Repository navigation
fix(ci): run the fuzzer in fork mode so a known crash doesn't end the run - #32
Merged
Merged
Conversation
… run UBSAN_OPTIONS=halt_on_error=1 kills the process on the first sanitizer hit, so a single easily-rediscovered third-party bug was ending nearly every nightly run in seconds instead of the intended 30-minute budget. -fork=1 -ignore_crashes/-ignore_ooms/-ignore_timeouts respawns a fresh worker after a crash instead of dying, so real fuzzing can continue. The classification step now runs unconditionally (fuzz itself now exits 0 even when it recorded crashes) and only fails the job if one of the collected crash artifacts actually implicates our own code.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 nightly fuzz job has effectively never run for its intended 30-minute budget:
UBSAN_OPTIONS=halt_on_error=1kills the whole process on the first sanitizer hit, and the corpus has settled on an input that trivially rediscovers the same known third-party bug (third_party/libjpeg/jdhuff.c:529, an integer-overflow) almost every single run, ending it in seconds.Switches to libFuzzer's fork mode (
-fork=1 -ignore_crashes=1 -ignore_ooms=1 -ignore_timeouts=1): a crashing worker gets respawned instead of taking down the whole run, so real fuzzing continues for the full time budget. The trade-off: the fuzz binary itself now exits 0 even when it collected crashes, so the classification step (only fail the job if a crash implicates our own code, not a vendored dependency) now runs unconditionally instead of being gated on the fuzz step failing.