feat: Add envvar for retrying git operations

go.mod

@@ -1,8 +1,9 @@
 module github.com/cert-manager/klone
 
-go 1.21.1
+go 1.23
 
 require (
+	github.com/cenkalti/backoff/v5 v5.0.1
 	github.com/rogpeppe/go-internal v1.11.0
 	github.com/spf13/cobra v1.7.0
 	gopkg.in/yaml.v3 v3.0.1

go.sum

@@ -1,3 +1,5 @@
+github.com/cenkalti/backoff/v5 v5.0.1 h1:kGZdCHH1+eW+Yd0wftimjMuhg9zidDvNF5aGdnkkb+U=
+github.com/cenkalti/backoff/v5 v5.0.1/go.mod h1:rkhZdG3JZukswDf7f0cwqPNk4K0sa+F97BxZthm/crw=
 github.com/cpuguy83/go-md2man/v2 v2.0.2/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
 github.com/inconshreveable/mousetrap v1.1.0 h1:wN+x4NVGpMsO7ErUn/mUI3vEoE6Jt13X2s0bqwp9tc8=
 github.com/inconshreveable/mousetrap v1.1.0/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=

pkg/cache/clone.go

@@ -18,6 +18,7 @@ func calculateCacheKey(src mod.KloneSource) string {
 }
 
 func getCacheDir() (string, error) {
+	// TODO: add centralized config management defining env vars + maybe a global config file for klone
 	if cacheDir := os.Getenv("KLONE_CACHE_DIR"); cacheDir != "" {
 		return filepath.Abs(filepath.Clean(cacheDir))
 	}

pkg/download/git/clone.go

@@ -8,6 +8,10 @@ import (
 	"os"
 	"os/exec"
 	"path/filepath"
+	"strconv"
+	"time"
+
+	"github.com/cenkalti/backoff/v5"
 
 	"github.com/cert-manager/klone/pkg/mod"
 )
@@ -24,26 +28,55 @@ func Get(ctx context.Context, targetPath string, src mod.KloneSource) (string, e
 	return filepath.Join(targetPath, src.RepoPath), nil
 }
 
+const gitRetryDelay = 5 * time.Second
+
 func runGitCmd(ctx context.Context, root string, stdout io.Writer, stderr io.Writer, args ...string) error {
-	cmd := exec.CommandContext(ctx, "git", args...)
+	do := func() (struct{}, error) {
+		// dummy return value to match the interface of backoff.Operation
+		ret := struct{}{}
 
-	cmd.Dir = root
-	cmd.Env = append(os.Environ(), cmd.Env...)
-	// Disable Git terminal prompts in case we're running with a tty
-	cmd.Env = append(cmd.Env, "GIT_TERMINAL_PROMPT=false")
+		cmd := exec.CommandContext(ctx, "git", args...)
 
-	cmd.Stdout = stdout
-	cmd.Stderr = stderr
+		cmd.Dir = root
+		cmd.Env = append(os.Environ(), cmd.Env...)
+		// Disable Git terminal prompts in case we're running with a tty
+		cmd.Env = append(cmd.Env, "GIT_TERMINAL_PROMPT=false")
 
-	if err := cmd.Start(); err != nil {
-		return err
+		cmd.Stdout = stdout
+		cmd.Stderr = stderr
+
+		if err := cmd.Start(); err != nil {
+			return ret, err
+		}
+
+		if err := cmd.Wait(); err != nil {
+			return ret, fmt.Errorf("git command failed: %v", err)
+		}
+
+		return ret, nil
 	}
 
-	if err := cmd.Wait(); err != nil {
-		return fmt.Errorf("git command failed: %v", err)
+	_, err := backoff.Retry(ctx, do, backoff.WithMaxTries(getRetryCount()), backoff.WithBackOff(backoff.NewConstantBackOff(gitRetryDelay)))
+	return err
+}
+
+func getRetryCount() uint {
+	// TODO: add centralized config management defining env vars + maybe a global config file for klone
+	retryCountRaw := os.Getenv("KLONE_GIT_RETRY_ATTEMPTS")
+	if retryCountRaw == "" {
+		return 1
 	}
 
-	return nil
+	retryCount, err := strconv.Atoi(retryCountRaw)
+	if err != nil {
+		return 1
+	}
+
+	if retryCount <= 0 {
+		return 1
+	}
+
+	return uint(retryCount)
 }
 
 func sparseCheckout(ctx context.Context, root string, repoURL string, branch string, patterns []string) error {