Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: Tidb6.1.6版本启动失败
[TiDB Usage Environment] Production Environment
[TiDB Version] v6.1.6
[Reproduction Path] After deployment, TiKV fails to start, and PD reports errors such as unable to synchronize.
[Encountered Problem: Symptoms and Impact] After deployment, TiKV fails to start, and PD reports errors such as unable to synchronize.
[Resource Configuration] 32vCpu 128G 4T
[Attachments: Screenshots/Logs/Monitoring]
tikv.log (7.8 MB)
tiup-cluster-debug-2023-07-11-19-19-48.log (2.8 MB)
pd.log (687.4 KB)
Check if the server time is out of sync.
What is the current status of the cluster? It reports that PD is not responding. Please check the network, especially if the firewall is enabled.
The firewall is not enabled.
Did you send the TiKV logs?
It’s the top tikv.log, right?
Could you please provide the complete content of the log file located at /var/log/tidb-deploy/tikv-20160 for us to review?
It’s all the same error: PD failed to respond. It still looks like a network issue. Check the connectivity to the PD 2379 port on the TiKV host. For example, you can use telnet cdh-mastor01 2379 to see if it is being refused.
TiKV and PD are on the same batch of hosts, and communication is not an issue.
I’m at a loss for now, let’s see if any other experts can solve it.
Is it no longer necessary to use real IPs for deployment now…?
I haven’t deployed it like this before, so I’m not sure how many pitfalls there might be. I also haven’t seen it deployed like this very often…
Change to IP address to solve it 
In a TiDB cluster, the heartbeat and information transmission between nodes are completed through IP connectivity.
Domain names are unreliable.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.
|
username: TiDBer_PD4AfHX8
|
username: 
|
username: 
|
username: 
|
username: 
|
username: