Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: tikv扩容后，分区表如何处理？
[TiDB Usage Environment] Production
[TiDB Version] v5.0.0
[Encountered Problem]
After expanding TiKV, it was found that the data volume of the new TiKV nodes did not increase, while the data volume of the original TiKV nodes kept increasing. The database mainly has a partitioned table. I understand that all the data is written to the original three TiKVs, and the newly added TiKV has no data written to it. In this case, is it necessary to recreate the partitioned table?
Take a look at the Grafana charts on port 3000:
PD → statistics-balance → store region count
PD → statistics-balance → store region size
You can forcibly schedule the data from the original TiKV node to the new TiKV node, which can be done through PD commands.
PD has built-in balancing scheduling capabilities, so you don’t need to handle it separately. It will automatically process when certain trigger conditions are met, but you can also handle it manually.
I found that the newly added TiKV keeps logging information about connecting to PD.
I found that the newly added TiKV is constantly trying to connect to the PD node, and the network between them is accessible.
Are the network and ports accessible? Check if the several ports of TiDB are connected.
Are there only Info logs? Are there any error logs?
|
username: TiDBer_lW81TNHi
|
username: 
|
username: 
|
username: