AIVA Automated Testing System developed by Y Soft Corporation is a cloud software solution designed for end-to-end testing of web applications and is operated as a cloud platform for anyone to use. The system already has a strong observability layer that was recently extended to gather frontend telemetry. However, this telemetry data is currently utilized primarily for internal debugging and infrastructure monitoring rather than for communicating system health to end users. AIVA is now preparing to launch a SaaS application and is interested in having a user-facing status page.

Status page apps (e.g., statuspage.io) are a well-known commodity. However, most status page applications focus on utilizing back-end telemetry, smoke tests or synthetic monitoring and reflect the status and availability of backend services. AIVA always had a philosophy of testing applications from the end-user perspective and we would like to apply this philosophy to the status page, i.e. show the system status from the perspective of real users.

The objective of this thesis is to utilize existing frontend telemetry (reflecting the real user perspective) to interpret the health status of AIVA’s key components via a simple user-facing status page, taking into account the entire end to end infrastructure and recognizing that different users may have different experience with AIVA, depending on their location, connectivity and other factors.

The thesis shall answer the following questions:

1. What are the limits of the current status page implementation, with respect to back-end telemetry, synthetic monitoring or smoke tests as observed by the end users? For example, when a system status is reported as fully working (often indicated by a green color), does that really mean that any user connecting from any location can access such system and have good-enough user experience?
2. Is front-end telemetry as implemented in AIVA sufficient to implement user-facing status page? What changes had to implemented in the AIVA front-end telemetry to implement a user-facing status page?
3. In general, is it feasible to implement a status page solely using front-end telemetry for a SaaS system? What are the key pre-requisites, enablers, design principles and tools need to be applied to enable this approach in a general case?
4. How can the functionality of such status page be reliably tested?
5. How can such status page implementation address security and integrity concerns, such as fake or unreliable telemetry?
6. What are the next steps for AIVA to bring such status page implementation into production?

The answers to these questions shall be well explained and justified, based on experiments and implementation (proof of concept) prototypes. Notably, it is expected for this thesis to contain a simple implementation of an AIVA status page taking into account user experiences from a global community of users.