Random Disconnections on RDS 2016 - Loads of ESENT errors

Ben 16

hi,

recently I've been having loads of rds disconnections on different servers (but not all) where all the users are disconnected and unable to login, when trying to login via console or mstsc /admin they fail as well (console just refreshes and removed the password from the field over and over) the only thing within the logs I can seem to find is a load (sometimes around 100) ESENT errors with different error numbers 482 is a usual one and the errors seem to say webcachetiles and insufficient resources. kind of run out of ideas with this... anything to do with AV or MRT?

any thoughts/help?

Andy YOU 3,071 Reputation points

2021-08-02T03:46:26.83+00:00

HI

1.Could you please refer below document and find disconnect reason code or other related logS then share them in here?

Windows RDP-Related Event Logs: Identification, Tracking, and Investigation
https://ponderthebits.com/2018/02/windows-rdp-related-event-logs-identification-tracking-and-investigation/

Remote Desktop Client Troubleshooting: Disconnect Codes and Reasons
https://social.technet.microsoft.com/wiki/contents/articles/37870.remote-desktop-client-troubleshooting-disconnect-codes-and-reasons.aspx

2.Can we remote access the issue server 2016 by using local administrator account?

3.What's difference between both normal RD session host server and issue RD session host server?
(1)We can enter below powershell command on both normal server and issue server then check if there is any difference about windows update patch.
get-hotfix
Peter Flion 6 Reputation points

2021-08-20T07:43:40.61+00:00

Hi everyone.

We also have disconnects on multiple machines with the esent errors. Needing restart to be able to work again. Starting as stated begin july 2021 .
Tried running without antivirus / dsim /restorehealth etc.. nothing helps to stop/prevent it from happening.

If anyone has a solution to this, please also inform us.

Peter Flion
System Engineer

<Removed for Privacy>
IT-Guy-9874 86 Reputation points

2021-08-23T07:02:54.077+00:00

Hello,
Same issue, we are also in Belgium.
No solution. We opened a case beginning of July at Microsoft support, no solution.
We have the issue on vmware and hyper-v VM.
All those servers are using RDS services. OS are 2016 and 2019 Std.
We have decided to stop updating our customer's servers.
ManagedTechMSP 1 Reputation point

2021-08-24T21:44:51.023+00:00

Anyone had any luck with Microsoft yet?

We unfortunately haven't - nearly 2 months this has been ongoing for, our end users are getting super frustrated.
Lee Peden 1 Reputation point

2021-08-25T12:03:06.807+00:00

II changed the size of the VM, from a D16s v3 to 2x D8s v3 and this appears to have solved my issue on one of my collections. I've now made changed to another RDS collection from D16s v3 to a D32s v3. The correlation here being a D16s. Now, my question is, how may on this thread have their VM's running on a D16s v3?

17 answers

Joey Kuijpers 6 Reputation points

2021-11-04T14:39:52.673+00:00

We are experiencing this issue somewhere since March 2021. Somewhere around July/August it has gotten worse. We did not directly link it to O365 but issues with Hyper-V Dynamic Memory because we are seeing issue within the VM when it hot swaps memory. After doing some forced updates of O365 in the MonthyEnterprise Channel we started to link it to O365. We are downgrading to version 16.0.14026.20308 and so far we see everything is gotten much more stable. Weird things is we have 100+ VMs some with identical configs (OS+O365 versions) some having issue others having none.

I want to file a Premier Support ticket and since we are Tier 1 CSP see if I can get this escalated to get some more info on this.
If anyone wants to share their ticketnumbers from MSFT I will reference them.
Please sign in to rate this answer.
Peter Flion 6 Reputation points

2021-11-04T14:48:18.233+00:00

Thank you for the feedback.

The update after 16.0.14026.20308 experiences a weird memory leak which finally results in esent errors followed by rdp disconnections till you reboot. And as you said .. some servers do have the problem, others don't.
We use administration policy to prevent updates, I just wonder when Microsoft will fix this. Till then no office updates.

Joey Kuijpers 6 Reputation points

2021-11-04T14:57:19.227+00:00

Since we are managing a lot of VMs we use dynamic memory on them. Thing is that the whole VM will become unresponsive at some point and crashes. So this support the suspicion it has something to do with a memory leak. Probably leaks into a part of the dynamic memory what is swapped out after a while which results into the issue of becoming unresponsive.
Did you ever get some evidence regards to the memory leak? If so I would like to refer that into the ticket with MSFT.
Sign in to comment
Mark Burgess 1 Reputation point

2021-12-07T13:27:23.41+00:00

We have been experiencing this exact issue and rolled back Office Apps as per this page. Does anyone have any further update to if this has been fixed yet? If we update Office right now i'm expecting it to start all over again!!

@Joey Kuijpers - did you get anything useful from Microsoft?
Please sign in to rate this answer.
Roy Charman 1 Reputation point

2021-12-07T17:03:20.577+00:00

Nothing useful from Microsoft sorry

Joey Kuijpers 6 Reputation points

2021-12-08T07:11:23.683+00:00

Unfortunately, no they are not willing to figure this one out to the bottom as it appears. For now they've only confirmed that the workaround is the way to go. They just let me test the Current Channel latest release, but same issue still occurs.

Roy Charman 1 Reputation point

2021-12-08T09:08:57.95+00:00

Exactly the same experience I had, And that's with premier support
Sign in to comment
Mark Burgess 1 Reputation point

2021-12-08T14:00:12.313+00:00

Thank you both for your response. Premier my A**! I guess they don't know how they broke it. Hopefully it will be fixed soon as it's crazy that we can't update Office now.
Please sign in to rate this answer.

0 comments No comments
Sign in to comment
Christian Hagemeier 1 Reputation point

2021-12-08T19:27:53.757+00:00

Having the same problem. Luckily only two RDS hosts are affected at the moment.
Will downgrade now and hope it fixed for the moment.
Please sign in to rate this answer.

0 comments No comments
Sign in to comment
IT-Guy-9874 86 Reputation points

2022-02-18T09:38:25.063+00:00

Hi everyone,

The downgrade has resolved the issues for us. After some month, the 365 ProPlus did not makes any server crashes. Hopefully.

But, we have some customers that are now using Office 2021 LTSC SPLA on their server.
We have installed this version on some Windows Server 2019 and 2022.... and we have the same crash as we had with 365 ProPlus.
2016 server is not compatible with Office 2021 LTSC so we only know for server 2019 & 2022.

For some big customers we already downgraded to Office 2019, but we still have one small server with the 2021 LTSC version and it crashes every +- 14 days.

Does anyone of you have the same issue?
Please sign in to rate this answer.

0 comments No comments
Sign in to comment

Share via

Random Disconnections on RDS 2016 - Loads of ESENT errors

17 answers