Can't monitor DM-Worker status

[mizuki-masuda]

Bug Report

Please answer these questions before submitting your issue. Thanks!

1. What did you do?

Cannot monitor DM-worker from Node_exporter.

----- topology.yml -----
global:
user: "tidb" ##DM Master,Workerノードに作成されるOSユーザ
ssh_port: 22 ##SSHで利用するポート
deploy_dir: "/dm/deploy" ##deployフォルダが作成されるパス
data_dir: "/dm/data" ##dataフォルダが作成されるパス

monitored: # 各ノードのCPU、メモリ、Disk容量などを監視する場合に必要な設定
node_exporter_port: 19100
blackbox_exporter_port: 19115
deploy_dir: /home/tidb/dm/deploy/monitor-19100
data_dir: /dm/data/monitor-19100
log_dir: /dm/deploy/monitor-19100/log

master_servers:

• host: 172.36.20.137 ##DM Masterノード1のプライベート IPv4 アドレスを記載

worker_servers:

• host: 172.36.20.97 ##DM Workerノードのプライベート IPv4 アドレスを記載

monitoring_servers:

• host: 172.36.20.137 ##DM Masterノードのプライベート IPv4 アドレスを記載

grafana_servers:

• host: 172.36.20.137 ##DM Masterノードのプライベート IPv4 アドレスを記載

alertmanager_servers:

• host: 172.36.20.137 ##DM Masterノードのプライベート IPv4 アドレスを記載

---

----- prometheus.yml -----
- targets:
- '172.36.20.137:19100'
labels:
group: 'node_exporter'

2. What did you expect to see?

Can monitor DM-worker from Node_exporter.

3. What did you see instead?

No, can't see it.

4. What version of TiUP are you using (tiup --version)?

1.5.6 tiup
Go Version: go1.16.7
Git Ref: v1.5.6
GitHash: 1d1478d

[AstroProfundis]

In

tiup/pkg/cluster/spec/monitoring.go

Line 267 in 5e97222

if servers, found := topoHasField("Masters"); found {

the dm worker and master nodes are not added to the uniqueHost set, where node_exporter configs are generated from. (and there is a typo in the function name as well...)