Odd situation around RealVNC connection component

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Announcements

Welcome Idira to LIVEcommunity!

Our community just got stronger! Explore new identity security resources, connect with experts, and discover more ways to learn and collaborate.

Meet Idira

Odd situation around RealVNC connection component

Hi all,

 

I'm facing an issue in a very specific PSM use case that I'd like to have your views on to know if:

a. There is a bug or misconfiguration somewhere and the behavior is unexpected

b. The behavior is by design and there is / there is not a solution or workaround to it

 

I'm using a customized "PSM AutoIt Dispatcher Skeleton" to establish VNC connections using a RealVNC client installed on the PSM server.

This works pretty well, no troubles so far.

 

The very specific use case is this one:

  • A person uses the AD UserU to authenticate in CyberArk pCloud and gets access to SafeU
  • In SafeU an AccountA for the VNC platforms allows to connect to ClientA with VNC, and a second AccountB to ClientB.
  • A first VNC connection is established using AccountA. So far so good.
  • Same person uses AccountB to connect to ClientB with VNC. The problem starts here.
  • If both PSM sessions for connecting AccountB and AccountA are running on the same PSM server (odds are 50/50),
  • then the same shadow user is shared and the second RealVNC client opens in the PSM session created for connecting to ClientA.
  • The AutoIt script in the second PSM session waits for the RealVNC window to appear, which never does since the client starts in a different session,
  • it eventually times out and the second PSM session closes.

 

At first glance the behaviour looks sound: same CyberArk user starts two PSM sessions on the same PSM server so the same shadow user is being used.

That's a situation that doesn't occur for the RDP connection component since it doesn't rely or need a shadow user so same CyberArk user can start several RDP sessions on the same PSM server without issues.

 

Have you already been in the same situation with a connection component that relies on the shadow user ?

What did you do to fix or workaround the issue ?

 

Ideas that come to mind but that are less than satisfactory:

  • Force AccountA & AccountB to run on a different PSM servers: problem is that I have a single regional PSM farm with two nodes and a hardware load-balancer. If I force a PSM server, I'm loosing the high-availability provided by the farm.
  • Use a jump server where the RealVNC client is installed: I would use a shared jump server in this case because creating a new VM just for one use case doesn't make any economical sense. The user given access to this jump server would therefore have broader network access to resources in our network compared to being constrained by current solution to connect only to ClientA or B.

 

Sorry, that was a long one. I would really appreciate if you read it up to this point and could share your views.

 

Best regards

Marc

 

 

 

1 accepted solution

Accepted Solutions

Thanks for the feedback, you're a lifesaver.

 

If by outside of CyberArk you meant running my au3 script with the /test script on the PSM server then this is what I tried.

Running it twice would basically reproduce the same issue I saw in PSM session, just that they occurred in my single admin RDP session on the server.

So I started tweaking the au3 script to get rid of that issue and I got rid of it in test mode.

The shock came when I tried again opening two PSM sessions with the same CyberArk user toward two VNC hosts and this time, each VNC client managed to connect within its own PSM session, not merging with the other one like before.

I haven't changed anything to the shadow user or any other aspect of the scenario, just modified the au3 script, so call me a bit puzzled here.

 

I really thought that in this specific scenario, the session isolation was no longer possible but I do have two separate PSMConnect user sessions on that PSM, albeit relying on same shadow user.

The most likely explanation for my problem, was that the VNC window handle and PID retrieval in the au3 got mixed up when two sessions were running on the same PSM which caused this "permeability" between the PSM sessions, and fixing the script got rid of the issue.

 

Again, many thanks for fueling the reflection and allowing me to find the solution.

 

View solution in original post

3 REPLIES 3

L7 Applicator

Hey @Marc Fom​ 

Very well explained and written question! Also, rather Interesting.

 

I'm curious, Does the VNC connection act the same outside of CyberArk?

 

For instance, if you run the program with the same user from the same machine, and attempt to log in twice, does it react the same? If so, then it would be suggested to run the sessions on two different machines, as you suggest above.

 

-Matt

 

 

Thanks for the feedback, you're a lifesaver.

 

If by outside of CyberArk you meant running my au3 script with the /test script on the PSM server then this is what I tried.

Running it twice would basically reproduce the same issue I saw in PSM session, just that they occurred in my single admin RDP session on the server.

So I started tweaking the au3 script to get rid of that issue and I got rid of it in test mode.

The shock came when I tried again opening two PSM sessions with the same CyberArk user toward two VNC hosts and this time, each VNC client managed to connect within its own PSM session, not merging with the other one like before.

I haven't changed anything to the shadow user or any other aspect of the scenario, just modified the au3 script, so call me a bit puzzled here.

 

I really thought that in this specific scenario, the session isolation was no longer possible but I do have two separate PSMConnect user sessions on that PSM, albeit relying on same shadow user.

The most likely explanation for my problem, was that the VNC window handle and PID retrieval in the au3 got mixed up when two sessions were running on the same PSM which caused this "permeability" between the PSM sessions, and fixing the script got rid of the issue.

 

Again, many thanks for fueling the reflection and allowing me to find the solution.

 

Great! Very nice to hear it's working now. I'm glad I could help lead you to the right path.

 

-Matt

  • 1 accepted solution
  • 86 Views
  • 3 replies
  • 0 Likes
Like what you see?

Show your appreciation!

Click Like if a post is helpful to you or if you just want to show your support.

Click Accept as Solution to acknowledge that the answer to your question has been provided.

The button appears next to the replies on topics you’ve started. The member who gave the solution and all future visitors to this topic will appreciate it!

These simple actions take just seconds of your time, but go a long way in showing appreciation for community members and the LIVEcommunity as a whole!

The LIVEcommunity thanks you for your participation!