Our client has more than 5 robots and each robot is located on a different server. Windows server-related problems can sometimes occur on these servers, and the servers need to be restarted sometimes. When it is restarted, the problems are solved, but when it is restarted, the jobs running in the live environment on the robot are also blocked.
I couldn’t find any resources on the internet on how to do robot server maintenance with the best practices. What are you doing about the maintenance of the robot servers?
probably no rocket science but we schedule the maintenance monthly and on sundays. Sundays there are fewer processes and even fewer “live” processes.
In case you have no free time slots for maintenance. Then I suggest to block timeslots for maintenance and in that time deactivate the robot from the orchestrator. Therefore, the processes run on different robots or are aligned and run when the robot is available again.
But I am also interested in further best practices regarding robot maintenance!
Also check this documentation, maybe it is related to your Windows sessions: Windows Sessions
I Assume these are all unattended bots. Few things are required to check before deploy the bots.
- Windows OS version - 2016 and above
- Server Configuration - Minimum 2 Core CPU
- Assistant/Robot Version installed - 20.10.x and above
- Orchestrator Version - 20.10.x and above
All mentioned above should be UiPath recommendation as per the suggested pre-requisites.
In your case, you have already deployed.
Hence Check all the bots are running 24 hrs?
If no when the server will be free?
You Said Windows server-related problems could you please explain about exact issue you are facing.
Thanks for your answer, all the requirements are already provided by customer.
I mean sometimes windows taskbar frozen, sometimes we get cpu usage alert, sometimes there are only dark window and robot could not do any ui interaction on server.
All the robots are running for 7/24. We can create free slots for restart but, i want to know if its crucial, what is the best practices.
taskbar frozen cpu usage hits high these are showing server is not having required configuration or may be not utilized then allocated resources.
Usually bot tries to open huge size excel files or tries to edit in huge size file it will happen.
Request you to identify why it is getting struck and frozen. There might be bot is doing something which consumes all the available resources in server.
For time being create a bat file and schedule it that delete temp files every 24hrs.
Several approaches can be taken here. First, if you’re using windows server you should look into high density robots, this allows for you to use one machine for multiple robot windows sessions, reference: High-Density Robots. This way you can disable a machine in Orchestrator conduct your maintenance and bring it back without interruption.
There is also maintenance API calls to stop robots current job and place the Orchestrator into maintenance mode. This is a bigger “hammer” however, I have used it for environment upgrades etc. Reference: Maintenance mode
I also created a script to manage the maintenance mode reference marketplace here: Orchestrator Maintenance Mode - RPA Component | UiPath Marketplace
Let me know your thoughts.