TiDB Cloud: ERROR 9003 (HY000) TiKV Server is Busy under high load

TL;DR

If it recurs at steady load, the fix is more TiKV capacity, not more retries. Check the TiKV status, monitoring data and logs to find the hotspot: unbalanced regions, too many concurrent scans, or disk IO saturation.

The error

ERROR 9003 (HY000) TiKV Server is Busy under high load

Fix

1. If it recurs at steady load, the fix is more TiKV capacity, not more retries. Expected: You get the expected result; the problem is gone.

2. Re-run the original operation and confirm the error is gone. Expected: no error, normal output.

When to use

• You hit this exact error with TiDB Cloud.
• The symptom matches: ERROR 9003 (HY000) TiKV Server is Busy under high load.

When NOT to use

• A different error message from TiDB Cloud; the cause here is specific to this error.
• Unrelated TiDB Cloud issues (different feature, different failure).
• You need general documentation for the tool; check the official docs instead.

Compatibility

Reported against TiDB Cloud. Source: https://github.com/pingcap/docs/blob/HEAD/error-codes.md.

Variant phrasings

ERROR 9003 (HY000) TiKV Server is Busy under high load

Why it happens

Queries to TiDB Cloud fail with ERROR 9003 (HY000): TiKV Server is Busy. This occurs when the TiKV workload is too high to keep up with incoming requests.

Edge cases

• If your error message differs even slightly, this is probably a different issue; search the exact text.
• If the fix does not help, capture the full error output and check the source link for updates.

Source

https://github.com/pingcap/docs/blob/HEAD/error-codes.md

Source: Queries to TiDB Cloud fail with ERROR 9003 (HY000): TiKV Server is Busy. This occurs when the TiKV workload is too high to keep up with incoming requests.

Verified resolution: [PingCAP docs, Error Codes and Troubleshooting]: Check the TiKV status, monitoring data and logs to find the hotspot: unbalanced regions, too many concurrent scans, or disk IO saturation. Reduce the concurrent client count or split the hot transaction, and retry with backoff since this is transient. If it recurs at steady load, the fix is more TiKV capacity, not more retries.

Source URL: https://github.com/pingcap/docs/blob/HEAD/error-codes.md