Cockroachdb cluster 1.1.4 3 nodes always show suspect status

I’m currently testing cockroachdb cluster version 1.1.4 with 3 nodes. All run on 3 docker hosts. Docker hosts is 3 vm on 3 physical servers.
From management web, all 3 nodes always show suspect status.
All docker hosts are ntp synchronized.
There is no load to the cluster.
Any idea what causing this?

Best Regards

Nevermind, I found that it caused by the client browser is not synced.

Why the client not being sync cause the cluster nodes indicated as suspect?
It is quite confusing.

Hi stanley,

Thanks for the report! This issue is a known issue and is being tracked here: