The query section
This section is all about tuning the inner workings of the query engine. Some are fairly straightforward to understand, such as how long a given query can run before being aborted (--query.timeout), or how many queries can run simultaneously (--query.max-concurrency).
However, two of them set limits that can have non-obvious consequences. The first is --query.max-samples, which was introduced in Prometheus 2.5.0, that sets the maximum number of samples that can be loaded onto memory. This was done as a way of capping the maximum memory the query subsystem can use (by using it together with --query.max-concurrency) to try and prevent the dreaded query-of-death—a query that loaded so much data to memory that it made Prometheus ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access