ICEBERG_VACUUM_MORE_RUNS_NEEDED

0

At every 30 minutes it saying "ICEBERG_VACUUM_MORE_RUNS_NEEDED: Removed 20000 files in this round of vacuum" but when I calculate the my table metadata size it didn't changed before and after the vacuum run. Can anyone tell me what to do ?. we have metadata size around 4.5 gb in icerberg table

Naresh
已提問 2 個月前檢視次數 244 次
2 個答案
0

VACCUUM deletes orphan files and expired snapshots, but it won't delete snapshots otherwise (you would lose history).
You can expire snapshosts by configuring in the table when that should happen or by expiring them manually:
https://iceberg.apache.org/docs/latest/spark-procedures/#expire_snapshots

profile pictureAWS
專家
已回答 2 個月前
0

Hello,

The error is expected as it is the behaviour of the command by design. There is a limitation on number of files set as of now to avoid query timeout while the command performs deletion of files in a single run. It is suggested to run the command multiple times until all the files are removed.

AWS
Ankur_J
已回答 2 個月前

您尚未登入。 登入 去張貼答案。

一個好的回答可以清楚地回答問題並提供建設性的意見回饋,同時有助於提問者的專業成長。

回答問題指南