TiKV OOM Issue

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

Original topic: tikv oom问题

| username: Jolyne

[TiDB Usage Environment] Production Environment / Testing / Poc
Production Environment
[TiDB Version]
v5.2.1
[Reproduction Path] What operations were performed when the issue occurred
Recently adjusted some parameters related to rocksdb, with memory usage at 45% (kv is 64G), but found that the block-cache monitoring far exceeds 45%, reaching 60~70. Also noticed that the response size exceeds the network outbound. I would like to ask what other adjustments are needed in this situation.
[Encountered Issue: Issue Phenomenon and Impact]
[Resource Configuration]
[Attachments: Screenshots/Logs/Monitoring]




| username: h5n1 | Original post link

| username: Jolyne | Original post link

:joy: I was slowly troubleshooting it based on your article and made some adjustments, but the problem still persists.

| username: Jolyne | Original post link

Is my disk latency too high?

| username: h5n1 | Original post link

Try to reduce the block cache size first. What type of hard drive are you using?

| username: Jolyne | Original post link

Ordinary disk on Huawei Cloud :sweat_smile:

| username: h5n1 | Original post link

That should be fine. My regular SAS disk latency-


For the OOM issue, I suggest you also upgrade to the latest version 5.2 and see.

| username: Jolyne | Original post link

Okay, I’ll apply for a new machine soon and set up a new cluster to see how it goes.

| username: system | Original post link

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