Long running profiles / scripts / api calls / callbacks time out over tunnel but work on local robot

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

Scenario 1:

I am attempting to execute a profile that exists on the "nexec" profile that takes a few minutes to complete.
This profile works fine when run in a "local" environment, where the request to nexec does not traverse a hub tunnel.
However, the same profile on an identical server that is connected via a tunnel times out with a communication error.

 

---

 Scenario 2: 

We are deploying monitoring profiles to target robots using UIMAPI and probe callbacks. For callbacks that require long time to execute (Eg. many attibutes) or if a scripted task executes many callbacks in the same function and this is done towards targets robots that are Tunneled the operation fails. 
 However, the same task on an identical server that is not connected via tunnel works perfectly fine. 

 

 

• Setting the profile timeout in the probe, callback or script does not help.
• Setting the probe request timeout in the probe utility does not help.
• Is there a config key that can be adjusted?

Environment

• DX UIM 23.4.*
• nexec any version
• uimapi any version

Cause

This is related to an issue where the UIM hub tunnel's SSL sessions can time out before the nexec profile finishes, causing an error on the tunnel.

 

Resolution

In hub.cfg the timeout for the SSL session can be set as follows:

hub.cfg:

<tunnel>
 session_reply_wait_timeout = 300

 

This should be set identically on both the client and server side of the tunnel.