Chesnay Schepler created FLINK-28095:
Summary: Replace IOUtils dependency on oss filesystem
Key: FLINK-28095
URL: https://issues.apache.org/jira/browse/FLINK-28095
Project: Flink
Issue
Chesnay Schepler created FLINK-28091:
Summary: Thread leak in
CheckpointResourcesCleanupRunnerTest#testCancellationAfterStart
Key: FLINK-28091
URL: https://issues.apache.org/jira/browse/FLINK-28091
+1
On 15/06/2022 10:49, Lijie Wang wrote:
Hi everyone,
We've received some additional concerns since the last vote [1], and
therefore made a lot of changes to design. You can find the details in [2]
and the discussions in [3].
Now I'd like to start a new vote thread for FLIP-224. The vote
according to this information. Besides that, it
cannot help with the stage that first observed and affected the slow
nodes.
Thanks,
Zhu
Chesnay Schepler 于2022年6月10日周五 17:04写道：
1)
It's true that if we handle this entirely in the scheduler we may get a bunch
of slow slots from the RM.
My point
My bad, our bylaws actually state that release votes must have a minimum
3 days duration.
On 14/06/2022 14:46, Chesnay Schepler wrote:
Yes, pretty much.
Mind you that the 72h voting duration is a recommendation by the ASF;
it's not a strict rule.
AFAICT we also haven't locked this down
, Jun 14, 2022 at 2:16 PM Chesnay Schepler
wrote:
On 14/06/2022 14:10, Gyula Fóra wrote:
For the operator the main logic (and the bugs) are part of the operator
docker image and the helm charts associated with it. It would be nice
to be
able to have lightweight patch releases that only contain
On 14/06/2022 14:10, Gyula Fóra wrote:
For the operator the main logic (and the bugs) are part of the operator
docker image and the helm charts associated with it. It would be nice to be
able to have lightweight patch releases that only contain the docker
image + updated Helm chart.
This would
On 14/06/2022 14:10, Gyula Fóra wrote:
Another alternative would be to simply not have these patch releases and
encourage users to pick up the latest automatic image builds from
https://ghcr.io/apache/flink-kubernetes-operator for the current release
branch, but I feel that production users
Chesnay Schepler created FLINK-28052:
Summary: Remove AdaptiveSchedulerTest#RunFailedJobListener
Key: FLINK-28052
URL: https://issues.apache.org/jira/browse/FLINK-28052
Project: Flink
Chesnay Schepler created FLINK-28042:
Summary: Create an extension for resetting HiveConf
Key: FLINK-28042
URL: https://issues.apache.org/jira/browse/FLINK-28042
Project: Flink
Issue
@Nick Have you checked whether this really affects Flink`?
Asking since the regressions doesn't categorically affect every mixed
code base.
On 13/06/2022 16:50, Nick wrote:
I think upgrading to 2.12.15 would be the correct choice, as 2.12.16 has a
known regression for projects compiled with
Chesnay Schepler created FLINK-28016:
Summary: Support Maven 3.3+
Key: FLINK-28016
URL: https://issues.apache.org/jira/browse/FLINK-28016
Project: Flink
Issue Type: Technical Debt
Chesnay Schepler created FLINK-28006:
Summary: Run architecture tests in single JVM
Key: FLINK-28006
URL: https://issues.apache.org/jira/browse/FLINK-28006
Project: Flink
Issue Type: Sub
Chesnay Schepler created FLINK-28005:
Summary: Reduce network memory in UnalignedCheckpointITCase
Key: FLINK-28005
URL: https://issues.apache.org/jira/browse/FLINK-28005
Project: Flink
Chesnay Schepler created FLINK-28004:
Summary: Reduce CI heap space by 25%
Key: FLINK-28004
URL: https://issues.apache.org/jira/browse/FLINK-28004
Project: Flink
Issue Type: Technical
Chesnay Schepler 于2022年6月8日周三 18:41写道：
+1
Thank you for proposing this. I can take care of the PMC-side of
things.
On 08/06/2022 12:37, Jingsong Li wrote:
+1
Thanks David for volunteering to manage the release.
Best,
Jingsong
On Wed, Jun 8, 2022 at 6:21 PM Jark Wu wrote:
Hi David
(i.e. pass the blocklist
information directly into slot manager).
5. Regarding the BlocklistTracker. I also agree with you, the BlocklistTracker
does not need to be aware of the executor, and the timeout actions can be
done outside.
Chesnay Schepler 于2022年5月20日周五 17:34写道：
I have a number o
Chesnay Schepler created FLINK-27974:
Summary: Potentially wrong classloader being used to create
dynamic table sources
Key: FLINK-27974
URL: https://issues.apache.org/jira/browse/FLINK-27974
Chesnay Schepler created FLINK-27972:
Summary: Race condition between task/savepoint notification failure
Key: FLINK-27972
URL: https://issues.apache.org/jira/browse/FLINK-27972
Project: Flink
Chesnay Schepler created FLINK-27954:
Summary: JobVertexFlameGraphHandler does not work on standby
Dispatcher
Key: FLINK-27954
URL: https://issues.apache.org/jira/browse/FLINK-27954
Project
+1
Thank you for proposing this. I can take care of the PMC-side of things.
On 08/06/2022 12:37, Jingsong Li wrote:
+1
Thanks David for volunteering to manage the release.
Best,
Jingsong
On Wed, Jun 8, 2022 at 6:21 PM Jark Wu wrote:
Hi David, thank you for driving the release.
+1 for the
lcome
if someone has another good idea which adds less complexity.
If you are still concerned about the value of this feature, I'm fine a
open a survey in the user mailing lists to see how users think about
it.
What do you think?
Thanks,
Zhu
Chesnay Schepler 于2022年6月7日周二 15:13写道：
I've had some
You are on the right path with using the --allowNonRestoredState flag;
we'll just have to find the right place to put it w.r.t. your setup.
Which docker images are you using (flink/statefun/something custom), and
how do you submit the job?
On 03/06/2022 01:17, Bhavani Balasubramanyam wrote:
Chesnay Schepler created FLINK-27933:
Summary: Savepoint OperationResult should be serializable
Key: FLINK-27933
URL: https://issues.apache.org/jira/browse/FLINK-27933
Project: Flink
Zhu
Chesnay Schepler 于2022年5月27日周五 14:54写道：
-1 to put a lid on things for now, because I'm not quite done yet with
the discussion.
On 27/05/2022 05:25, Yangze Guo wrote:
+1 (binding)
Best,
Yangze Guo
On Thu, May 26, 2022 at 3:54 PM Yun Gao wrote:
Thanks Lijie and Zhu for driving the
I'm not necessarily against collecting ideas/topics around the community
side of things and discussing them in some way, but Im very much not a
friend of immediately starting out such an effort with synchronous meetings.
IMO we should only setup such meetings after we have identified a clear
Chesnay Schepler created FLINK-27829:
Summary: Remove Calcite usages in flink-python
Key: FLINK-27829
URL: https://issues.apache.org/jira/browse/FLINK-27829
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27819:
Summary: Generate better operationIds for OpenAPI spec
Key: FLINK-27819
URL: https://issues.apache.org/jira/browse/FLINK-27819
Project: Flink
Issue
Chesnay Schepler created FLINK-27818:
Summary: Model enums as references in OpenAPI spec
Key: FLINK-27818
URL: https://issues.apache.org/jira/browse/FLINK-27818
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27811:
Summary: Remove netty dependency in flink-test-utils
Key: FLINK-27811
URL: https://issues.apache.org/jira/browse/FLINK-27811
Project: Flink
Issue
Chesnay Schepler created FLINK-27810:
Summary: Elasticsearch e2e jars bundle way more than they should
Key: FLINK-27810
URL: https://issues.apache.org/jira/browse/FLINK-27810
Project: Flink
Chesnay Schepler created FLINK-27809:
Summary: Clarify that cluster-id is mandatory for Kubernetes HA in
standalone mode
Key: FLINK-27809
URL: https://issues.apache.org/jira/browse/FLINK-27809
Chesnay Schepler created FLINK-27808:
Summary: Allow "kubernetes" as setting for HA_MODE
Key: FLINK-27808
URL: https://issues.apache.org/jira/browse/FLINK-27808
Project: Flink
-1 to put a lid on things for now, because I'm not quite done yet with
the discussion.
On 27/05/2022 05:25, Yangze Guo wrote:
+1 (binding)
Best,
Yangze Guo
On Thu, May 26, 2022 at 3:54 PM Yun Gao wrote:
Thanks Lijie and Zhu for driving the FLIP!
The blocked list functionality helps reduce
at 3:56 PM Chesnay Schepler wrote:
I'd be more inclined to move the FlinkContainerTestEnvironment to
flink-(connector-)test-utils for the time being, because the vast
majority of stuff in flink-end-to-end-tests-common isn't really in a
state where we want to share it.
On 25/05/2022 11:55
I'd be more inclined to move the FlinkContainerTestEnvironment to
flink-(connector-)test-utils for the time being, because the vast
majority of stuff in flink-end-to-end-tests-common isn't really in a
state where we want to share it.
On 25/05/2022 11:55, Alexander Fedulov wrote:
Hi everyone,
On Wed, May 25, 2022 at 8:54 PM Chesnay Schepler
wrote:
Yes; but that's also a limitation of the current fine-grained
recovery.
My suggestion was primarily aimed at jobs that have no/few blocking
exchanges, where users would currently have to explicitly configure
Chesnay Schepler created FLINK-27778:
Summary: table-planner should explicitly depend on
table-code-splitter
Key: FLINK-27778
URL: https://issues.apache.org/jira/browse/FLINK-27778
Project: Flink
) as the previous execution,
thus downstreams cannot resume consuming the data.
Best,
Xintong
On Wed, May 25, 2022 at 3:05 PM Chesnay Schepler wrote:
Will this also allow spilling everything to disk while also forwarding
data to the next task?
This would allow us to improve fine-grained recovery
Will this also allow spilling everything to disk while also forwarding
data to the next task?
This would allow us to improve fine-grained recovery by no longer being
constrained to pipelined regions.
On 25/05/2022 05:55, weijie guo wrote:
Hi All,
Thank you for your attention and feedback.
This is currently not supported, no.
There are some conceptual issues.
For example, session mode can't isolate the logs between jobs, sou you'd
end up with gargantuan logs being archived for every job.
Similarly, and form of log rotation would reduce the value of such a
mechanism.
I would
Chesnay Schepler created FLINK-27763:
Summary: Remove netty bundling in
flink-streaming-kinesis-tests
Key: FLINK-27763
URL: https://issues.apache.org/jira/browse/FLINK-27763
Project: Flink
See https://issues.apache.org/jira/browse/FLINK-27751
On 24/05/2022 08:12, rui fan wrote:
Hi masters,
All compile_ci of flink-ci[1] failed today, seemingly for the same reason:
05:03:16,982 ERROR org.apache.flink.tools.ci.suffixcheck.ScalaSuffixChecker
[] - Violations found:
Scala-free
Chesnay Schepler created FLINK-27753:
Summary: Scala suffix checks should check for maven errors
Key: FLINK-27753
URL: https://issues.apache.org/jira/browse/FLINK-27753
Project: Flink
They are all different representations of a job.
StreamGraph is the representation made by DataStream API, which gets
converted into the JobGraph (which is API-agnositic), that is submitted
to Flink. From the JobGraph we then create an ExecutionGraph, which is
the runtime representation of a
It's not necessarily required that the scala byte code matches the
version of the java byte code.
By and large such inconsistencies are inevitable w.r.t. external libraries.
