Skip to content

fix(monitor): skip dirs without underscore in Update instead of panicking #2371

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
hami-robot merged 1 commit into from
Aug 6, 2026
Merged

fix(monitor): skip dirs without underscore in Update instead of panicking #2371

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 7 additions & 3 deletions pkg/monitor/nvidia/cudevshr.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -185,8 +185,13 @@ func (l *ContainerLister) Update() error {
if !entry.IsDir() {
continue
}
parts := strings.SplitN(entry.Name(), "_", 2)
if len(parts) != 2 || parts[0] == "" || parts[1] == "" {
klog.Warningf("Skipping dir with unexpected name format: %s", entry.Name())
continue
}
dirName := filepath.Join(l.containerPath, entry.Name())
podUID := strings.Split(entry.Name(), "_")[0]
podUID := parts[0]
if !podUIDs[podUID] {
dirInfo, err := os.Stat(dirName)
if err == nil && dirInfo.ModTime().Add(resyncInterval).After(time.Now()) {
Expand All @@ -209,11 +214,10 @@ func (l *ContainerLister) Update() error {
continue
}
if usage == nil {
// no cuInit in container
continue
}
usage.PodUID = podUID
usage.ContainerName = strings.Split(entry.Name(), "_")[1]
usage.ContainerName = parts[1]
l.containers[entry.Name()] = usage
klog.Infof("Adding ctr dirname %s in monitorpath", dirName)
}
Expand Down
30 changes: 30 additions & 0 deletions pkg/monitor/nvidia/cudevshr_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -432,4 +432,34 @@ func Test_ContainerLister_Update(t *testing.T) {
assert.Equal(t, got.ContainerName, "mycontainer")
defer func() { _ = syscall.Munmap(got.data) }()
})

t.Run("dir without underscore in name is skipped", func(t *testing.T) {
dir := t.TempDir()
pod := &corev1.Pod{ObjectMeta: metav1.ObjectMeta{Name: "p", Namespace: "default", UID: "nodashes"}}
ctrDir := filepath.Join(dir, "nodashes")
assert.NilError(t, os.Mkdir(ctrDir, 0755))
writeCacheFile(t, ctrDir, "x.cache", headerBytes(v1CacheFileSize, SharedRegionMagicFlag, 1, 0))
l := &ContainerLister{
containerPath: dir,
containers: map[string]*ContainerUsage{},
podLister: newTestPodLister(pod),
}
assert.NilError(t, l.Update())
assert.Equal(t, len(l.containers), 0)
})
Comment thread
coderabbitai[bot] marked this conversation as resolved.

t.Run("dir with trailing underscore is skipped", func(t *testing.T) {
dir := t.TempDir()
pod := &corev1.Pod{ObjectMeta: metav1.ObjectMeta{Name: "p", Namespace: "default", UID: "uid"}}
ctrDir := filepath.Join(dir, "uid_")
assert.NilError(t, os.Mkdir(ctrDir, 0755))
writeCacheFile(t, ctrDir, "x.cache", headerBytes(v1CacheFileSize, SharedRegionMagicFlag, 1, 0))
l := &ContainerLister{
containerPath: dir,
containers: map[string]*ContainerUsage{},
podLister: newTestPodLister(pod),
}
assert.NilError(t, l.Update())
assert.Equal(t, len(l.containers), 0)
})
}
Loading