TiDB 7.5 Service Fails to Start on Euler System Deployment

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

Original topic: 欧拉系统部署TIDB7.5无法启动TIDB服务

| username: kelvin

[TiDB Usage Environment] Virtual Machine Testing
[TiDB Version] v7.5.0
[Reproduction Path]
[Encountered Issue: TiDB service cannot start on Euler system deployment of TiDB 7.5, PD and TiKV start normally]
[Resource Configuration]
[Attachments: Screenshots/Logs/Monitoring]


| username: h5n1 | Original post link

Is this a fresh deployment or an upgrade? Refer to tidb升级到7.5.0, tidb servers升级报错了怎么处理 - #5,来自 zhanggame1 - TiDB 的问答社区

| username: kelvin | Original post link

Newly deployed

| username: zhanggame1 | Original post link

Destroy and redeploy once again.

| username: kelvin | Original post link

After destroying and redeploying, the same error still occurs.

| username: h5n1 | Original post link

Try adding resource-control.enabled: false under the TiKV configuration in the YAML file.

| username: Kongdom | Original post link

Refer to this and take a look:

| username: tidb狂热爱好者 | Original post link

You can import this yourself, right?

| username: kelvin | Original post link

There is no problem deploying TiDB on Euler 2203, but the TiDB server cannot be started. It’s so frustrating.

| username: Billmay表妹 | Original post link

Which version is the Euler system?

| username: kelvin | Original post link

The x86_64 one

| username: ShawnYan | Original post link

I would like to confirm which version of EulerOS you are using? Please provide the specific OS version number.

| username: kelvin | Original post link

I feel like my operating system is too new, right?

| username: 托马斯滑板鞋 | Original post link

Your kernel version is on par with Rocky Linux 9.2 :joy:

| username: 有猫万事足 | Original post link

At the beginning, the logs reported a timeout. Try adjusting the --wait-timeout parameter and see. You can check the explanation through tiup cluster --help.

–wait-timeout uint Timeout in seconds to wait for an operation to complete, ignored for operations that don’t fit. (default 120)

It seems like the table creation wasn’t completed within 2 minutes during the first startup, and then it couldn’t start anymore. If a TiDB instance can’t start at all, there’s no way to log in and create tables. It feels like the absence of the tidb_runaway_watch table is not the root cause of the failure to start.

From here, it doesn’t seem like this error was propagated up, causing the entire process to exit. It just simply logged the error.

| username: 小龙虾爱大龙虾 | Original post link

See if there are any other logs.

| username: zhanggame1 | Original post link

Try deploying version 7.1.2, that should be fine. The table creation failure is unlikely to be an operating system issue.

| username: ShawnYan | Original post link

His kernel is even newer than Rocky 9.3

[shawnyan@rocky9 ~]$ cat /etc/system-release
Rocky Linux release 9.3 (Blue Onyx)
[shawnyan@rocky9 ~]$ uname -r
5.14.0-362.8.1.el9_3.x86_64
| username: ShawnYan | Original post link

It’s not impossible, why not try using 22.03 first?

| username: Kongdom | Original post link

:joy: My understanding of deployment is that it means being able to run. You can first refer to the suggestions of other classmates and try lowering the operating system version.