testutils/lint: TestLint failed

[cockroach-teamcity]

testutils/lint.TestLint failed with artifacts on master @ 5c0af43059fba81d2f146923f724694c106cc5d9:

=== RUN   TestLint
--- FAIL: TestLint (1089.35s)
=== RUN   TestLint/TestStaticCheck
    lint_test.go:1773: 
        pkg/sql/copy.go:433:8: const copyBatchRowSize is unused (U1000)
    --- FAIL: TestLint/TestStaticCheck (884.23s)

Parameters: GOFLAGS=-json

Help

See also: How To Investigate a Go Test Failure (internal)

/cc @cockroachdb/test-eng

This test on roachdash | Improve this report!

Jira issue: CRDB-19038

[cucaroach]

I'm so confused, that line of code doesn't contain that variable on master but it does at that SHA. I have no idea what to do with this...

[srosenberg]

It was trigged by this build [1] manually? The changes are not yet in master.

[1] https://teamcity.cockroachdb.com/buildConfiguration/Cockroach_UnitTests/6246057

[cucaroach]

Still confused, that commit even has old/wrong commit notes:

    Release note (sql change): COPY FROM operations are now atomic by
    default instead of being segmented into 100 row transactions. Setting
    the sql.copyfrom.atomic cluster setting to false to get the old
    behavior.

On master its:

    Release note (backward-incompatible change): COPY FROM operations are
    now atomic by default instead of being segmented into 100 row
    transactions. Set the copy_from_atomic_enabled session setting to
    false to get the old behavior.

Its like bors constructed a merge node with an older version of my PR or something?

[srosenberg]

Its like bors constructed a merge node with an older version of my PR or something?

That's likely what happened. Bors uses a workaround [1] which first creates staging.tmp (with master and PRs), then copies it over to staging. Technically, staging.tmp shouldn't trigger CI, but it does sometimes [2]. It might be worth checking with #dev-inf if this is a known issue.

[1] https://forum.bors.tech/t/what-is-the-staging-tmp-branch-used-for/65
[2] https://teamcity.cockroachdb.com/buildConfiguration/Cockroach_UnitTests?branch=staging.tmp

[yuzefovich]

This was fixed in #85704.

[cucaroach]

Okay so one mystery solved, thanks @yuzefovich, although gotta wonder what's up with the linter and why didn't it fire on my PR?