fix: stop leaking per-job docker networks (#1124)

Leaked per-job networks each hold a subnet of the daemon's address pool and nothing reclaimed them, so a host eventually fails every job with `all predefined address pools have been fully subnetted`. This is what CI hit in https://gitea.com/gitea/runner/actions/runs/742177.

- teardown no longer loses a container, and its network with it: a failed `ContainerRemove` was reported as success, a container whose id was never learned was skipped, and the daemon's own `AutoRemove` teardown was raced
- the idle cleanup reclaims what teardown cannot: networks carry `com.gitea.runner.uuid`, so a runner only touches its own, and a cutoff keeps a job starting during the pass out of scope
- pull failures reported mid-stream were discarded, surfacing later as a confusing `No such image`; they now propagate, and fall back to a local copy instead of failing the job
- `NetworkCreate` retries pool exhaustion briefly, then says which knobs to turn
- digest-pinned images are not re-pulled, and removal kills first so it never waits out Podman's stop timeout (measured 10.1s → 0.09s per container)

---------

Co-authored-by: bircni <bircni@icloud.com>
Reviewed-on: https://gitea.com/gitea/runner/pulls/1124
Reviewed-by: bircni <bircni@icloud.com>
This commit is contained in:
silverwind
2026-07-30 16:25:47 +00:00
parent da4037899a
commit b7a3bf98bc
14 changed files with 395 additions and 64 deletions

View File

@@ -6,11 +6,15 @@ package container
import (
"context"
"io"
"strings"
"testing"
"github.com/docker/cli/cli/config"
log "github.com/sirupsen/logrus"
"github.com/sirupsen/logrus/hooks/test"
assert "github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func init() {
@@ -65,3 +69,21 @@ func TestGetImagePullOptions(t *testing.T) {
assert.NoError(t, err, "Failed to create ImagePullOptions") //nolint:testifylint // pre-existing issue from nektos/act
assert.Equal(t, "eyJ1c2VybmFtZSI6InVzZXJuYW1lIiwicGFzc3dvcmQiOiJwYXNzd29yZFxuIiwic2VydmVyYWRkcmVzcyI6Imh0dHBzOi8vaW5kZXguZG9ja2VyLmlvL3YxLyJ9", options.RegistryAuth, "RegistryAuth should be taken from local docker config")
}
// A digest-pinned image is immutable, so its local copy is always current.
func TestIsPinnedImage(t *testing.T) {
assert.True(t, isPinnedImage("alpine@sha256:28bd5fe8b56d1bd048e5babf5b10710ebe0bae67db86916198a6eec434943f8b"))
assert.False(t, isPinnedImage("alpine:latest"))
}
// The pull path reports a failure the daemon sent mid-stream, so it must carry the reason
// whichever of the two shapes the daemon used.
func TestLogDockerResponseError(t *testing.T) {
logger, _ := test.NewNullLogger()
streamErr := func(line string) error {
return logDockerResponse(logger, io.NopCloser(strings.NewReader(line)), false)
}
require.EqualError(t, streamErr(`{"error":"toomanyrequests: rate limit exceeded"}`), "toomanyrequests: rate limit exceeded")
require.EqualError(t, streamErr(`{"errorDetail":{"message":"unexpected EOF"}}`), "unexpected EOF")
require.NoError(t, streamErr(`{"status":"Downloading"}`))
}