Congratulations zhijiang :)
Best,
Yang
yangtao.yt 于2019年7月23日周二 上午10:17写道：
> Congrats, Zhejiang!
>
> Best,
> Tao Yang
>
> > 在 2019年7月23日，上午9:46，boshu Zheng 写道：
> >
> > Congratulations Zhijiang
> >
> > 发自我的 iPhone
> >
> >> 在 2019年7月23日，上午12:55，Xuefu Z 写道：
> >>
> >> Congratulations, Zhijiang!
>
Hi xintong,
Thanks for your detailed proposal. After all the memory configuration are
introduced, it will be more powerful to control the flink memory usage. I
just have few questions about it.
- Native and Direct Memory
We do not differentiate user direct memory and native memory. They ar
Hey Fabian,
Sounds great! It will be more easier to build an end-to-end play ground
with docker.
I prefer option 1.
We need to build the official docker images and push to docker hub after
every version is released. It could be used to play with docker compose,
kubernetes, etc.
Today we coul
Hi all,
Currently cloud native architectures has been introduced to many companies
in production. They use kubernetes to run deep learning, web server, etc.
If we could deploy the per-job/session flink cluster on kubernetes to make
it mix-run with other workloads, the cluster resource utilization
t; > Thanks Yang. K8s natively integration is very necessary and important for
> > the adoption of flink IMO.
> > I notice that the design doc is written in 2018, is there any changes or
> > update ?
> >
> > >>> Download the flink release binary and create the
re any changes or
> > update ?
> >
> > >>> Download the flink release binary and create the ~/.kube/config file
> > corresponding to the k8s cluster. It is all what you need.
> >
> > How can I specify which k8s cluster to run in case I have multiple k8s
>
hould not be affected
> by
> > this and I would hope that this is the only point of interaction a
> > streaming job would have with the MemoryManager.
> >
> > Concerning the second open question about setting or not setting a max
> > direct memory limit, I would
>
> [1] https://issues.apache.org/jira/browse/FLINK-10407
> [2] https://issues.apache.org/jira/browse/FLINK-9953
> [3]
>
> https://lists.apache.org/thread.html/9b873f9dc1dd56d79e0f71418b123def896ed02f57e84461bc0bacb0@%3Cdev.flink.apache.org%3E
>
> Cheers,
> Till
>
> On Mo
Hi Cristian,
I think you could have a try on standalone cluster on X(underlying cluster
framework). By now it works well for kubernetes[1].
If your custom distributed system support to start an application based on
yaml/json configuration, then it will be very simple to start a session
cluster. Fo
Hi Xintong,
Thanks for your detailed proposal. I think many users are suffering from
waste of resources. The resource spec of all task managers are same and we
have to increase all task managers to make the heavy one more stable. So we
will benefit from the fine grained resource management a lot.
>From the user's perspective, it is really confused about the scope of
per-job cluster.
If it means a flink cluster with single job, so that we could get better
isolation.
Now it does not matter how we deploy the cluster, directly deploy(mode1)
or start a flink cluster and then submit job throu
d be helpful if you check Stephan's last comment,
> > > > which states that isolation is important.
> > > >
> > > > For per-job mode, we run a dedicated cluster(maybe it
> > > > should have been a couple of JM and TMs during FLIP-6
> > >
utility. Setting larger -XX:MaxDirectMemorySize
> > will
> > > > not
> > > > > > > > reduce
> > > > > > > > > > the
> > > > > > > > > > > > sizes of the other memory pools.
> > >
sion.sh -d -n 2 -tm 512 -s 4 -nm
> > flink-session-example
> > > -i flink:latest -kD kubernetes.service.exposed.type=NODE_PORT
> >
> > Is the -n option means number of TaskManager?
> > Do we pre-running taskmanager pods or requesting and launching
> taskmanager
>
Hi Dadashov,
Regarding your questions.
> Q1 Do all those 800 nodes download of batch of 3 at a time
The 800+ containers will be allocated on different yarn nodes. By default,
the LocalResourceVisibility is APPLICATION, so they will be downloaded only
once and shared for all taskmanager conta
of Flink processes easier. However, I think
> that
> >> >> adding
> >> >> > > > this functionality should not be part of this FLIP because it
> >> would
> >> >> > > simply
> >> >> > > > widen the scope un
Thanks for creating this FLIP.
This sounds like a useful feature to make the Flink applications running on
YARN cluster more securely.
However, I think we still miss some important parts in the FLIP.
1. Which hadoop versions this FLIP relies on?
2. We need to describe a bit more about how the YAR
ore descriptive.
> 3. ACLs apply to logs as well as permissions to kill the application.
> Also, in the PR we are setting ACLs for Task Manager
> (createTaskExecutorContext) as well as Job Manager (startAppMaster).
>
> Thanks,
> Archit Goyal
>
> From: Yang Wang
> Date:
+1 (binding)
Best,
Yang
Becket Qin 于2023年6月6日周二 10:35写道：
> +1 (binding)
>
> Thanks for driving the FLIP, Archit.
>
> Cheers,
>
> Jiangjie (Becket) Qin
>
> On Tue, Jun 6, 2023 at 4:33 AM Venkatakrishnan Sowrirajan <
> vsowr...@asu.edu>
> wrote:
>
> > Thanks for starting the vote on this one, Arc
Sorry for the late reply. I am in favor of introducing such a built-in
resource localization mechanism
based on Flink FileSystem. Then FLINK-28915[1] could be the second step
which will download
the jars and dependencies to the JobManager/TaskManager local directory
before working.
