fix: bound blocking calls and stop failing silently (#1174)

Jobs occasionally go silent ([example](https://gitea.com/gitea/runner/actions/runs/805045/jobs/1055123)) mid-run and Gitea reaped them after `ZOMBIE_TASK_TIMEOUT`, with no error in the log. This contains a number of related fixes, all with full test coverage:

1. Bound every RPC to Gitea with a timeout, a stalled report otherwise parked logs and heartbeats for the whole job.
2. Cap `runner.fetch_timeout` at that ceiling.
3. Let only the daemon loop close its own channel, the race panicked the process.
4. Stop the job on any terminal server result, not just `RESULT_CANCELLED`.
5. Report that result instead of relabelling it as cancelled.
6. Log reporting failures once at each end of an outage instead of discarding them.
7. Clamp the acknowledged log index, a too-large ack panicked on a slice bound.
8. Stop reading server health from a `FetchTask` deadline, it marked the runner healthy and reset the error backoff on a timeout.
9. Return an error from the Docker version probe instead of a `logrus` panic.
10. Pass the context to go-git's fetch and pull.
11. Fail the clone when a refresh dies on a cancelled context.
12. Set `terminationGracePeriodSeconds` in the Kubernetes examples.

Also contains a deprecation fix for goreleaser.

Reviewed-on: https://gitea.com/gitea/runner/pulls/1174
Reviewed-by: bircni <bircni@icloud.com>
Co-authored-by: silverwind <me@silverwind.io>
This commit is contained in:
silverwind
2026-08-19 11:13:21 +00:00
committed by silverwind
parent 3f70822458
commit bc8161c673
21 changed files with 422 additions and 63 deletions
+7 -5
View File
@@ -17,7 +17,7 @@ import (
"gitea.dev/actionslib/runner/v1/runnerv1connect"
)
func getHTTPClient(endpoint string, insecure bool) *http.Client {
func getHTTPClient(endpoint string, insecure bool, timeout time.Duration) *http.Client {
transport := &http.Transport{
Proxy: http.ProxyFromEnvironment,
MaxIdleConns: 10,
@@ -29,11 +29,13 @@ func getHTTPClient(endpoint string, insecure bool) *http.Client {
InsecureSkipVerify: true,
}
}
return &http.Client{Transport: transport}
return &http.Client{Transport: transport, Timeout: timeout}
}
// New returns a new runner client.
func New(endpoint string, insecure bool, uuid, token string, opts ...connect.ClientOption) *HTTPClient {
// New returns a new runner client. timeout bounds every RPC: without it a
// stalled connection parks the reporter for the whole job context, so logs and
// heartbeats stop together and the task is reaped as a zombie.
func New(endpoint string, insecure bool, uuid, token string, timeout time.Duration, opts ...connect.ClientOption) *HTTPClient {
baseURL := strings.TrimRight(endpoint, "/") + "/api/actions"
opts = append(opts, connect.WithInterceptors(connect.UnaryInterceptorFunc(func(next connect.UnaryFunc) connect.UnaryFunc {
@@ -49,7 +51,7 @@ func New(endpoint string, insecure bool, uuid, token string, opts ...connect.Cli
}
})))
httpClient := getHTTPClient(endpoint, insecure)
httpClient := getHTTPClient(endpoint, insecure, timeout)
return &HTTPClient{
PingServiceClient: pingv1connect.NewPingServiceClient(
httpClient,