Bo Cui created HUDI-4078:
----------------------------
Summary: BootstrapOperator cannot load all index data
Key: HUDI-4078
URL: https://issues.apache.org/jira/browse/HUDI-4078
Project: Apache Hudi
Issue Type: Bug
Reporter: Bo Cui
the bootstrapOperator can not obtain all the parquet and and log from the
hoodieTable#getSliceView()#getLatestFileSlicesBeforeOrOn
Procedure:
1) write 10k records to the HUDI table by stream mode.
create table（） with (
'table.type' = 'MERGE_ON_READ',
'index.bootstrap.enabled' = 'true',
'archive.max_commits' = '4200',
'archive.min_commits' = '4000',
'clean.retain_commits' = '3999',
...
)
2) stop job, and delete the last compaction commit, like
`.hoodie/20220505131426.commit`
3) restart job without chk/savepoint and not write data.
4) Observe how much index data is loaded to the bootstrapOperator.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
