pingcap / pingcap/tidb

br node OOM when restore 7.5 cluster backup with 3000 tables with partitions and statistics

Open
#50,289 2 comments 0 reactions 0 assignees View on GitHub
component/br severity/moderate type/bug
Dominant language
Go
Stars
40.5k
Forks
6.2k
PR merge metrics
PR metrics pending

Description

## Bug Report

Please answer these questions before submitting your issue. Thanks!

### 1. Minimal reproduce step (Required)
1. There is a 7.5.0 br full backup data for a cluster about 3k+ tables, each table has some partitions and statstics.
2. Use v7.6.0-pre br (8c 32g) to restore the backup

### 2. What did you expect to see? (Required)
1. br node should not OOM.

### 3. What did you see instead (Required)
2. br node OOM.

### 4. What is your TiDB version? (Required)
br v7.6.0-pre

Contributor guide

Open the contributing guide

Research direction

Start by reproducing the v7.6.0-pre BR restore with the 7.5.0 full backup described in the issue, using the stated 8-core, 32 GB environment, and monitor BR memory usage. Done means the backup containing roughly 3,000 partitioned tables with statistics restores without the BR node running out of memory.

Written by the indexing model from the issue text.

Assessment

Tech stack
go
Domain
databases
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.