On 20/05/2022 12:23, Ran Tao wrote:
Hi, Martijn. Even if we upgrade scala to 2.12.15 to support Java17, it just
fix the
the ability to users to setup the
exact
time
of
blocklist end
through RestAPI
-
Not being tied to a single value of
*cluster.resource-blacklist.item.timeout*
On Mon, 2 May 2022 at 14:17, Chesnay Schepler
<
ches...@apache.org>
wrote:
I do share the concern between blurri
some notes:
- the ES repo currently depends on 1.16-SNAPSHOT; so we'd have to go
back to 1.15 to do such a release as Konstantin proposed. This may have
repercussions on the architecture tests.
- the version scheme in the repo does not yet match what was discussed
in the original proposal
Chesnay Schepler created FLINK-27533:
Summary: Unstable
AdaptiveSchedulerSimpleITCase#testJobCancellationWhileRestartingSucceeds
Key: FLINK-27533
URL: https://issues.apache.org/jira/browse/FLINK-27533
Chesnay Schepler created FLINK-27532:
Summary: Drop flink-clients test-jar
Key: FLINK-27532
URL: https://issues.apache.org/jira/browse/FLINK-27532
Project: Flink
Issue Type: Technical
Chesnay Schepler created FLINK-27501:
Summary: [JUnit5 Migration] SerializerTestBase
Key: FLINK-27501
URL: https://issues.apache.org/jira/browse/FLINK-27501
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27488:
Summary: Migrate flink-docker CI to github actions
Key: FLINK-27488
URL: https://issues.apache.org/jira/browse/FLINK-27488
Project: Flink
Issue Type
+1 to backport it now; the release is already being finalized after all.
On 04/05/2022 10:13, Timo Walther wrote:
Hi everyone,
The 1.15 release is almost out. We should still discuss whether we
want to backport scalafmt to the release-1.15 branch. Currently, it is
quite cumbersome to
Chesnay Schepler created FLINK-27487:
Summary: KafkaMetricWrappers do incorrect cast
Key: FLINK-27487
URL: https://issues.apache.org/jira/browse/FLINK-27487
Project: Flink
Issue Type
:30 PM Chesnay Schepler
wrote:
I still think that's mostly a process issue.
Of course we can be blind-sided if we do the QA for a release artifact
after the release has been finalized.
But that's a clearly broken process from the get-go.
At the very least we should already open a PR when the RC
Chesnay Schepler created FLINK-27477:
Summary: Drop flink-yarn test-jar
Key: FLINK-27477
URL: https://issues.apache.org/jira/browse/FLINK-27477
Project: Flink
Issue Type: Technical Debt
Chesnay Schepler created FLINK-27470:
Summary: [JUnit5 Migration] Module:
flink-statebackend-heap-spillable
Key: FLINK-27470
URL: https://issues.apache.org/jira/browse/FLINK-27470
Project: Flink
Chesnay Schepler created FLINK-27469:
Summary: Remove CliFrontendRunWithYarnTest
Key: FLINK-27469
URL: https://issues.apache.org/jira/browse/FLINK-27469
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27467:
Summary: Remove CliFrontendTestBase
Key: FLINK-27467
URL: https://issues.apache.org/jira/browse/FLINK-27467
Project: Flink
Issue Type: Technical
I do share the concern between blurring the lines a bit.
That said, I'd prefer to not have any auto-detection and only have an
opt-in mechanism
to manually block processes/nodes. To me this sounds yet again like one
of those
magical mechanisms that will rarely work just right.
An external
Chesnay Schepler created FLINK-27466:
Summary: JDBC metaspace leak fix is misleading
Key: FLINK-27466
URL: https://issues.apache.org/jira/browse/FLINK-27466
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27459:
Summary: JsonJobGraphGenerationTest doesn't reset context
environment
Key: FLINK-27459
URL: https://issues.apache.org/jira/browse/FLINK-27459
Project: Flink
Chesnay Schepler created FLINK-27455:
Summary: [JUnit5 Migration] SnapshotMigrationTestBase
Key: FLINK-27455
URL: https://issues.apache.org/jira/browse/FLINK-27455
Project: Flink
Issue
Chesnay Schepler created FLINK-27454:
Summary: Remove inheritance from TestBaseUtils
Key: FLINK-27454
URL: https://issues.apache.org/jira/browse/FLINK-27454
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27453:
Summary: Cleanup TestBaseUtils
Key: FLINK-27453
URL: https://issues.apache.org/jira/browse/FLINK-27453
Project: Flink
Issue Type: Technical Debt
Chesnay Schepler created FLINK-27452:
Summary: Move Te
Key: FLINK-27452
URL: https://issues.apache.org/jira/browse/FLINK-27452
Project: Flink
Issue Type: Technical Debt
Personally I'm tempted to just link to
https://projects.apache.org/committee.html?flink, if at all.
