Issues with Starting dm-work

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

Original topic: dm-work 启动问题

| username: 宸凡_22

[TiDB Usage Environment] Production Environment / Testing / PoC
[TiDB Version]
[Reproduction Path] What operations were performed when the issue occurred
[Encountered Issue: Issue Phenomenon and Impact]
[Resource Configuration] Enter TiDB Dashboard - Cluster Info - Hosts and take a screenshot of this page
[Attachments: Screenshots / Logs / Monitoring]

The above image shows the frontend display.
In the dm-meta database, it was found that there is no lightning_checkpoint_list.

In the above image, the information related to dm127 is from the local database, while dm62 is from a database on another server. The dm127 database can synchronize to TiDB; the dm62 database cannot. I suspect there might be some differences. However, no valuable clues can be found from the backend logs. The only difference found is in the above image.

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

Check the logs again, there should be something there.

| username: keyway | Original post link

If there isn’t much data, try reinstalling and see.

| username: zhang_2023 | Original post link

What is the detailed error?

| username: 宸凡_22 | Original post link

Created data source

| username: 宸凡_22 | Original post link

The master log is shown in the image above, and there are no log updates in the worker.

| username: 宸凡_22 | Original post link

It has nothing to do with DM installation. I can normally synchronize when I install a MySQL database locally.

| username: 宸凡_22 | Original post link

From my experience with synchronizing multiple databases, it seems that the issue is caused by a missing component in the database that needs to be synchronized. However, there are no relevant prompts in the logs, so it’s unclear what the problem is.

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

Check the corresponding dm-worker logs, there shouldn’t be nothing at all, right?

| username: 宸凡_22 | Original post link

In the master-err log, I see this error:
{“level”:“warn”,“ts”:1706532916.9048774,“caller”:“fileutil/fileutil.go:57”,“msg”:“check file permission”,“error”:“directory "/datafile01/05_DB_DataFiles/03_Tidb/dm/data/dm-master-8261" exist, but the permission is "drwxr-xr-x". The recommended permission is "-rwx------" to prevent possible unprivileged access to the data”}

| username: 宸凡_22 | Original post link

dm-worker.log (11.7 KB)

| username: 有猫万事足 | Original post link

Actually, looking at the number of tables, the number of dm_meta tables for these two tasks doesn’t seem quite right. Normally, there should be 4. You have one with 3 and one with 2.


Still, I suggest

tiup dmctl query-status

to check the result.

export DM_MASTER_ADDR=“xx.xx.xx.xx:8261”

Remember to set this environment variable, so you don’t have to add --master-addr xx.xx.xx.xx:8261 every time you use tiup dmctl, which is more convenient.

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

Can’t see the problem :smiley_cat: