 thin clients to deploy.


April 9th, 2013 7:14pm

So I am in the process of deploying an RDS farm. Should I enable the User Profile Disk or not? I did try it in the test setup before hand and it corrupted the Administrator profile so bad I had to completely trash it. I have now started from scratch. I have RDS installed on one server running Windows Server 2012 on VMware and waiting to add it to another physical server.

Has anyone had luck with solving this issue?

I am onsite ready to pull the trigger as I have 18 thin clients to deploy.


Free Windows Admin Tool Kit Click here and download it now
April 9th, 2013 10:14pm

I too have had this problem on a pure 2012 platform. Everything was fine until a bad shutdown one day and ever since then it hasn't worked no matter what we tried. We did remove the bad profiles (ones with .bak) in Regedit under HKLM\software\Microsoft\WindowsNT\CurrentVersion\Profilelist\ , gut that only allowed us to connect to the corrupted profiles again. It still will not update the changes to the UPDs. This solution has great potential, but without a fix for this it's not viable for a production environment.

How about it MS? Are you going to fix this or not?

April 14th, 2013 10:29pm

I too have had this problem on a pure 2012 platform. Everything was fine until a bad shutdown one day and ever since then it hasn't worked no matter what we tried. We did remove the bad profiles (ones with .bak) in Regedit under HKLM\software\Microsoft\WindowsNT\CurrentVersion\Profilelist\ , gut that only allowed us to connect to the corrupted profiles again. It still will not update the changes to the UPDs. This solution has great potential, but without a fix for this it's not viable for a production environment.

How about it MS? Are you going to fix this or not?

Free Windows Admin Tool Kit Click here and download it now
April 15th, 2013 1:29am

Any News? Same issue here at two of our clients. In this case UPD is not useable...thanks.

April 19th, 2013 5:17pm

We are seeing the same issue on a Windows Server 2008 R2 SP1 with Hyper-V machine, Windows Server 2012 Datacenter edition with a 5 node RDS Session Collection Farm.



Free Windows Admin Tool Kit Click here and download it now
April 29th, 2013 5:40pm

Hi Chris!

First off, I'd like to thank you for your persistence with this issue.  We are about to deploy a two-VM RDS farm, both servers running 2012 Standard using UPD's, and are having the EXACT same problem!

I was just posting here to re-enforce the fact that this is a HUGE issue that needs to be addressed by Microsoft ASAP.  Thanks for the time you put into this, and thanks to everyone else for responding.  Makes me feel better that we are not the only ones experiencing this issue.

May 17th, 2013 5:49pm

There are 3 issues here that we are actively tracking:

1. The event ID 137 and 140 from NTFS that seem to indicate that we've lost data from the UPD when there has been no data loss.  To date, we do not have any cases where real data has been lost, but if you are seeing data loss from the UPD then please open a case with Microsoft support so that we can investigate.  It is likely that these events can be safely ignored if you are not seeing data lost, but I do not have a confirmation on that yet.

2. Users reporting corruption of UPDs after an unplanned outage of the file server where the UPDs are stored OR the Session Host server that had the UPD loaded (system hang, bugcheck, etc.)  In all cases, once the servers are stabilized, the problems with UPDs are resolved.

3. Admins logging on directly to multiple session hosts that are in collections where UPDs are enabled. This causes the UPD to be loaded for the admin session and any subsuquent sessions for the same admin account will fail to load the UPD.  User will be notified that a temporary profile was loaded.  This occurs when logging on to the host via RDP or by using VMconnect (Hyper-V console).  This is by design. 

In all cases, please do not accept your case being closed if you are hosting your Windows Server 2012 on VMware, or reopen your case that might have been closed previously.  If the issue does turn out to be a bug, ask to have the cost of the support incident refunded.

Free Windows Admin Tool Kit Click here and download it now
May 18th, 2013 2:25am

Hi

Im in the same boat, i have experienced a few profile corruptions becasue of this bug, ive had one just this morning. Now the user gets temp profile. My recovery process is this. Reboot both RD Servers and File Server [UPD Store] after hours, then restore profile from a previous version.