I'm not sure overall whether this listing really provides value in the
first place.
On 28/04/2022 13:58, Xintong Song wrote:
Hi Flink Committers & PMC members,
I just noticed that the list
I'm strongly against a release process
where such an important piece depends on the approval of a 3rd party.
Thank you~
Xintong Song
[1] https://issues.apache.org/jira/browse/FLINK-20650
[2] https://github.com/docker-library/official-images/pull/9249
On Thu, Apr 28, 2022 at 2:43 PM Chesnay Sche
We could just stop releasing to apache/flink and only go for the
official-images route.
On 28/04/2022 07:43, Xintong Song wrote:
Forgot to mention that, we have also proposed to use one shared account and
limit its access to the PMC members, like what we do with the PyPI account.
Chesnay Schepler created FLINK-27433:
Summary: E2E log artifacts contain several thousand rocksdb logs
Key: FLINK-27433
URL: https://issues.apache.org/jira/browse/FLINK-27433
Project: Flink
Chesnay Schepler created FLINK-27432:
Summary: Replace Time with Duration in TaskSlotTable
Key: FLINK-27432
URL: https://issues.apache.org/jira/browse/FLINK-27432
Project: Flink
Issue
Chesnay Schepler created FLINK-27431:
Summary: Allow Duration for RpcTimeouts
Key: FLINK-27431
URL: https://issues.apache.org/jira/browse/FLINK-27431
Project: Flink
Issue Type: Sub-task
Chesnay Schepler created FLINK-27430:
Summary: Remove unused MockResourceManagerRuntimeServices#timeout
Key: FLINK-27430
URL: https://issues.apache.org/jira/browse/FLINK-27430
Project: Flink
Chesnay Schepler created FLINK-27428:
Summary: Deduplicate #toDuration(Time)
Key: FLINK-27428
URL: https://issues.apache.org/jira/browse/FLINK-27428
Project: Flink
Issue Type: Technical
Chesnay Schepler created FLINK-27427:
Summary: Remove timeouts from RpcUtils#terminate*
Key: FLINK-27427
URL: https://issues.apache.org/jira/browse/FLINK-27427
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27426:
Summary: Replace Time with Duration in RPC system internals
Key: FLINK-27426
URL: https://issues.apache.org/jira/browse/FLINK-27426
Project: Flink
Just wanted to point out that when we release 1.0.0 we inevitably also
have to think about what compatibility guarantees we want to give and
how we intend to enforce them.
Additionally it would be good to think about the general approach of
releases; how often are minor/patch releases made,
Chesnay Schepler created FLINK-27398:
Summary: Move completeness test to flink-scala
Key: FLINK-27398
URL: https://issues.apache.org/jira/browse/FLINK-27398
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27390:
Summary: Remove unused flink-tests dependencies
Key: FLINK-27390
URL: https://issues.apache.org/jira/browse/FLINK-27390
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27389:
Summary: Move test utils from flink-test to flink-test-utils
Key: FLINK-27389
URL: https://issues.apache.org/jira/browse/FLINK-27389
Project: Flink
we
want
to depend on.
Thank you~
Xintong Song
On Fri, Apr 22, 2022 at 4:07 PM Chesnay Schepler
wrote:
I'm overall against the idea of creating a fork.
It implies quite some maintenance overhead, like dealing with unstable
tests, CI, licensing etc. and the overall release
I'm overall against the idea of creating a fork.
