fix: stop racing the daemon when removing containers (#1093)

Containers set `HostConfig.AutoRemove` but act also removes them explicitly, so the two removers race and the loser logs a 409 `removal of container X is already in progress` — seen at the end of nearly every `uses: docker://` step.

The explicit remove is redundant for `docker://` steps and docker actions (`Start(true)` already awaited exit), so it's skipped. Job and service containers keep both removers — their `sleep` entrypoint needs `AutoRemove` as a fallback reaper — so there the race is inherent and `remove()` now treats `NotFound` and `Conflict` as success.

---------

Co-authored-by: bircni <bircni@icloud.com>
Reviewed-on: https://gitea.com/gitea/runner/pulls/1093
Reviewed-by: bircni <bircni@icloud.com>
Co-authored-by: silverwind <me@silverwind.io>
This commit is contained in:
silverwind
2026-07-21 11:03:15 +00:00
committed by silverwind
parent 0c08b0f2da
commit 89467c9dd0
6 changed files with 124 additions and 3 deletions

View File

@@ -16,6 +16,7 @@ import (
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/mock"
"github.com/stretchr/testify/require"
)
func TestStepDockerMain(t *testing.T) {
@@ -118,6 +119,43 @@ func TestStepDockerMain(t *testing.T) {
cm.AssertExpectations(t)
}
// With AutoRemove the daemon reaps the container on exit, so act must not remove it afterwards.
func TestStepDockerAutoRemove(t *testing.T) {
orig := ContainerNewContainer
defer func() { ContainerNewContainer = orig }()
for _, tc := range []struct {
autoRemove bool
removes int
}{
{false, 2}, // stale + post-run
{true, 1}, // post-run skipped
} {
cm := &containerMock{}
ContainerNewContainer = func(*container.NewContainerInput) container.ExecutionsEnvironment { return cm }
sd := &stepDocker{
RunContext: &RunContext{
Config: &Config{AutoRemove: tc.autoRemove},
Run: &model.Run{JobID: "1", Workflow: &model.Workflow{Jobs: map[string]*model.Job{"1": {}}}},
JobContainer: cm,
},
Step: &model.Step{ID: "1", Uses: "docker://node:14"},
}
removes := 0
cm.On("Pull", false).Return(func(context.Context) error { return nil })
cm.On("Remove").Return(func(context.Context) error { removes++; return nil })
cm.On("Create", []string(nil), []string(nil)).Return(func(context.Context) error { return nil })
cm.On("Start", true).Return(func(context.Context) error { return nil })
cm.On("Close").Return(func(context.Context) error { return nil })
require.NoError(t, sd.runUsesContainer()(context.Background()))
cm.AssertExpectations(t)
assert.Equal(t, tc.removes, removes)
}
}
func TestStepDockerNewStepContainerAllocatePTY(t *testing.T) {
for _, tc := range []struct {
name string