Environemt is this;
2x IBM Servers and a IBM DS3524 SAN. VMWARE 5.1 on both hosts. And Server 2012 for the RD Servers.
My UPD is on a Windows SBS 2011 server. Majority of the time users login and out without issue, but so far over a few months ive had profile corruptions.
If they are at the satrt of the day i cannot resolve and have to configure the temp profile for them to use until i can fix it after hours.

Ive done a bit of research online and there a lot of people with same issue. Microsoft cannot wash their hands of this. We need a fix.

Hopefully we can keep this thread alive and get a solution posted too.

May 22nd, 2013 3:01am

Hello,

same problems here.

RDS-Farm, 3 RD-Servers in it. very huge problems with the UPD and corruptions. (3 of 3 from dgeddes list)

My temp workaround: set 2 of the 3 servers to decline new connections. all users are now on the same server but thats not exactly my understanding of a working RD-Farm with loadbalancing.

That might work for the next 4-5 weeks but more users are about to migrate to the new farm if this happens it's going to be ugly.

Help will be appreciated!

Free Windows Admin Tool Kit Click here and download it now
May 23rd, 2013 3:09pm

Hello,

Same issue here. TS Farm with 2 RDSH Servers User Profile Disk on shared folder.

All servers are virtual on Hyper-v 3.0. Everything works fine except event log error. No corruptions is detected to user profile disk

Any updates?

May 31st, 2013 12:07pm

Hello,

here the same problem.

We have 4 RD-Servers and 2 Domain-Controllers running on 2 ESX 5.1 servers with 128GB RAM each.

Our temp. solution is to deaktivate the UPD - but the UPD is a great feature.

Microsoft, please come up with a solution.

Free Windows Admin Tool Kit Click here and download it now
June 26th, 2013 12:18pm

I have the same problem at a customer site. However, our UPD location that gives an error is on a Server 2003 front end for a Dell NAS.

When I point UPD to the session broker host, as a test, it works fine.

Setup:

2 RDS session hosts - physical machines

1 RDS session broker - ESXi guest (Server 2012)

Great feature - if it works.

Is Server 2003 not a valid choice for the file share? I have both the session hosts and broker as full control of NTFS and Sharing and even made those three servers local administrators on the 2003 NAS host. On the Session Host (temporary test location of UPD) I have only listed the NTFS and Sharing for the servers - did not add to local admin group.
  • Edited by CNY-IT-Guy Monday, July 01, 2013 2:01 PM update
July 1st, 2013 1:51pm

I have the same problem at a customer site. However, our UPD location that gives an error is on a Server 2003 front end for a Dell NAS.

When I point UPD to the session broker host, as a test, it works fine.

Setup:

2 RDS session hosts - physical machines

1 RDS session broker - ESXi guest (Server 2012)

Great feature - if it works.

Is Server 2003 not a valid choice for the file share? I have both the session hosts and broker as full control of NTFS and Sharing and even made those three servers local administrators on the 2003 NAS host. On the Session Host (temporary test location of UPD) I have only listed the NTFS and Sharing for the servers - did not add to local admin group.
Free Windows Admin Tool Kit Click here and download it now
July 1st, 2013 4:51pm

My issue was that the UPD share did not have the proper permissions. Someone had removed system from NTFS. All is well now.
July 2nd, 2013 1:50pm

Seeing the same issue.  I think it can be fixed by setting the virtual disks to quick removal and disabling write caching - but I can't figure out how to make that change and make it persist.  If we could set the virtual disk to discover with Quick removal enabled, that'd be worth testing.
Free Windows Admin Tool Kit Click here and download it now
July 25th, 2013 11:00pm

We had to use the same "temp" solution as Flow.Ryan. Spent days trying to get it to work and we have Hyper-V 2012 Data center edition with Hyper-V VMs. Would like to use UPD but it does not work. Had to constantly battle with temp user profiles issues.
July 26th, 2013 8:58pm

Seeing the same errors.