It implies quite some maintenance overhead, like dealing with unstable
tests, CI, licensing etc. and the overall release overhead.
Is there no alternative where we can collaborate more with the calcite
guys, like verifying new features so bugs
Chesnay Schepler created FLINK-27351:
Summary: TaskExecutorBuilder configures unreasonable amount of
memory
Key: FLINK-27351
URL: https://issues.apache.org/jira/browse/FLINK-27351
Project: Flink
Chesnay Schepler created FLINK-27342:
Summary: Link to Apache privacy policy
Key: FLINK-27342
URL: https://issues.apache.org/jira/browse/FLINK-27342
Project: Flink
Issue Type: Technical
Chesnay Schepler created FLINK-27339:
Summary: Some classes don't have a package
Key: FLINK-27339
URL: https://issues.apache.org/jira/browse/FLINK-27339
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27326:
Summary: Remove LocalExecutor#createWithFactory
Key: FLINK-27326
URL: https://issues.apache.org/jira/browse/FLINK-27326
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27325:
Summary: Remove unnecessary forkCount settings
Key: FLINK-27325
URL: https://issues.apache.org/jira/browse/FLINK-27325
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27324:
Summary: [JUnit5 Migration] Module: flink-rpc-akka-loader
Key: FLINK-27324
URL: https://issues.apache.org/jira/browse/FLINK-27324
Project: Flink
Chesnay Schepler created FLINK-27321:
Summary: [JUnit5 Migration] Module: java-ci-tools
Key: FLINK-27321
URL: https://issues.apache.org/jira/browse/FLINK-27321
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27317:
Summary: Snapshot deployment fails due to .scalafmt.conf not being
found
Key: FLINK-27317
URL: https://issues.apache.org/jira/browse/FLINK-27317
Project
Chesnay Schepler created FLINK-27253:
Summary: Remove custom surefire config from connector-cassandra
Key: FLINK-27253
URL: https://issues.apache.org/jira/browse/FLINK-27253
Project: Flink
Chesnay Schepler created FLINK-27252:
Summary: Remove surefire fork options from connector-hive
Key: FLINK-27252
URL: https://issues.apache.org/jira/browse/FLINK-27252
Project: Flink
Chesnay Schepler created FLINK-27250:
Summary: Remove custom surefire config from sql-parser[-hive]
Key: FLINK-27250
URL: https://issues.apache.org/jira/browse/FLINK-27250
Project: Flink
Chesnay Schepler created FLINK-27234:
Summary: Enable fork-reuse for connector-jdbc
Key: FLINK-27234
URL: https://issues.apache.org/jira/browse/FLINK-27234
Project: Flink
Issue Type
Re why CI didn't pick up on it:
Currently, excessive NOTICE entries don't fail CI, only missing ones do.
They should be logged though!
The reason being that only missing entries are problems (for us, in a
legal sense).
That would be easy to change though.
On 13/04/2022 17:40, Dawid
Chesnay Schepler created FLINK-27232:
Summary: .scalafmt.conf can
Key: FLINK-27232
URL: https://issues.apache.org/jira/browse/FLINK-27232
Project: Flink
Issue Type: Technical Debt
Could you expand a bit on possible alternative implementations that
require this interface to become public, opposed to providing more
built-in ways to subscribe?
On 13/04/2022 11:26, Qingsheng Ren wrote:
Thanks for the proposal Mason! I think exposing `KafkaSubscriber` as public API
is
Chesnay Schepler created FLINK-27228:
Summary: Redistributed modules across CI profiles
Key: FLINK-27228
URL: https://issues.apache.org/jira/browse/FLINK-27228
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27225:
Summary: Remove redundant reuseForks settings
Key: FLINK-27225
URL: https://issues.apache.org/jira/browse/FLINK-27225
Project: Flink
Issue Type
Chesnay Schepler created FLINK-27224:
Summary: Drop redundant flink.forkCountTestPackage property
Key: FLINK-27224
URL: https://issues.apache.org/jira/browse/FLINK-27224
Project: Flink
501 - 600 of 3157 matches
Mail list logo
