Quick Start with TiDB: Simulating Production Environment Cluster Deployment on a Single Machine - Startup Failure

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: 快速上手TiDB–在单机上模拟部署生产环境集群–启动失败

| username: 随缘天空

[TiDB Usage Environment] Single machine Linux system, CentOS 7.6
[TiDB Version] v7.1.0
[Reproduction Path] Refer to the Quick Start with TiDB - Simulate Deployment of Production Environment Cluster on a Single Machine
[Encountered Issue: Phenomenon and Impact] The first 7 steps were executed successfully, but the 8th step to start the cluster reported an error, with the TiKV cluster startup timing out.
[Resource Configuration]
[Attachments: Screenshots/Logs/Monitoring]



| username: xfworld | Original post link

You can check why the TiKV node failed to start, there is a log address in the place you highlighted…


A TiKV node can be configured with just one… if it’s just for experiencing it…

Additionally, you can refer to the hybrid deployment documentation:

| username: 随缘天空 | Original post link

The last screenshot shows the error information in the error log, indicating a startup timeout. However, the specific cause of the timeout is unknown. I set the timeout to 10 minutes when starting, but it still reports a timeout error: tiup cluster start Tidb-cluster --wait-timeout 600.

| username: 像风一样的男子 | Original post link

Check the error log for the PD startup.

| username: xfworld | Original post link

You are talking about the logs of tiup, not tikv…

| username: zhanggame1 | Original post link

Check the PD logs.

| username: 裤衩儿飞上天 | Original post link

Please provide the corresponding TiKV logs for review.

| username: Kongdom | Original post link

Could it be due to insufficient single-machine resources? Or should we check the log information under /tidb-deploy/tikv-20162/log?

| username: redgame | Original post link

Is there any other error in the tikv.log or tikv_stderr.log?

| username: 随缘天空 | Original post link

tikv.log (293.3 KB)

| username: 随缘天空 | Original post link

The logs show that the PD response failed, but the console indicates that the PD component started successfully.


tikv.log (293.3 KB)

| username: 随缘天空 | Original post link

pd.log (137.4 KB)
pd_stderr.log (2.4 KB)

| username: 随缘天空 | Original post link

Uploaded, please take a look.

| username: 随缘天空 | Original post link

The logs have been uploaded, but the error message says it cannot connect to PD. However, the third-to-last line in the PD log says it cannot fetch the cluster from remote. I’m not sure what the exact reason is.

| username: xfworld | Original post link

What are the machine specifications?

| username: Kongdom | Original post link

Is the firewall turned off?

| username: 随缘天空 | Original post link

Turned off, not sure if it’s a system configuration issue. I deployed it on my own Tencent Cloud CentOS 7.6 system.
Configuration: CPU 2 cores, memory 8GB, hard disk around 80GB.

| username: 随缘天空 | Original post link

Configuration: 2-core CPU, 8GB memory, around 80GB hard disk.
I deployed it on my own Tencent Cloud CentOS 7.6 system, referring to the official documentation: Quick Start with TiDB - Simulate Deployment of Production Environment Cluster on a Single Machine. The first few steps were executed successfully, but an error occurred at step 8 when starting the cluster.

| username: Kongdom | Original post link

:thinking: If it’s a cloud server, I remember the cloud server management platform also has a firewall. Check it out.

| username: xfworld | Original post link

First, rule out firewall issues.