Does anyone know how to map the Volume GUIDs in the event to an actual file or device? (either VHD file or physical volume). The errors I'm seeing always reference the same Volume GUID, so my suspicion is there's something wrong with a specific (single) vhdx.

Thx in advance!

M.

Free Windows Admin Tool Kit Click here and download it now
August 6th, 2013 12:38am

To convert the UPD SID (file name) to username and vice versa check this out: http://blogs.technet.com/b/heyscriptingguy/archive/2010/10/12/use-powershell-to-translate-a-user-s-sid-to-an-active-directory-account-name.aspx

August 9th, 2013 8:49pm

Hi,

To help with the SID's, I also wrote this little script to generate an overview of all the UPD files with the corresponding users. It's available on TechNet Gallery here:

http://gallery.technet.microsoft.com/Retrieve-usernames-for-a-94780a9e#content

Free Windows Admin Tool Kit Click here and download it now
August 9th, 2013 9:05pm

Does anyone know if this UPD problem still exists in 2012 R2 ?  Hoping perhaps someone has installed the Preview and tested this.

Thanks - Tom

August 9th, 2013 10:34pm

Same problems here.  What was really bad was during testing, the Administrator accounts started to not be able to connect to their UPD's, so they kept using temp profiles.  I really hope 2012 R2 fixes these problems.
Free Windows Admin Tool Kit Click here and download it now
September 19th, 2013 3:07pm

It looks like Microsoft released a hotfix for this issue:

http://support.microsoft.com/kb/2896328

November 14th, 2013 6:57am

Just tried it.

Unfortunately doesn't seem to work :-(

Free Windows Admin Tool Kit Click here and download it now
November 14th, 2013 12:21pm

Hi Guys

I have only one session host with local UPD and I'm getting the issue too. What occurs with user profiles if I turn off the UPD features ?

November 19th, 2013 2:43pm

If you diable UPD then Profiles will be stored locally on server.

You can find them in the users folder.

Or you can set up roaming peofiles (not reccomeded) and or folder redirection.

Free Windows Admin Tool Kit Click here and download it now
November 19th, 2013 2:49pm

I understand that the profiles will be stored locally, but I just want to make sure that the existing profiles would be saved on their "actual state" . I fear the users loose their profile customization
November 19th, 2013 3:16pm

Same problem here. 3 RDS servers with a connection broker. the UPD are stored on a Windows server 2012 and all the rights of the shared folder are correctly.
Free Windows Admin Tool Kit Click here and download it now
November 28th, 2013 9:57am

Same problem here. Tried several different options, no luck. Only thing working is local profile.

Hope Microsoft comes with a decent fix, as it now renders the complete idea of a broker with multiple session hosts completely useless.

November 29th, 2013 7:51pm

Any one with a solution?

In our case the problem is only related to thick clients ( Windows XP SP3) and not with our thin clients ( WIndows 7).

We will try to investigate thiqs problem

Free Windows Admin Tool Kit Click here and download it now
December 2nd, 2013 1:18pm

I opened a case with Premium support regarding ths issue. I was told that it is a know case and I will be updated when a solution is found. 3 Months later, still the same.

This morning I ran a test enviroment on Windows 2012R2 and guess what still the same issue! It seems that Microsoft don't have a clue how to fix this issue.

Microsoft, do you want to wake up and fix the issue?

December 2nd, 2013 4:08pm

Microsoft,

A lot of users with the same problem, you need to solve this issue asap!!

Free Windows Admin Tool Kit Click here and download it now
December 3rd, 2013 8:07am

I also checked out these hotfix info (without installing it):

http://support.microsoft.com/kb/2896328/en-us

It seems to be the same symptoms. Who else installed this fix and who is helped with it?
In the description i don't see any remarks about 'UPD'

Also it is not good news to read that in 2012 R2 this isn't solved.

What is the alternative? Switch off UPD and set the user profiles local on the RDS host under C:\Users?

How to set a quota on these user profiles?

December 5th, 2013 5:27pm

Anyone a working solutions?

