milvus: [Bug]: [chaos]Flush hangs at e2e test after deleting pulsar pod kill chaos

Is there an existing issue for this?

  • I have searched the existing issues

Environment

- Milvus version: master-20211225-132b3c2
- Deployment mode(standalone or cluster): cluster
- SDK version(e.g. pymilvus v2.0.0rc2): 2.0.0rc9.dev24
- OS(Ubuntu or CentOS): 
- CPU/Memory: 
- GPU: 
- Others:

Current Behavior

Flush hangs at e2e test after deleting datacoord or rootcoord pod kill chaos

Expected Behavior

all operations work well after deleting chaos

Steps To Reproduce

# reproduce on local k8s env
cd `tests/python_client/chaos`
modify milvus tag in `cluster-values.yaml` to `master-20211225-132b3c2`
run command `bash chaos_test.sh rootcoord pod_kill`

# run GitHub action
https://github.com/milvus-io/milvus/actions/workflows/pod-kill-chaos-test.yaml

Anything else?

datacoord log: https://github.com/milvus-io/milvus/suites/4764821710/artifacts/132476087

rootcoord log: https://github.com/milvus-io/milvus/suites/4764821710/artifacts/132476097

log path is k8s_logs/chaos_test

About this issue

  • Original URL
  • State: closed
  • Created 3 years ago
  • Comments: 43 (42 by maintainers)

Commits related to this issue

Most upvoted comments

Seek cost 9mins

@XuanYang-cn Would you help to figure out the reason why it cost so long? 👍