kubernetes: "kubectl logs -f" gets stuck for high output logs

Is this a BUG REPORT or FEATURE REQUEST?:

Uncomment only one, leave it on its own line:

/kind bug /area kubectl

/kind feature

What happened: The output of kubectl logs -f gets stuck for high output logs. Restarting the command doesn’t fix the issue. What you expected to happen: To continue getting output from the command. How to reproduce it (as minimally and precisely as possible): $ gcloud container clusters create --cluster-version=1.8.6-gke.0 test $ kubectl run debug --image=alpine – /bin/sh -c “while true ; do date ; sleep 0.001 ; done” $ kubectl logs <debug’s_pods> -f

Anything else we need to know?: The command gets stuck after around 5 minutes running. It doesn’t happen for less frequents logs (e.g. using “sleep 0.01” instead of “sleep 0.001”). This was tested in GKE (Kubernetes Engine) and Kubernetes on GCE clusters.

Environment:

• Kubernetes version (use kubectl version): v1.8.6 on GKE and v1.9.2 on GCE

• Cloud provider or hardware configuration: Google Cloud Platform

• OS (e.g. from /etc/os-release): Container-Optimized OS (GKE version 63, GCE version 60)

• Kernel (e.g. uname -a): GKE nodes: Linux gke-cluster-1-default-pool-4643f8f5-s95q 4.4.86+ #1 SMP Thu Dec 7 20:11:11 PST 2017 x86_64 Intel® Xeon® CPU @ 2.30GHz GenuineIntel GNU/Linux GCE master and nodes: Linux kubernetes-master 4.4.64+ #1 SMP Wed Aug 30 20:27:36 PDT 2017 x86_64 Intel® Xeon® CPU @ 2.30GHz GenuineIntel GNU/Linux Linux kubernetes-minion-group-c0fv 4.4.64+ #1 SMP Wed Aug 30 20:27:36 PDT 2017 x86_64 Intel® Xeon® CPU @ 2.30GHz GenuineIntel GNU/Linux

• Install tools: Followed https://kubernetes.io/docs/getting-started-guides/gce/ for GCE cluster

• Others: