Repeated Filter Effects Query Hashing

[landenai]

Environment

SaaS (https://sentry.io/)

Steps to Reproduce

Customer has two N+1 performance issues that appear almost identical, except the preceding DB span has more repetitions of the same filter thus influencing the grouping and causing confusion.

Expected Result

Repeating filters do not impact query hashing

Actual Result

Repeating filters cause seemingly identical N+1 issues to be grouped separately

Product Area

Unknown

Link

No response

DSN

No response

Version

No response

[linear]

ENG-5438 Repeated Filter Effects Query Hashing

[mrduncan]

This is working as initially designed however there are cases where an ORM may apply a repeated where clause which would be nice to account for in our fingerprinting, for example the following would group together:

WHERE (id = ? AND status = ?) OR (id = ? AND status = ?)

WHERE (id = ? AND status = ?) OR (id = ? AND status = ?) OR (id = ? AND status = ?)

In the mean time, these detectors can be disabled: https://docs.sentry.io/product/issues/issue-details/performance-issues/