Can't connect some machines and users to orchestrator after version upgrade

[Irene]

Hi everyone,

This is going to be very vague / high level as a topic, but at this stage I’m looking for other people’s experiences to check in which direction we can continue troubleshooting.

Since the version upgrade to 23.4, some of our machines could be reconnected to Orchestrator with no issues, but 1 machine cannot connect anymore. A user is also unable to connect anymore to Orchestrator since the upgrade, but for other users there’s apparently no issue. Unfortunately I don’t have more info at the moment, although some actions were already performed (disconnect/reconnect, restarts, trying to roll back the version on the machine, …). I’m just wondering if anyone has experienced similar issues? If so, any suggestions on what additional steps could be performed to further investigate, or any clue on potential root causes for to?

Thanks!
Irene

[prateek.mehandiratta9 (Prateek Mehandiratta)]

Hi @Irene

Did you try provisioning the particular machine again? See if that helps.

Thanks.

[Irene]

Thanks! That was already done but it did not change anything…

[Anil_G (Anil Gorthi)]

@Irene

You can check the event logs to check what error you are getting while connecting…

Or even you can verify the log data in UiPath folders to know more about the error

Cheers

[Rahul_Unnikrishnan]

@Irene Please compare the assistant version in all the machines and see whether all are updated to the same version.

Also are you able to see the robot as unavailable or offline?
Thanks

[Irene]

Thanks! This is ongoing, the logs of the event viewer were extracted and are being analyzed. I don’t know yet what’s the outcome of this analysis, but at least this is in progress

[Irene]

Good point, but the same version was installed on all machines. Still, 3 could reconnect but 1 could not. On this last machine, we tried rolling back to the previous version (just in case) but it couldn’t connect anymore.

The robots are not impacted as far as I know: there’s a template machine connected to 4 VMs, and all robots are using the same template machine. At the moment, only 3 of the VMs are up and running, while the 4th one is still disconnected, so the robots can still connect to the other ones and run. Should I check something more specific related to this?

[Irene]

FYI, the error raised while connecting is a timeout. Root cause is still unknown, so no resolution yet.
Users created before the upgrade to 23.4 can still connect with no issues.
Users created after the upgrade are unable to connect UiPath Studio/Assistant to the Orchestrator.

[Irene]

After further investigation, the connection to Orchestrator doesn’t work at all when the URL of Orchestrator is a local intranet address. However the same intranet address works perfectly when putting it in Chrome instead, so the connection to Orchestrator works perfectly from a browser on the same machine.

Here’s the two methods we’ve tested:

And here’s the corresponding error logs:

Has anyone experienced anything similar?

[Irene]

Finally solved: it was a malformed value in the NO_PROXY environment variable: semicolon instead of comma. Replacing the value in the environment variable and restarting all UiPath processes solved the issue.