How to manually handle a large number of GC error messages in TiDB logs

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

Original topic: tidb日志中有大量gc错误信息,这种怎么手动处理

| username: Jolyne

[TiDB Usage Environment] Production Environment / Testing / Poc
Production Environment
[TiDB Version]
5.2.1
[Reproduction Path] What operations were performed when the issue occurred
Recently, we found a large number of gc_worker errors in the logs. Will this affect the cluster performance, and can it be handled manually? (Due to security issues, IT previously disabled the SSH protocol on the server, which seems to have caused the connection refused error)
[Encountered Issue: Issue Phenomenon and Impact]
[Resource Configuration]
[Attachments: Screenshots/Logs/Monitoring]

| username: liuis | Original post link

Refer to this post: TiDB 节点大量[gc worker] delete range failed 报错信息 - TiDB 的问答社区

| username: liuis | Original post link

Is it because there is too much data and the GC can’t keep up with the cleanup? It seems like upgrading the version solved it according to the history.

| username: Jolyne | Original post link

I think the error seems to have occurred recently, and there hasn’t been any new data added lately. Our data is all T+1, and data processing isn’t that frequent usually.

| username: tidb菜鸟一只 | Original post link

Can you still connect to the TiKV ports? 172.21.10.139:20166 and 172.21.10.140:20166

| username: WalterWj | Original post link

During cluster restart, I remember this version has GC-related issues.

Consider upgrading later.

| username: buddyyuan | Original post link

It looks like this error is due to being unable to connect to TiKV. Try using telnet to check.

| username: Jolyne | Original post link

I see that it’s connected, so it’s very strange why it keeps reporting connection failure.

| username: Jolyne | Original post link

Ah? Well… I’ll try upgrading to a higher version later. Although this error is reported, it is still functioning normally at the moment, but I’m not sure if this will have a significant impact on the current cluster’s performance.

| username: buddyyuan | Original post link

I just remembered, this issue seems to be related to gc.enable-compaction-filter. I suggest upgrading.

| username: WalterWj | Original post link

Wasting space capacity and some SQL may scan MVCC data, leading to performance degradation. Don’t delay.

| username: Jolyne | Original post link

OK, thanks.

| username: system | Original post link

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