Repository navigation
test: deflake probes that fail only on master merges - #973
Merged
Merged
Conversation
The same-slot txid search stopped after half a million draws of a 16-bit mix, which misses about once in two thousand runs. The RPC work-queue check raced a fast batch and sometimes saw only HTTP 200. A live follow session never counted an unknown BIP324 short id, so peer_session_with sat on the CRAP gate and coverage jitter failed the master merge.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Master merges failed after green PR CI in three places that are still in the tree:
tiponly_same_page_foreigner_records_probe_diagsearches for a txid whose mix shares a tiny head's page and in-page slot (16 bits, 1 in 65536). Half a million draws miss about once in 2000 runs. The search now stops at the first hit, with a cap many times the mean. This failed the test job on the merge of docs: queue rust-bitcoin gaps by who they hit #947.rpc_work_queue_exceededfired a 256-callgetblockcountbatch and hoped a parallel probe would overlap it. On a fast run every status was 200. The check now holds the only permit until the probe has seen HTTP 503.peer_session_withscored CRAP 30.7 on the merge of store: keep block filters and sp tweaks on the cold volume #970 (coverage 70.07%) after the same tree passed on the PR. A live follow session now sends Core's unrecognized short id, checks*other*accounting, and stays connected.The Esplora
Address already in usefailures on the merges of #931 and #934 were before listeners moved to port 0. No later master test job failed that way.Local fmt, clippy, deny, and ast-grep passed. The workspace suite is left to GitHub CI.
Test plan
cargo test -p rbitcoin-store --lib tiponly_same_page_foreigner_records_probe_diagcargo test -p rbitcoin-rpc --lib rpc_work_queue_exceededcargo test -p rbitcoin-net --lib unknown_v2_short_id_counts_other_and_staysqc,test,windows,macos,coverage)