Skip to content

Refactor Manager StartWorkspace #11547

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
merged 1 commit into from
Jul 22, 2022
Merged

Refactor Manager StartWorkspace #11547

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
15 changes: 8 additions & 7 deletions components/ws-manager-api/typescript/src/promisified-client.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -117,16 +117,17 @@ export class PromisifiedWorkspaceManagerClient implements Disposable {
this.client.startWorkspace(
request,
withTracing({ span }),
this.getDefaultUnaryOptions(),
{
// Important!!!!: client timeout must be higher than ws-manager to be able to process any error
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🧡

// https://github.com/gitpod-io/gitpod/blob/main/components/ws-manager/pkg/manager/manager.go#L171
deadline: new Date(new Date().getTime() + 60000*11),
interceptors: this.interceptor,
},
(err, resp) => {
span.finish();
if (err) {
if (attempt < 3 && err.message.indexOf("already exists") !== -1) {
// lets wait a bit more
} else {
TraceContext.setError(ctx, err);
reject(err);
}
TraceContext.setError(ctx, err);
reject(err);
} else {
resolve(resp);
}
Expand Down
16 changes: 13 additions & 3 deletions components/ws-manager/pkg/manager/manager.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -164,7 +164,13 @@ func (m *Manager) Close() {
}

// StartWorkspace creates a new running workspace within the manager's cluster
func (m *Manager) StartWorkspace(ctx context.Context, req *api.StartWorkspaceRequest) (res *api.StartWorkspaceResponse, err error) {
func (m *Manager) StartWorkspace(_ context.Context, req *api.StartWorkspaceRequest) (res *api.StartWorkspaceResponse, err error) {
// We cannot use the passed context because we need to decouple the timeouts
// Create a context with a high timeout value to be able to wait for scale-up events in the cluster (slow operation)
// Important!!!: this timeout must be lower than https://github.com/gitpod-io/gitpod/blob/main/components/ws-manager-api/typescript/src/promisified-client.ts#L122
ctx, cancel := context.WithTimeout(context.Background(), 10*time.Minute)
defer cancel()

owi := log.LogContext(req.Metadata.Owner, req.Metadata.MetaId, req.Id, req.Metadata.GetProject(), req.Metadata.GetTeam())
clog := log.WithFields(owi)
span, ctx := tracing.FromContext(ctx, "StartWorkspace")
Expand Down Expand Up @@ -314,8 +320,12 @@ func (m *Manager) StartWorkspace(ctx context.Context, req *api.StartWorkspaceReq
return nil, err
}

err = wait.PollWithContext(ctx, 100*time.Millisecond, 10*time.Minute, podRunning(m.Clientset, pod.Name, pod.Namespace))
// if we reach this point the pod is created
// in case the context is canceled or a timeout happens we should delete the pod?

err = wait.PollImmediateWithContext(ctx, 100*time.Millisecond, 7*time.Minute, podRunning(m.Clientset, pod.Name, pod.Namespace))
if err != nil {
clog.WithError(err).WithField("req", req).WithField("pod", pod.Name).Warn("was unable to start workspace")
return nil, xerrors.Errorf("workspace pod never reached Running state: %w", err)
}

Expand All @@ -326,7 +336,7 @@ func (m *Manager) StartWorkspace(ctx context.Context, req *api.StartWorkspaceReq
return nil, xerrors.Errorf("unable to get workspace pod %s: %w", pod.Name, err)
}

err = wait.PollWithContext(ctx, 100*time.Millisecond, 5*time.Minute, pvcRunning(m.Clientset, pvc.Name, pvc.Namespace))
err = wait.PollImmediateWithContext(ctx, 100*time.Millisecond, 5*time.Minute, pvcRunning(m.Clientset, pvc.Name, pvc.Namespace))
if err != nil {
if startContext.VolumeSnapshot != nil && startContext.VolumeSnapshot.VolumeSnapshotName != "" {
m.eventRecorder.Eventf(pod, corev1.EventTypeWarning, "PersistentVolumeClaim", "PVC %q restore from volume snapshot %q failed %v", pvc.Name, startContext.VolumeSnapshot.VolumeSnapshotName, err)
Expand Down