How to specify the request frequency when using BR for data recovery?

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

Original topic: 使用br做数据恢复时,怎么指定请求频率呢?

| username: 泰迪比爱好者

【TiDB Usage Environment】Production Environment / Testing / Poc
【TiDB Version】v7.1.1
【Reproduction Path】Operations performed that led to the issue
【Encountered Issue: Issue Phenomenon and Impact】
【Resource Configuration】
【Attachments: Screenshots / Logs / Monitoring】

| username: 泰迪比爱好者 | Original post link

I have specified the minimum rate limit as 2 and the maximum as 1024, but it still disconnects the same way.

| username: 舞动梦灵 | Original post link

How much data?

| username: 泰迪比爱好者 | Original post link

22GB, 200 million rows of data.

| username: 舞动梦灵 | Original post link

Just one table?? That shouldn’t be. I always restore the entire database and haven’t encountered any errors.

| username: 泰迪比爱好者 | Original post link

I’m also puzzled and think there might be some configuration that needs optimization.

| username: 舞动梦灵 | Original post link

Is there any error log information?

| username: 泰迪比爱好者 | Original post link

This is the error log. I set up a MinIO locally and then performed data recovery from MinIO.

| username: 舞动梦灵 | Original post link

Are all the permissions correct? I haven’t used MinIO. I didn’t have any issues with NFS for small full backups, 6G. Check the official documentation to see if the S3-related permissions are relevant.

| username: Fly-bird | Original post link

It seems that there is an issue with your backup files, as the logs indicate they cannot be read.

| username: zhanggame1 | Original post link

How about switching to NFS?

| username: yeminhua | Original post link

Using NFS is both simple and stable. I use high-capacity NFS with a 10-gigabit network, ratelimit 256, and backups and restores of a 30TB database are stable.

| username: 舞动梦灵 | Original post link

Hey bro, how long did your backup take? I’m planning to migrate. With 3 replicas totaling 40T, each one is estimated to be around 14T. How long will it take to complete the backup and restore?

| username: yeminhua | Original post link

I have a 30TB cluster with 3 replicas, and the backup is around 10TB. With a lower speed limit, it normally takes about 10 hours to complete the backup, at a rate of around 400MB per second. This doesn’t affect normal usage.

| username: zhanggame1 | Original post link

It mainly depends on your network bandwidth and disk I/O capability. Without bottlenecks, it can reach several hundred megabytes per second.