Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix issues with several TAP tests - Closes #4830 #4881

Merged
merged 9 commits into from
Mar 24, 2025
Merged

Conversation

JavierJF
Copy link
Collaborator

This PR fixes some consistency issues or design problems found with the following tests:

  • set_character_set-t
  • test_backend_conn_ping-t
  • set_testing-240-t
  • test_utf8mb4_as_ci-4841-t
  • test_backend_conn_ping-t
  • test_cacert_load_and_verify_duration-t
  • mysql-protocol_compression_level-t

It also addresses the request for adding warning messages on #4830.

renecannao and others added 9 commits March 21, 2025 12:40
Number of connections being maintained by the test now assumes
(number_of_intervals - 1). This spare interval decreases the number of
connections actively maintained and should make the test more resilient
in to small network stalls.
Test requires a MySQL 8 instance, this is now parametrized via
TAP_MYSQL8_BACKEND_HG env variable as in other TAP tests.
Also adds a status variable exporting the OpenSSL version in use.
- Deterministic test table: Invalid limits and tables with variable
  rows ('information_schema') were used to create the original table
  size, this is now replaced with a deterministic table.
- Compression overhead measurement: Previous test design generated a
  very large resultset (> 20Gb), performing a single measurement per
  resultset. This resulted in very inaccurate measurements, with radical
  variations between different systems. A resultset of the previous size
  can interfere in multiple ways with measurements, like network
  saturation, bufferbloat, and queuing delays at application level, all
  things that are out of the scope of the test measurement. This have
  been reworked into multiple smaller packets (40Mb) big enough so that
  the increased latency of compression can be measured and averaged.
- Fixed test checks: Previous test checks were meaningless, performing
  only a non-zero comparison between performance differences. This gave
  no insights about ProxySQL vs MySQL differences or allowed for simple
  coherence checks (non-compress time < compress time). All these checks
  have new confidence ranges based on experimental values.
- Improved general logging based on other TAP tests conventions and
  refactored most of the functions.
@JavierJF JavierJF force-pushed the v3.0-ci_fixes_172 branch from dd21b47 to 1aaa16a Compare March 21, 2025 18:47
@renecannao renecannao merged commit a876b1f into v3.0 Mar 24, 2025
6 of 7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants