Error When Installing Multiple Directories in TiDB v5.2.1

Translated
1

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

Original topic: TIDB v5.2.1版本安装多目录报错

| username: Hacker_g3b9VBO9

cat topology.yaml
global:
user: “tidb”
ssh_port: 22
deploy_dir: “/alidata/tidb-deploy”
data_dir: “/alidata/tidb-data”
server_configs:
tidb:
log.slow-threshold: 300
tikv:
readpool.storage.use-unified-pool: false
readpool.coprocessor.use-unified-pool: true
pd:
replication.location-labels: [“host”]
schedule.leader-schedule-limit: 4
schedule.region-schedule-limit: 2048
schedule.replica-schedule-limit: 64
pd_servers:

  • host: 172.16.7.103

  • host: 172.16.7.104

  • host: 172.16.7.105
    tidb_servers:

  • host: 172.16.7.101

  • host: 172.16.7.102
    tikv_servers:

  • host: 172.16.7.103
    port: 20160
    status_port: 20180
    deploy_dir: “/alidata/tidb-deploy/tikv-20160”
    data_dir: “/alidata/tidb-data/tikv-20160”
    log_dir: “/alidata/tidb-deploy/tikv-20160/log”
    numa_node: “0”
    config:
    server.labels: { host: “tikv1” }

  • host: 172.16.7.103
    port: 20161
    status_port: 20181
    deploy_dir: “/alidata/tidb-deploy/tikv-20161”
    data_dir: “/alidata/tidb-data/tikv-20161”
    log_dir: “/alidata/tidb-deploy/tikv-20161/log”
    numa_node: “1”
    config:
    server.labels: { host: “tikv1” }

  • host: 172.16.7.104
    port: 20160
    status_port: 20180
    deploy_dir: “/alidata/tidb-deploy/tikv-20160”
    data_dir: “/alidata/tidb-data/tikv-20160”
    log_dir: “/alidata/tidb-deploy/tikv-20160/log”
    numa_node: “0”
    config:
    server.labels: { host: “tikv2” }

  • host: 172.16.7.104
    port: 20161
    status_port: 20181
    deploy_dir: “/alidata/tidb-deploy/tikv-20161”
    data_dir: “/alidata/tidb-data/tikv-20161”
    log_dir: “/alidata/tidb-deploy/tikv-20161/log”
    numa_node: “1”
    config:
    server.labels: { host: “tikv2” }

  • host: 172.16.7.105
    port: 20160
    status_port: 20180
    deploy_dir: “/alidata/tidb-deploy/tikv-20160”
    data_dir: “/alidata/tidb-data/tikv-20160”
    log_dir: “/alidata/tidb-deploy/tikv-20160/log”
    numa_node: “0”
    config:
    server.labels: { host: “tikv3” }

  • host: 172.16.7.105
    port: 20161
    status_port: 20181
    deploy_dir: “/alidata/tidb-deploy/tikv-20161”
    data_dir: “/alidata/tidb-data/tikv-20161”
    log_dir: “/alidata/tidb-deploy/tikv-20161/log”
    numa_node: “1”
    config:
    server.labels: { host: “tikv3” }

monitoring_servers:

  • host: 172.16.7.101

grafana_servers:

  • host: 172.16.7.101

alertmanager_servers:

  • host: 172.16.7.101
2
| username: GreenGuan | Original post link

Please change all numa_node to 0.

3
| username: xingzhenxiang | Original post link

Post the logs and check with lscpu.

4
| username: Hacker_g3b9VBO9 | Original post link

Image
Image1380×378 35 KB

5
| username: xingzhenxiang | Original post link

Because there is only node0

image
image1380×388 46.4 KB

GreenGuan provided a solution that can resolve the issue

numa_node: “1”

Please change all numa_node to 0

6
| username: xingzhenxiang | Original post link

For this image, the CPU must include node1’s write numa_node: “1” to take effect.

7
| username: Hacker_g3b9VBO9 | Original post link

What is the command to clear everything?

8
| username: xingzhenxiang | Original post link

Can’t you just replace the configuration file?

9
| username: Hacker_g3b9VBO9 | Original post link

Uninstall it to prevent any future issues.

10
| username: h5n1 | Original post link

Modify the numactl parameters in the run_tikv.sh script under the tikv deploy directory.

11
| username: xingzhenxiang | Original post link

I cleaned it manually, I’ve had it installed for a long time.

12
| username: system | Original post link

This topic will be automatically closed 60 days after the last reply. No new replies are allowed.