Free Windows Admin Tool Kit Click here and download it now
December 11th, 2013 3:24pm

We too are having this issue!  Anything from Microsoft yet on a fix????
December 13th, 2013 7:24pm

I hope this issue has not been forgotten about.  I am beginning to run into this issue where UPD's are not loading and instead temporary profiles are loading for our users.  It happens sporadically.  Initially i thought it was due to a profile disconnecting rather than logging off and then trying to log onto a different RDS host server, but that no longer seems to be the case.  Below is our server set up:

1 2012 management server

1 2012 RDS broker server also handling licensing and RD web access

2 2012 RDS Host servers

1 2008 Server hosting the share UPD's are stored on

This is a production environment with over 100 users accessing at any given time.  The issue is minor as of now, with only about 3 profiles having this issue per week, but they are starting to add up.  Has MS found a solution yet?

Free Windows Admin Tool Kit Click here and download it now
January 10th, 2014 2:20pm

Yes, I hope this has not been forgotten about either!  This is causing us a lot of issues.  Microsoft, what's up???

January 14th, 2014 10:07pm

All,

After a call to Microsfot, we found a solution for our problem.

The problem with the VHD only appears on Windows XP SP3 clients, not on the clients from windows 8 or windows 7.

After 2 weeks support of Microsoft, the issue is solved.

We have received a workaround from Microsoft.

The VHD problem has been solved by excluding the VHD from the AV.

Free Windows Admin Tool Kit Click here and download it now
January 16th, 2014 8:06am

Hi

I also having issues with UPD on 2012 R2 which happen to be running in ESXi 5.1 U2.

Got ticket open with msft and waiting for update to what the solution is.

I believe there is hotfix for 2012 R1 servers however the problem appears to have been reintroduced in 2012 R2.

Don't know if it works as all my servers are 2012 R2, here's the link anyway http://support.microsoft.com/kb/2896328

Carl

January 21st, 2014 4:49pm

Everyone complaining about temp profile issues need to first figure out why they are getting it in the first place. There can be a number of reasons for the same. Correct way is to look for the following event in event viewer on your Session Host Server: 

Log Name:      Microsoft-Windows-TerminalServices-RemoteConnectionManager/Admin
Source:        Microsoft-Windows-TerminalServices-RemoteConnectionManager
Date:          xx/xx/xxxx xx:xx:xx
Event ID:      20493
Task Category: None
Level:         Error
Keywords:     
User:          SYSTEM
Computer:      rds.domain.local
Description:
Remote Desktop Services could not apply a user desktop for a user account with a SID of S-1-5-21-2871297466-3009485072-1309590870-500. A temporary profile was enforced for the user. Verify that the user profile disk settings are correct. The error code is 0x800700B7.0

The error code can be converted using ERR.exe tool from Microsoft and then you can find the problem.

For example the above error means (STIERR_HANDLEEXISTS) i.e. a handle is already open to that VHDX file and cannot be mounted. Further troubleshooting based on error code.

Free Windows Admin Tool Kit Click here and download it now
January 22nd, 2014 3:20pm

Manish, don't try to be smart about solving this. This problem is here since UPD was introduced in WS2012 and apparently persists still in R2. This has been confirmed by plenty of people with various configurations.

For example in our environment, the RemoteConnectionManager log does not contain any errors.

To those who opened a ticket with MS - did you get a hotfix/resolution? (aside from Samon900, but we're not using AV at all).

February 4th, 2014 2:41pm

Found this thread after deploying a W2012 R2 RDS farm with UPD. All servers in the farm are getting all the errors described in this thread. Is there any hope for a fix?
Free Windows Admin Tool Kit Click here and download it now
February 14th, 2014 6:38pm

Same issue here. Can't believe that there is no reaction from MS. This thread is about over a year!
February 18th, 2014 9:57am

Exact same error for me aswell in our environment at one customer.
They are using Windows 2012 R1 in a farm using two RDS servers.

As we see it it's because the UPD is not released from RDS1 when the user is told to go to RDS2 when hitting the broker.

Free Windows Admin Tool Kit Click here and download it now
February 26th, 2014 12:44pm

Same issue ....
March 28th, 2014 2:14pm

Any news? We have a customer with the same problem. Has anyone tried the hotfix?

