TiDB Node Continuously Reports Error: Region Count Getting from PD is 0

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

Original topic: TiDB节点持续报错region count getting from PD is 0

| username: Kongdom

[TiDB Usage Environment] Production Environment
[TiDB Version] v6.5.3
[Reproduction Path] The cluster has undergone data center migration, PD reconstruction, expansion, and contraction.
[Encountered Problem: Phenomenon and Impact]
The cluster has deployed TiFlash nodes but is not using them. Currently, TiDB continuously reports errors, approximately once per millisecond.

[2023/06/28 17:16:24.574 +08:00] [ERROR] [ddl_tiflash_api.go:396] ["get tiflash sync progress failed"] [error="region count getting from PD is 0"] [tableID=8765] [IsPartition=false]
[2023/06/28 17:16:24.575 +08:00] [ERROR] [ddl_tiflash_api.go:396] ["get tiflash sync progress failed"] [error="region count getting from PD is 0"] [tableID=8765] [IsPartition=false]
[2023/06/28 17:16:24.576 +08:00] [ERROR] [ddl_tiflash_api.go:396] ["get tiflash sync progress failed"] [error="region count getting from PD is 0"] [tableID=8765] [IsPartition=false]
[2023/06/28 17:16:24.577 +08:00] [ERROR] [ddl_tiflash_api.go:396] ["get tiflash sync progress failed"] [error="region count getting from PD is 0"] [tableID=8765] [IsPartition=false]
[2023/06/28 17:16:24.578 +08:00] [ERROR] [ddl_tiflash_api.go:396] ["get tiflash sync progress failed"] [error="region count getting from PD is 0"] [tableID=8765] [IsPartition=false]
| username: robert | Original post link

Has this phenomenon always existed, or did it only last for a period of time?

| username: Kongdom | Original post link

It has always existed, and many 300M log files have already been generated.

| username: tidb菜鸟一只 | Original post link

Is it all about the 8765 table? What kind of table is this, and have you enabled the TiFlash replica?

| username: Kongdom | Original post link

:sweat_smile: It’s a temporary table. The awkward part is that the TiFlash node has already been scaled down, and the cluster no longer has any TiFlash nodes.

| username: tidb菜鸟一只 | Original post link

Are there any other temporary tables? Check the error, and what is the difference with this table? Has the error stopped after scaling down the TiFlash node?

| username: 有猫万事足 | Original post link

“select * from information_schema.tiflash_replica tr where tr.TABLE_ID=8765;”

Check which table it is and whether there is a TiFlash replica on this table.

If there is, it means you have set a TiFlash replica for this table but there is no TiFlash available. Removing the replica should solve the issue.

| username: Kongdom | Original post link

:astonished: Yes, there is data in this table, which shouldn’t be the case. I remember it should check for tiflash replicas when going offline. Let me make some changes and try again.

| username: 有猫万事足 | Original post link

I just realized it’s a temporary table, and the table is gone, so it seems more like a bug. :joy:

| username: Kongdom | Original post link

It is not a real temporary table; it is a physical table that is only used temporarily.

| username: Kongdom | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.