- On June 23, 2021
We can all agree that robots are the future. Why would someone do a repetitive task when they can make a robot do it? Wouldn’t that make life much easier? Well, that is exactly the kind decision we made. Whenever we talk about automation, we think about minimal human intervention, and how fast it is. It would be ideal for our day-to-day life and would allow us to reach our goals efficiently and quickly as well.
It is imperative to mention the fact that in our field, automation would save a lot of time and get things done faster yet efficiently. As TechOps and Support engineers, we decided to automate some of the tickets that we receive daily. In our scope of work, the number of tickets that are opened every day is increasing. However, that does not mean that they are all new. The tickets get repeated but with different cases. So why would we spend so much time repeating the work, when a simple robot can do it with a click of a button? This is why automation would be of great use to us. In other words, we can look into the roles of Robotic Process Automation (RPA):
• It automates everyday processes.
• It is a digital assistant that does routine tasks.
• It boosts efficiency by saving time.
• It records the tasks and performs them.
Nonetheless, is this enough to choose automation over human involvement when handling repetitive tickets? There are a few advantages to RPA tools that proved to show that we should pursue this option and automate tickets. These advantages included:
• Multiple processes can be automated at once.
• Clearing the simple tasks that eat up part of every worker’s day Easy to model, scale and deploy.
• No training period is required, unlike machine learning programs.
In summary, RPA tools automate repetitive tasks, that can save up human efforts and involvement which is exactly why we need automation in our day-to-day work experience.
Now that we established why automation is a good option, we evaluated several RPA tools and proceeded with Robocorp as its key features and advantages were useful with the tickets we are dealing with. It is simple and user-friendly and that is shown in its coding options whether we can use Python or Language Server Protocol for robot framework (LSP).
How we used RPA
The diagram summarizes the self-healing system that goes in the following cycle: the monitoring tool detects an issue, sends an alert to the user and if the issue has a corresponding automated task, it will get triggered. If not, then the user would have to interfere in order to solve it.
We chose a few common tickets to automate using RPA and Python scripting. We would go over the scenario of the resolution of the ticket in order to reproduce it using RPA. For instance, if we needed to open a certain browser with a link, we would do so using RPA. Any passwords used would be encrypted in a vault and accessed through the code so that there would be confidentiality within the process of resolution. In order to move from one step to another and implement UI web-based navigations, we would use screen scraping/pictures. We would take a screenshot of the item that needs to be clicked on and use it as a legend to compare the items on the screen. Once it is located, we move the mouse to its coordinates and click on it. We were able to connect to VPNs and access remote machines as well as execute bash scripts and database queries. There were calculations and comparisons done on excel sheets that would save us a lot of time as well.
For instance, there was a task where we needed to compare records from the database. The scenario was as follows: we connected to a VPN and accessed the machine using a remote desktop. As previously mentioned, the passwords are located in a virtual vault and accessed by code. Once connected, we located the database icon, opened it, and selected the required schema to run the necessary queries. We then copied the output into an excel sheet and conducted the comparisons. These calculations allowed us to reach the conclusion with a simple output sentence. All we needed to do is run the RPA flow and wait for the final output to confirm success or failure. Based on this other RPA flows were triggered to complete the remaining activities.
The automated tasks leave no room for human error as we wrote the RPA scripts and tested it out, which proved to be efficient.
With RPA, we have automated highly repetitive tasks. Automation has simplified all our common tickets. Some of the tickets would take us 20 minutes manually, but when automated, with a click of a button, it would take only 1-2 minutes to execute. This means that not only is it efficient, but also time saving. We would run the robot and work on something else in the meantime. This allowed us to shift our focus on gaining more knowledge and working on preventive SRE activities in order not to consume effort on tedious ticket handling tasks.