Regards, Christoph

Free Windows Admin Tool Kit Click here and download it now
April 9th, 2014 9:00am

I applied the hotfix (kb 2896328) but it did not resolve the issues.  We don't see as many of these issues but we are still getting them periodically. All systems are either Windows7 or Windows8. 

Anyone have any ideas?

May 19th, 2014 1:16pm

Same issue and a real pain. Opened a case with Microsoft, they seem to have no clue about the issue or the way to troubleshoot it. On first contact without understanding the issue completely i was asked to install the hotfix from kb 2896328. now we are at the phase of "further" data collection. the same issue exists in 2012 R2 as well. Microsoft please fix it or at the very least admit it it is an issue and you cannot fix it or don't have the time to! Even though the company i work for is a gold partner, after about two weeks of doing nothing, i was told to open a case under PREMIER contract!well, i am still struggling with Microsoft to progress it.

Free Windows Admin Tool Kit Click here and download it now
May 28th, 2014 6:34pm

Same issue server 2012 R2... Pretty frustrating that is a known issue with know hotfixes since our MS licensing cost just as much the servers.   
June 5th, 2014 1:53am

Hello,

Same issue here, with 2012 and 2012 r2.

It can't survive any unplanned problem.

Quite often, one of the RDP server keep the vhdx open/locked on the filer, and then bad things happen.

Free Windows Admin Tool Kit Click here and download it now
June 16th, 2014 7:13pm

Hi all,

I am facing exactly in the same Issues. We have 5 RDS Server with 2012 and use UPD. We have application which try to write information to the UPD, when user is logging off. Data get lost, UPD are in user by a different RDS Server, or completely un accessible. Next log on will end in temp profile, which will not work. So sometimes we have to kick off 300 Users to reboot the server to lock off the UPD.
This is a disaster. We have to react very fast. We could not got on like this.

I did the following workaround to stabilize al little bit the situation.

In des Collection Settings, a set the disconnection time to 16 hours, because my users normaly do not sign off.

I disabled log on with temp profiles. So better the user could not log on, than having a not working profile at all.

We open a business critical ticket with MS, but only data collection happened. This is very poor.

What did all the other do in that post? did you switch back to roaming profiles? Is this worth it?

Any idea will help! Thanks.

  

June 23rd, 2014 8:22am

Thorsten, we're enduring the problem. However annoying and total failure on Microsoft's part (I can't see how this could get through testing and is present in 2 OS generations ...), it's still little less annoying than having to deal with roaming profiles.

I've previously asked if others got any resolution, but noone cared to respond (or forgot about this thread).

Best thing we came up is to spread sessions over as many RDSH as possible and reboot them regularly to lessen the impact. And regularly cleanup messed up entries in the profileslist registy entry.

Free Windows Admin Tool Kit Click here and download it now
July 2nd, 2014 2:07pm

We experience less to zero problems after an outage with our User Profile Disks when we use the option: "Store all user settings and data on the user profile disk". Other then when we use the other checkbox (even though all checkboxes underneath are checked). 

Our test setup was, an RDSH 2012 r2 server(s), with an 2012r2 server holding the share.

We first tried the "store all user settings" for the User Profile Disk, we logged several accounts onto our RDSH Server we then disconnected our server which holds the share to simulate an outage. After about 5 minutes we connected the server again holding the share. When the share comes back online, we logged off the users and logged them back in. The result was a valid profile for all the accounts including their last changes from before the share got disconnected. The exact same test was done with the option: "Store only the following folders on the user profile disk" with all checkboxes underneath checked, which gave us corrupt profiles.. We also did these tests by giving the server holding the share a reboot instead of an disconnect from network. This gave us the same result. Each test was done at least 2 to 3 times..

I hope others can confirm this as well in their situation, as it seems to be a bug within that specific checkbox.

July 5th, 2014 8:56pm

We experience less to zero problems after an outage with our User Profile Disks when we use the option: "Store all user settings and data on the user profile disk". Other then when we use the other checkbox (even though all checkboxes underneath are checked). 

Our test setup was, an RDSH 2012 r2 server(s), with an 2012r2 server holding the share.

We first tried the "store all user settings" for the User Profile Disk, we logged several accounts onto our RDSH Server we then disconnected our server which holds the share to simulate an outage. After about 5 minutes we connected the server again holding the share. When the share comes back online, we logged off the users and logged them back in. The result was a valid profile for all the accounts including their last changes from before the share got disconnected. The exact same test was done with the option: "Store only the following folders on the user profile disk" with all checkboxes underneath checked, which gave us corrupt profiles.. We also did these tests by giving the server holding the share a reboot instead of an disconnect from network. This gave us the same result. Each test was done at least 2 to 3 times..

I hope others can confirm this as well in their situation, as it seems to be a bug within that specific checkbox.

Looking at the above tests, the best way in my test cases were to exclude the default folders (f.e.: \Documents) if you would like to redirect them instead of using them inside your User Profile Disk. Redirects can be done through GPO Folder redirection options or GPO registry combined with Environment Variables.

Even after several attemps to simulate an outage, i get left with consistent profiles. I've now also included an hard reset of my virtual machine (that gave the same results as a disconnect of the network card). This was all done through using the: "Store all user settings in the user profile disk". The other option left us with inconsistent profiles. I hope others can confirm this as well..

Looks to me that the type of connection towards the User profile happens different. You can see this when looking at the C:\users\<username> and subfolders if you use user profile disks. The 2 options show different symbols as well.

Free Windows Admin Tool Kit Click here and download it now
July 6th, 2014 10:18am

Thanks for the tests Dennis, but in our case, we never used any other option beside the "Store all user settings in the user profile disk" and we still get the problems described in this thread.

How many users are we talking about? From our experience, the busier the server, the worse of an issue it is.

July 14th, 2014 7:04pm

Thanks for the tests Dennis, but in our case, we never used any other option beside the "Store all user settings in the user profile disk" and we still get the problems described in this thread.

How many users are we talking about? From our experience, the busier the server, the worse of an issue it is.

Well in that case, i don't suggest doing otherwise then using that specific option (saved us a lot of effort into fixing temp profiles after we've switched.) We have about 100 users at this point, this number will increase during the next few months. Our setup does seem to be stable now. Using fully patches 2012r2 RDSH and a 2012r2 fileshare.
Free Windows Admin Tool Kit Click here and download it now
July 21st, 2014 8:07am

We have the same issue here

ours is set to "store all users settings and data on the user profile disk" and we still get all the errors in the event log

It seems odd that MS would not come in with a solution here, as 'User Profile Disks' were a great idea, yet it seems that this problem is still around even well into 2012 R2

Our environment is Server 2012 hosted on Hyper-v 2012 - has anyone experienced this issue on hardware / not a VM?

if anyone has a solution, love to hear it!

July 31st, 2014 6:42am

Any update on this? Banging my head against the wall
Free Windows Admin Tool Kit Click here and download it now
September 16th, 2014 8:39pm

Also STILL having this issue, will be one year next month.  Have not encountered any profile corruption(knock on wood). We have always had the "Store all user settings and data on the user profile disk" enabled.

Our configuration is one RDS Server 2012 R1 VM hosting all roles on a Hyper-V Server 2012 R1 core machine.  Profile disks are stored locally on the C:\ of the RDS server.

Another issue I believe may be related is we are running wbadmin backups of the RDS 2012 host VM from the Hyper-V core server.  The wbadmin backups occasionally fail(maybe once a week).  I think the issue may be related as the failures either a while when a user is working late(still logged in) or forgets to log out(profile left open), need to monitor a bit more to verify which.

September 16th, 2014 9:22pm

Yep, same problem. RDS 2012 R2
Free Windows Admin Tool Kit Click here and download it now
October 7th, 2014 2:36pm

UPDATE: Opened a support case and got nowhere. Said they hope to have it resolved in the next YEAR. Judging by the beginning date of this thread I would say that is their standard answer. Pathetically bad.
November 17th, 2014 8:39pm

Two years later, any progress made?
Free Windows Admin Tool Kit Click here and download it now
December 10th, 2014 6:47pm

Seeing this on a session host (server 2012) hosted on a Windows 2012 R2 pServer using direct attached storage.

