Frontend Observability Working Group

Attributes

Property	Value
Date Created	2021-07-14
End Date	TBD
Slack	#wg_frontend-observability (only accessible from within the company)
Google Doc	Working Group Agenda (only accessible from within the company)

Charter

This working group will co-ordinate the organization of the effort to develop a frontend observability framework for Gitlab and establish guidelines for its usage. There is currently no method to instrument the frontend for the purposes of monitoring performance, especially across a range of devices and content. Error monitoring is possible but is mostly unused. This group will explore, define, and coordinate the development of system to make this instrumentation possible and document recommendations for use.

Scope and Definitions

This group will focus on creating a method for sending browser performance data to a frontend stats gateway that will allow the metrics to be accessed via our system, like the current backend metrics. Determining the details of this is part of the remit. Once this exists, the group will be responsible for documenting its use, as well as usage recommendations for error monitoring. This will allow frontend engineers to understand the performance of their code in various contexts and add another source of data to help find issues when new features are rolled out.

Stakeholders for this project include frontend engineers and SREs.

This group will not focus on instrumentation around user actions for product reasons; this is covered by Snowplow and has a different use case.

Definitions

Frontend Observability describes the process of being able to answer questions about the performance of frontend code "in the wild". It is provided via the instrumentation of frontend Javascript. This is a nice talk covering what it can mean, although the goals of this group will be focused around performance and errors.

Exit Criteria

This working group will have fulfilled its purpose when:

The @sentry/browser package is upgraded to version 6.10.0 or higher
Sentry is configured to track and report frontend errors reliably
Sentry is configured to monitor our performance with performance tracing
We have decided what client performance data to send to Prometheus
The aforementioned client performance data is sent to and tracked in Prometheus
A triaging / diagnosing process is created for sentry errors
Documentation is written that allows other frontend engineers to contribute

Roles and Responsibilities

The functional leads will be responsible for:

Representing the needs of individual stakeholders in their department/sub-dept
Gathering and consolidating feedback on specific proposals from their department/sub-dept
Communicating the output from the working group (if any) and answering questions from their dept/sub-dept

Ideally the functional lead is someone who is an IC that might be affected by the policy put in place. but anyone capable of representing a department or sub-department in the fashion mentioned above is welcome.

Working Group Role	Person	Stakeholder Department	Title
Executive Sponsor	Tim Zallmann		Senior Director
Facilitator	Sam Beckham	Frontend:Verify	Frontend Engineering Manager
Functional Lead	Sarah Groff Hennigh-Palermo	Frontend:Verify::Pipeline Authoring	Senior FE Engineer
Functional Lead	Drew Cimino	Backend:Verify::Pipeline Execution	Backend Engineer
Functional Lead	Andrew Newdigate	Infrastructure	Distinguished Engineer
Member	José Vargas López	Frontend:Verify::Pipeline Execution	Senior FE Engineer
Member	Thomas Randolph	Frontend:Create::Code Review	Senior FE Engineer
Member	Marius Bobin	Backend:Verify::Pipeline Execution	Backend Engineer
Member	Miguel Rincon	Frontend:Verify::Runner	Senior FE Engineer
Member	Dov Herskhovitch	Product Manager:Verify::Pipeline Authoring	Senior Product Manager