Are you collecting the right metrics?
Tuesday, October 22 at 13:30–17:00
Of course, it is important to collect metrics on your database clusters. Often, it isn't easy to understand how best mandatory, important and optional metrics should be split apart and handled. Depending on PostgreSQL cluster architectures, sometimes metrics that are typically useless can become of high or even critical importance.
We will discuss available metrics at Linux, container and PostgreSQL levels and sort out the usage of these metrics. This will help in choosing what to monitor, and why, along with setting up alerting thresholds and workflows. We will additionally cover the modification of PostgreSQL resource allocations using default cgroups mechanism.
Attendees will gain:
Attendees will gain the knowledge of important metrics to collect to ensure critical events are never missed. At the end of the session, users will be able to understand and follow how their system behaves. This will be to optimize observability at the system level as well as at the application or project level.
Technical requirements
No previous knowledge is required.
If you'd like to participate in the session, please bring your laptop to use as a playground with PostgreSQL installed, or an ssh client like OpenSSH and/or psql already installed.