Help soon?
January 9th, 2015 6:03pm

When this problem occurs, for while I solved doing this:

- do logoff from RDS (temporary profile)

- while this, go to your server with UPD folder shared; open Computer Management > System Tools > Shared Folders > Open Files, and find VHDX file from your user; it probably still opened twice (by same RDS), then Close the files (VHDX)

- come back to your RDS and try logon again

If you need to know what VHDX file is, put this script into notepad, and save with name showObjs.PS1 , then execute it like "showObjs.PS1 \\myserver\UPD" or change "$args[0]" by "\\myserver\UPD" on text. It will show you a list.

$UPDShare = $args[0]
$fc = new-object -com scripting.filesystemobject
$folder = $fc.getfolder($UPDShare)
"Username,SiD" > export.csv
foreach ($i in $folder.files)
{
  $sid = $i.Name
  $sid = $sid.Substring(5,$sid.Length-10)
  if ($sid -ne "template")
  {
    $securityidentifier = new-object security.principal.securityidentifier $sid
    $user = ( $securityidentifier.translate( [security.principal.ntaccount] ) )
    $user,$i.Name -join "," >> export.csv
  }
}
$a = Import-Csv export.csv
$a
pause

I hope have helped till Microsoft fix this BUG!

Free Windows Admin Tool Kit Click here and download it now
January 23rd, 2015 1:19pm

Just found this exact issue on our system.

UPD's are shared off a 2012R2 server.

The VHDX for the user in question did not show up in the shared files list, other users did, and they were all working fine.

The file existed, but the system process had it open.  I could not force it to close the handle

February 27th, 2015 8:56pm

Just a quick note:
Same problem seems to occur with non-UPD implementations on RDS 2012.  I have a customer with 2 RDS 2012 R2 host servers. The customer does noet use UPD, but roaming profiles + folder redirection.  Sometimes a user reports he/she is being logged on with a temporary profile.  When I check, I notice that the fileserver that hosts the redirected folders and roaming profiles, has files open in write mode for this user, even if the user is logged off from the RDS host.  Just closing theese open files in local-computer-share-management is all it takes so that the next logon of the user will be normal. (without temp profile)

Not clear why this happens, but users report that when it happens, their logoff time takes extremely long, like 10 mins...

I was reading this thread hoping that UPD might be a solution for my customer, but it seems that UPD suffer from the same problem: some kind of improper closure after RDS session end.

Free Windows Admin Tool Kit Click here and download it now
March 4th, 2015 3:04pm

We have been struggling with roaming profiles as desribed by hwelvaar and thought that  UPD  was the solution. - As for the roaming profile we were able to have a work around the temporary profiles that where created by installing a Windows 2008 R2 server and use it to store the roaming profiles instead of the Windows 2012 Server. 2012 and 2012 R2 seems to hold the files locked and release them after 10 - 30 minutes while the WIndows 2008 R2 release them within 1 -2 seconds.

We are testing UPD and are also using the same Windows 2008 R2  to save the profile disk. So far the UPD are release and there are no error. We only get temp profiles when we logging on the RDS with an Xp xlient (I know it is very bad).

We will be testing this week more and we update the post.

March 30th, 2015 12:39pm

Been following this thread for a while hoping for an update.  I guess UPD are not to be used in Production at this time.
Free Windows Admin Tool Kit Click here and download it now
April 14th, 2015 3:23pm

+1 here. Server 2012 environment hosted on vmware 5.1

ntfs issues in log but no corruption or tempfiles yet.

June 2nd, 2015 3:42pm

Running Windows 2012 R2 in vCloud and have had issues with every single user profile creating a temp. I wouldn't advice anyone try it in production until there is a fix that brings consistency.
Free Windows Admin Tool Kit Click here and download it now
June 7th, 2015 5:56pm

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics