TestQueriesGoToTheResolverSWWDNSBLResolverNames, one of the issue's six, now calls t.Parallel(). It queries a UDP stand-in outside any synctest bubble. Go starts a parallel test only after every test that is not parallel has ended, so it cannot overlap the bubbled DNSBL tests. No other parallel test of the reputation package queries a resolver. The comment at the top of internal/reputation/dnsbl_test.go says so.
The other five stay one at a time. So do ten DNSBL tests in synctest bubbles that the issue's scan missed. Each already gives its reason in a comment:
the two in internal/config/viper_test.go set environment variables, which Go refuses in a parallel test;
TestCorazaWritesNoFile sets TMPDIR for the whole process;
the inspection-time test compares processor time with reference work, which parallel tests would disturb;
TestASwitchNotPassedOnClosesTheAppsConnection would have a leaked connection closed by garbage collection that other tests start;
the bubbled DNSBL tests share Go's resolver, which fails when two bubbles query at once.
Every subtest inside a parallel test already calls t.Parallel(). The only subtests that do not are inside the inspection-time test and a bubbled DNSBL test, both one at a time.
Judgement call: I left the existing reason comments as they are. Each sits at its test or points to the comment at the top of its file.
Model: opus-5-5
closes https://git.eeqj.de/sneak/smallwebwaf/issues/406
`TestQueriesGoToTheResolverSWWDNSBLResolverNames`, one of the issue's six, now calls `t.Parallel()`. It queries a UDP stand-in outside any synctest bubble. Go starts a parallel test only after every test that is not parallel has ended, so it cannot overlap the bubbled DNSBL tests. No other parallel test of the `reputation` package queries a resolver. The comment at the top of `internal/reputation/dnsbl_test.go` says so.
The other five stay one at a time. So do ten DNSBL tests in synctest bubbles that the issue's scan missed. Each already gives its reason in a comment:
- the two in `internal/config/viper_test.go` set environment variables, which Go refuses in a parallel test;
- `TestCorazaWritesNoFile` sets `TMPDIR` for the whole process;
- the inspection-time test compares processor time with reference work, which parallel tests would disturb;
- `TestASwitchNotPassedOnClosesTheAppsConnection` would have a leaked connection closed by garbage collection that other tests start;
- the bubbled DNSBL tests share Go's resolver, which fails when two bubbles query at once.
Every subtest inside a parallel test already calls `t.Parallel()`. The only subtests that do not are inside the inspection-time test and a bubbled DNSBL test, both one at a time.
Judgement call: I left the existing reason comments as they are. Each sits at its test or points to the comment at the top of its file.
Model: opus-5-5
TestQueriesGoToTheResolverSWWDNSBLResolverNames queries a real UDP
stand-in outside any synctest bubble. A parallel test starts only once
every test that is not parallel has ended, so it can no longer overlap
the bubbled DNSBL tests, and no other parallel test of the package
queries a resolver.
The other tests that run one at a time stay so, each for the reason its
comment gives: environment variables, TMPDIR, timing, Go's resolver in
synctest bubbles, and garbage collection hiding a leaked connection.
Every subtest inside a parallel test already calls t.Parallel().
Model: opus-5-5
clawbot
self-assigned this 2026-11-17 06:32:24 +01:00
The resolver test now runs in parallel without overlapping the bubbled DNSBL tests, and every test that stays one at a time has a valid reason in a comment, which meets the definition of done of #406.
Model: opus-5-5
Verdict: PASS
Head: c125375ac7aab3b007ecbd6204109788ef3a8d89
The resolver test now runs in parallel without overlapping the bubbled DNSBL tests, and every test that stays one at a time has a valid reason in a comment, which meets the definition of done of https://git.eeqj.de/sneak/smallwebwaf/issues/406.
Model: opus-5-5
clawbot
merged commit c0d9d5787e into next2026-11-17 06:57:45 +01:00
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
closes #406
TestQueriesGoToTheResolverSWWDNSBLResolverNames, one of the issue's six, now callst.Parallel(). It queries a UDP stand-in outside any synctest bubble. Go starts a parallel test only after every test that is not parallel has ended, so it cannot overlap the bubbled DNSBL tests. No other parallel test of thereputationpackage queries a resolver. The comment at the top ofinternal/reputation/dnsbl_test.gosays so.The other five stay one at a time. So do ten DNSBL tests in synctest bubbles that the issue's scan missed. Each already gives its reason in a comment:
internal/config/viper_test.goset environment variables, which Go refuses in a parallel test;TestCorazaWritesNoFilesetsTMPDIRfor the whole process;TestASwitchNotPassedOnClosesTheAppsConnectionwould have a leaked connection closed by garbage collection that other tests start;Every subtest inside a parallel test already calls
t.Parallel(). The only subtests that do not are inside the inspection-time test and a bubbled DNSBL test, both one at a time.Judgement call: I left the existing reason comments as they are. Each sits at its test or points to the comment at the top of its file.
Model: opus-5-5
Verdict: PASS
Head:
c125375ac7The resolver test now runs in parallel without overlapping the bubbled DNSBL tests, and every test that stays one at a time has a valid reason in a comment, which meets the definition of done of #406.
Model: opus-5-5