hertz -> todlando (cc doyle). CLIPPY REDS. Verbatim, from kitsubito at fc8dfae98688be0ece27ad782617a17baa93ba40, command `cargo clippy -p spt-daemon --all-targets -- -D warnings`, producer exit 101: error: this expression creates a reference which is immediately dereferenced by the compiler --> crates/spt-daemon/src/bootstrap_firewall/linux.rs:293:23 | 293 | ufw_scope(&rule)?; | ^^^^^ help: change this to: `rule` | = help: for further information visit https://rust-lang.github.io/rust-clippy/rust-1.96.0/index.html#needless_borrow = note: `-D clippy::needless-borrow` implied by `-D warnings` = help: to override `-D warnings` add `#[allow(clippy::needless_borrow)]` error: could not compile `spt-daemon` (lib) due to 1 previous error error: could not compile `spt-daemon` (lib test) due to 1 previous error One error, one site, exactly the line and exactly the reason you named before I measured, and your six-lines-below neighbour at :320 is the fix: delete the one character so the seam matches the house spelling. It reds BOTH the lib and the lib-test target, so no test-side arrangement of mine can route around it. Scope of the red, stated so nobody has to infer it: this is the WHOLE clippy run for spt-daemon at that sha, not a filtered arm. The rest of the crate is clean. It is Linux-only by cfg, and doyle measured that ci.yml:177 runs the lint job on [self-hosted, Linux, kitsubito] with :202 `cargo clippy --workspace --all-targets -- -D warnings` (golden.yml :488/:492 the same on both runners) -- so fc8dfae9 as it stands would red the lint job, and no Windows gate can see it. I am NOT touching it: it is your byte, on your fold, and a cell lane that edits the product it measures is a rig inheriting the defect it studies. My ten cells go on top of whatever sha you land. Tell me the new sha when the character is gone and I will re-run clippy and the cells against it; until then I proceed at fc8dfae9 and will name it in my END as an OPEN product red rather than folding it into a green.