bpf: fix flaky nat4_port_allocation_tcp/udp tests#45112
Draft
vipul-21 wants to merge 1 commit intocilium:mainfrom
Draft
bpf: fix flaky nat4_port_allocation_tcp/udp tests#45112vipul-21 wants to merge 1 commit intocilium:mainfrom
vipul-21 wants to merge 1 commit intocilium:mainfrom
Conversation
The SNAT port allocation stress tests were flaky due to two issues: 1. The test used get_prandom_u32() for destination address selection in the callback, adding unnecessary run-to-run variance unrelated to the algorithm under test. Replace it with a seeded xorshift32 PRNG that produces a deterministic but non-trivial distribution, while the algorithm's own get_prandom_u32() for port selection on collision remains the sole source of randomness being tested. 2. The 50% checkpoint assertion (retries_50percent[SNAT_COLLISION_RETRIES] < 15) was too tight. The algorithm's internal random port picks can occasionally produce slightly more early collisions than expected. Relax the threshold from 15 to 25 (still negligible: 0.019% of 131K iterations at the 50% mark). Fixes: cilium#42599 Signed-off-by: Vipul Singh <singhvipul@microsoft.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The SNAT port allocation stress tests were flaky due to two issues:
The test used get_prandom_u32() for destination address selection in the callback, adding unnecessary run-to-run variance unrelated to the algorithm under test. Replace it with a seeded xorshift32 PRNG that produces a deterministic but non-trivial distribution, while the algorithm's own get_prandom_u32() for port selection on collision remains the sole source of randomness being tested.
The 50% checkpoint assertion (retries_50percent[SNAT_COLLISION_RETRIES] < 15) was too tight. The algorithm's internal random port picks can occasionally produce slightly more early collisions than expected. Relax the threshold from 15 to 25 (still negligible: 0.019% of 131K iterations at the 50% mark).
Fixes: #42599