Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: tidb集群单机部署出现无法登录dashboard的情况
How should I handle the situation where I cannot log in to the dashboard after deploying a TiDB cluster on a single machine? The 4000 port is actually in a running state, but the cluster status shows down. When trying to log in to the dashboard through the browser, it shows login failed: no running TiDB instance.
Check the logs in the TiDB directory to see what caused this abnormal state.
Is it the logs under this path? Which one should I specifically look at?
Check the tidb.log for any error logs, TiDB is not starting up.
It is obvious that the TiDB node did not start successfully.
The way this cluster is started causes some services to fail to bring up the ports.
It is displayed like this.
Please upload the corresponding tidb.log file.
It looks like an SSH permission issue. Have you set up mutual trust?
Is it a cluster deployed on different ports of a single machine?
Turn off the firewall and try setting up mutual trust.
Is the number of connections provided by SSH sufficient? If not, it will be rejected. This issue only occurs on a single machine.
I just tested increasing the maximum number of SSH terminal connections to 1000, and the same error occurred.
All I can say is to try more, it’s an SSH issue.
Thank you, boss. I’ll try again.
Do I still need to set up mutual trust for a single machine?
Check the status of the cluster components with tiup cluster display cluster_name
to see if they are all up.
Is it deploying all nodes on a single machine?
Looking at this picture, it involves using bash and systemctl commands. The main focus here is passwordless sudo and ssh. Can your control machine connect to TiDB via ssh without a password? What is the user specified in your configuration file for starting the cluster? Both sudo and ssh are related to the user in the configuration file.