Hi,
Very thanks Fabian for drafting this FLIP! It looks very good to me. I see
currently most of us agree with option 2, but I personally feel that option
3 may be better :-)
I have some small concerns for option 2
1. Developers understand that the cost is relatively high.
The merging of small fi
Caizhi Weng created FLINK-24759:
---
Summary: Metrics on web UI is not correct in a filesystem source
to filesystem sink streaming job
Key: FLINK-24759
URL: https://issues.apache.org/jira/browse/FLINK-24759
Jingsong Lee created FLINK-24758:
Summary: partition.time-extractor.kind support "MMdd"
Key: FLINK-24758
URL: https://issues.apache.org/jira/browse/FLINK-24758
Project: Flink
Issue Type:
Caizhi Weng created FLINK-24757:
---
Summary: Yarn application is not terminated after the job finishes
when submitting a yarn-per-job insert job with SQL client
Key: FLINK-24757
URL: https://issues.apache.org/jira/bro
Hi Fabian,
Thanks for drafting the FLIP!
## Few thoughts of user requirements
1.compact files from multiple checkpoints
This is what users need very much.
2.The compaction block the checkpointing
- Some scenarios are required. For example, the user expects the
output data to be consistent wit
Hi everyone,
I would like to start a vote on FLIP-189: SQL Client Usability Improvements
[1].
The FLIP was discussed in this thread [2].
FLIP-189 targets usability improvements of SQL Client such as parsing
improvement,
syntax highlighting, completion, prompts
The vote will be open for at least 7
Frederic Hemery created FLINK-24756:
---
Summary: Flink metric identifiers contain group variables.
Key: FLINK-24756
URL: https://issues.apache.org/jira/browse/FLINK-24756
Project: Flink
Issue
Aitozi created FLINK-24755:
--
Summary: sun.misc doesn't exist
Key: FLINK-24755
URL: https://issues.apache.org/jira/browse/FLINK-24755
Project: Flink
Issue Type: Bug
Components: Build System
Hi David and Till,
Thanks for your great feedback. One definitely confusing point in the FLIP is
who is doing the actual compaction. The compaction will not be done
by the CommittableAggregator operator but the committers so it should also not
affect the checkpointing duration or have a signif
xuyang created FLINK-24754:
--
Summary: PushDownSepcs are better to be print by a clear order.
Key: FLINK-24754
URL: https://issues.apache.org/jira/browse/FLINK-24754
Project: Flink
Issue Type: Improv
Marios Trivyzas created FLINK-24753:
---
Summary: Enforce CHAR/VARCHAR precision when outputing to a Sink
Key: FLINK-24753
URL: https://issues.apache.org/jira/browse/FLINK-24753
Project: Flink
Hi Timo,
I completely agree it would be great if we can propagate Calcite parser
config in
the way you have described. As you mentioned we could discuss this
when it comes to the implementation.
Meanwhile it looks like I can start voting (please correct me if I'm wrong).
I will start it a bit lat
Francesco Guardiani created FLINK-24752:
---
Summary: Cleanup ScalarOperatorGens#generateCast
Key: FLINK-24752
URL: https://issues.apache.org/jira/browse/FLINK-24752
Project: Flink
Issue T
Ideally, the compaction won't affect the checkpointing time. If the
compaction takes longer to complete, then the result could be published
with the next checkpoint. Of course, this would increase the end-to-end
latency.
I hope that with option 2, we can support both use cases: single task
compact
tao.yang03 created FLINK-24751:
--
Summary: flink SQL comile failed cause by:
java.lang.StackOverflowError
Key: FLINK-24751
URL: https://issues.apache.org/jira/browse/FLINK-24751
Project: Flink
Sergey Nuyanzin created FLINK-24750:
---
Summary: Single quotes converted to left/right single quotes
during docs generation
Key: FLINK-24750
URL: https://issues.apache.org/jira/browse/FLINK-24750
Proj
Hi Sergey,
thanks for your explanation.
Regarding keywords and other info: We should receive the information
from the Flink SQL parser directly. We have added a couple of new
keywords such as WATERMARK or MATCH_RECOGNIZE clauses. SQL92 would not
help a user understand why a column name needs
Hi Fabian, thanks for drafting the FLIP! This is a really nice and useful
topic to target ;)
Few thoughts on the option 2)
The file compaction is by definition quite costly IO bound operation. If I
understand the proposal correctly, the aggregation itself would run during
operator (aggregator) ch
Chesnay Schepler created FLINK-24749:
Summary: Reuse CheckpointStatsTracker across rescaling
Key: FLINK-24749
URL: https://issues.apache.org/jira/browse/FLINK-24749
Project: Flink
Issue T
Chesnay Schepler created FLINK-24748:
Summary: Remove
CheckpointStatsTracker#getJobCheckpointingConfiguration
Key: FLINK-24748
URL: https://issues.apache.org/jira/browse/FLINK-24748
Project: Flink
Thanks for creating this FLIP Fabian.
>From your description I would be in favour of option 2 for the following
reasons: Assuming that option 2 solves all our current problems, it seems
like the least invasive change and smallest in scope. Your main concern is
that it might not cover future use ca
Hi 李宇彬,
I think you are right. Thank you very much for the idea.
I came across MySQL[1] and PostgreSQL[2] prompts and also
found several interesting features like control symbols to change style,
showing current property value and different datetime formats.
I have added your proposals and my fin
Francesco Guardiani created FLINK-24747:
---
Summary: Add producedDataType to
SupportsProjectionPushDown.applyProjection
Key: FLINK-24747
URL: https://issues.apache.org/jira/browse/FLINK-24747
Proj
I have to admit that I cannot think of a better name for the adaptive batch
scheduler atm. Maybe it is good enough to call the two schedulers
AdaptiveBatchScheduler and AdaptiveStreamingScheduler to tell which
scheduler is used for which execution mode. It is true, though, that the
former is adapti
Chesnay Schepler created FLINK-24746:
Summary: Alibaba maven mirror is unstable
Key: FLINK-24746
URL: https://issues.apache.org/jira/browse/FLINK-24746
Project: Flink
Issue Type: Technica
Steven created FLINK-24745:
--
Summary: Add support for Oracle OGG json parser in flink-json
module
Key: FLINK-24745
URL: https://issues.apache.org/jira/browse/FLINK-24745
Project: Flink
Issue Type:
Till Rohrmann created FLINK-24744:
-
Summary:
FlinkKafkaProducerITCase.testMigrateFromAtLeastOnceToExactlyOnce fails on Azure
because org.apache.kafka.common.errors.UnknownTopicOrPartitionException: This
server does not host this topic-partition
Till Rohrmann created FLINK-24743:
-
Summary: New File Sink end-to-end test fails on Azure
Key: FLINK-24743
URL: https://issues.apache.org/jira/browse/FLINK-24743
Project: Flink
Issue Type: Bu
Hi David,
Thanks for your comments.
I personally think that "Adaptive" means: Flink automatically determines
the appropriate scheduling and execution plan based on some information.
The information can include both resource information and workload
information, rather than being limited to a ce
29 matches
Mail list logo
