DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1493220098
We use the offline method for compaction. It is normal to start the
compaction, but after running for a period of time, we found that the
compaction is delayed a lot, and the instant
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1493216537
Yes, I understand, we test the effect of different bucket numbers.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1491339024
No, we have stopped the job. The logs file cleanup logic is enabled by
default, as seen from the DAG diagram.
--
This is an automated message from the Apache Git Service.
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1490228324
We are currently stopped the offline Compaction operation. The bucket index
we use and the bucket num are set to 100. Currently, under this file group,
there are 13 parquet files, and
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1488144130
> > there is actually no filedId parquet file
>
> Confused by your words, can you re-organize it a little?
Sorry,We fund instant `20230328130031810 `has been compacted, but
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1487993254
Yes, we checked the Compaction archive file and found that the corresponding
commit has completed the Compaction, but there is actually no filedId parquet
file.
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1486562349
At present, our data is written in real time, and the Compaction is offline
through the Flink job, but we found that there are still some logs files
without Compaction.
Through
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1484881742
Thx for reply.
> * the `--service` param has no value, it is a non-valued param
we tried it, no problem.
--
This is an automated message from the Apache Git Service.
To
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1482523743
We found that part of the Instant Compaction was completed through `Hudi
Cli`. We have a few questions:
1.Why the` Data File Path` is null, through hudi cli?
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1482471170
When I used the` --service `parameter, I found that there was an error in
parsing:
DavidZ1 commented on issue #8267:
URL: https://github.com/apache/hudi/issues/8267#issuecomment-1482134241
Thank you for your answer.
I use the `HUDI CLI` tool to view the compact execution process of the
table, and found that many compacts are `inflight`. Does it mean that the
11 matches
Mail list logo