Dumpling or TiDB needs to improve table scan capability to avoid TiDB OOM #34814
Labels
duplicate
Issues or pull requests already exists.
type/feature-request
Categorizes issue or PR as related to a new feature.
Feature Request
Is your feature request related to a problem? Please describe:
During testing dumpling w/ Dynamic Regions feature, which means the tidb cluster config contains this
The tikv region sizes are > 1200MB.
The tidb-server was OOM when dump with 10 threads
/dumpling -h "downstream-tidb.brie-acceptance-long-time-tps-810231-1-41.svc" -P "4000" -u "root" --output "/dump_from_downstream" --filetype "csv" --filesize "256M" --rows "10000" --threads "10" --database "H"
. The same command will be just fine with region sizes are < 250MB. After I reduce the threads to 4, there is no OOM anymore.Describe the feature you'd like:
I'd like one of the following features, which might solve the problem:
Describe alternatives you've considered:
Teachability, Documentation, Adoption, Migration Strategy:
tidb log
dumpling log
The text was updated successfully, but these errors were encountered: