Repository navigation
rust: point the teardown-SIGSEGV retry at the renamed p2p test - #24263
Merged
Merged
Conversation
…de tests The #20973 stopgap retries a test that dies with SIGSEGV during teardown, after its body has passed. Its filter named `p2p_version::peers_negotiate_eth_69` exactly. #23046 renamed that test to `peers_negotiate_eth_71`, so the override has matched no test since then: a crash fails the job again, with no retry and no Slack alert. The same crash has now been seen on a second test of the same binary, `invalid_post_exec::test_buffered_malformed_post_exec_child_is_invalid`. Both tests start two nodes in one process with `setup(2)`. #20973 scoped the crash to the p2p test because it was then the only such test; #23186 added two more. The override now covers the three two-node tests of `reth-optimism-node::it`: - The p2p test, by an anchored regex on the eth version. The next version bump renames the test again and no longer disables the retry. - The two `invalid_post_exec` tests, by exact name. One has crashed. The other builds its two nodes the same way and has no crash on record. The crash has not been seen on a one-node test, so those tests stay under the no-retries policy. The two `invalid_post_exec` tests now install the fatal-signal handler, as the p2p test does. A retried crash then leaves the faulting stack in the JUnit report, which is what the Slack alert points to. The CircleCI flaky-retry step is unchanged. It greps the JUnit report for `<flaky` and reads the test names from the report. Only two YAML comments that described the old single-test scope change. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Contributor
Agent approval checkHead branch claude/keen-lehmann-9a3089 matches agent prefix claude/ Current head: |
…seen The override also covered `test_lagoon_buffered_child_with_wrong_post_exec_block_number_is_invalid`, which starts two nodes like the two tests that crashed but has no crash on record. A retry for it stretches a stopgap that is an exception to the no-retries policy. The override now covers a test only after the crash has been seen on it. That test keeps the fatal-signal handler. A crash there fails the job and leaves the faulting stack, which is the evidence needed to add the test to the override. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Keep only the change that restores the #20973 stopgap: the override filter names `peers_negotiate_eth_71`, the name the test has had since #23046. Take out the rest of this branch: the second test in the override, the crash handler in the `invalid_post_exec` tests, and the comment edits in `nextest.toml` and `rust-ci.yml`. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
teddyknox
marked this pull request as ready for review
November 18, 2026 20:35
sebastianst
approved these changes
Nov 18, 2026
teddyknox
enabled auto-merge
November 18, 2026 20:52
joshklop
approved these changes
Nov 18, 2026
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
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.
One line switches a merge-queue safeguard back on. It has been silently off since 2026-10-16.
reth-optimism-nodeintegration tests sometimes segfault at process exit, after the test has passed (#20973).op-reth-integration-testsis a required check, so a crash ejects whichever unrelated PR is in the merge queue. Before there was a mitigation, the crash recurred five times in about two months and repeatedly ejected PRs from the queue.#21909 added the mitigation: nextest retries the crashing test, and a Slack alert reports each retry with the stack. The retry selects the test by exact name,
peers_negotiate_eth_69. #23046 renamed the test topeers_negotiate_eth_71. nextest does not report a filter that matches nothing, so no check failed and nobody noticed.developtoday: the filter matches 0 tests. The next crash fails the required check, and the Slack alert cannot fire.p2p_version::peers_negotiate_eth_71. nextest retries a crash in a fresh process, the job stays green, and the alert fires.The crash is still live: the same signature was seen this month in an RWX shadow run of the suite, on another test of the same binary.
The change is one line of test-runner configuration. It has no effect on users or on product code.
Left out on purpose, to keep this to one line:
peers_negotiate_eth_69.Related: #20973
Review notes:
rust-code-reviewerandci-config-reviewerran on an earlier, wider version of this branch and left no open finding. They did not run on this one-line diff.🤖 Generated with Claude Code