The first step
+1 for this FLIP
Maybe a FLIP is an overkill for this enhancement.
Best,
Yang
Venkatakrishnan Sowrirajan 于2023年8月23日周三 01:44写道：
> Thanks for the FLIP, Archit.
>
> This is definitely quite a useful addition to the *yarn.provided.lib.dirs*
> . +1.
>
> IMO, except for the fact that *yarn.provide
+1 (binding)
Best,
Yang
Becket Qin 于2023年9月14日周四 11:01写道：
> +1 (binding)
>
> Thanks for the FLIP, Archit.
>
> Cheers,
>
> Jiangjie (Becket) Qin
>
>
> On Thu, Sep 14, 2023 at 10:31 AM Dong Lin wrote:
>
> > Thanks Archit for the FLIP.
> >
> > +1 (binding)
> >
> > Regards,
> > Dong
> >
> > On Thu
Since the users could always use the old Flink Kubernetes Operator version
along with old Flink versions, I am totally in favor of this proposal to
reduce maintenance burden.
Best,
Yang
Biao Geng 于2023年9月6日周三 18:15写道：
> +1 for the proposal.
>
> Best,
> Biao Geng
>
> Gyula Fóra 于2023年9月6日周三 16:
+1 (binding)
Best,
Yang
On Thu, Jan 11, 2024 at 9:53 AM liu ron wrote:
> +1 non-binding
>
> Best
> Ron
>
> Matthias Pohl 于2024年1月10日周三 23:05写道：
>
> > +1 (binding)
> >
> > On Wed, Jan 10, 2024 at 3:35 PM ConradJam wrote:
> >
> > > +1 non-binding
> > >
> > > Dawid Wysakowicz 于2024年1月10日周三 21:
+1 to make flink-docker repository self-contained, including the document.
And others refer
to it.
Best,
Yang
Till Rohrmann 于2020年2月9日周日 下午5:35写道：
> Sounds good to me Patrick. +1 for these changes.
>
> Cheers,
> Till
>
> On Fri, Feb 7, 2020 at 3:25 PM Patrick Lucas
> wrote:
>
> > Hi all,
> >
+1 non-binding
- Building from source with all tests skipped
- Build a custom image with 1.10-rc3
- K8s tests
* Deploy a standalone session cluster on K8s and submit multiple jobs
* Deploy a standalone per-job cluster
* Deploy a native session cluster on K8s with/without HA configure
Hi Rong Rong,
Thanks for starting this discussion. I think the log is an important part
of improving user
experience of Flink. The logs is very important for debugging problems or
checking the
expected output. Some users, especially for machine learning, print global
steps or
residual to the log
I tend to drop the windows bat script to start/stop the cluster. Since there
are various windows versions(e.g. xp, win7, win10, etc.) and do not have
sensational compatibility. Also we do not have the e2e tests to guarantee
the scripts work well.
On the other hand, i completely agree that local de
Thanks Stephan for kicking off the 1.11 release and Zhijiang and Piotr for
the volunteering.
A short release cycle is really good. Since we will not push many features
simultaneously in
every release.
+1 for feature freeze date for end of April.
Best,
Yang
Hequn Cheng 于2020年2月20日周四 下午8:27写道：
+1 non-binding
The log files list are shown in the new design. I like the idea. Since the
user defined logs(e.g. my-log.file)
could be found here now.
Best,
Yang
Yadong Xie 于2020年2月20日周四 下午5:56写道：
> Hi all
>
> I want to start the vote for FLIP-103, which proposes to provide better
> user log
Hi Canbing,
Thanks a lot for sharing your thoughts to improve the Flink on K8s native
integration.
Frankly speaking, your discussion title confuses me and i am wondering
whether you
want to refactor the whole design. However, after i dive into the details
and the provided
document, i realize tha
I second xintong's suggestion. When i open a PR, i also check the item list
in the template. It help to
know whether i should test the PR in a real cluster(Yarn/K8s/Mesos). Or i
should be more careful
when touching the per-record code paths.If we have some dependencies
changes, i will need to check
Congratulations！Jingsong. Well deserved.
Best,
Yang
Zhijiang 于2020年2月21日周五 下午1:18写道：
> Congrats Jingsong! Welcome on board!
>
> Best,
> Zhijiang
>
> --
> From:Zhenghua Gao
> Send Time:2020 Feb. 21 (Fri.) 12:49
> To:godfrey he
>
+1 non-binding
It is really straightforward.
Best,
Yang
Jark Wu 于2020年2月21日周五 下午1:52写道：
> +1 (binding)
>
> On Fri, 21 Feb 2020 at 11:51, Kurt Young wrote:
>
> > +1 (binding)
> >
> >
> >
> > On Fri, Feb 21, 2020 at 1:09 AM Zhijiang > .invalid>
> > wrote:
> >
> > > +1 (binding).
> > > It seems
> > >
> > > > +1 (non-binding)
> > > >
> > > > I like the ideas of having a list of all log files, and make them
> > > > downloadable.
> > > >
> > > > Thank you~
> > > >
> > > > Xintong Song
>
t to make sure, it is not a fundamentally different approach compared
> to
> > what we have right now, is it?
> >
> > If this is the case, then I think it makes sense to reuse code as much as
> > possible and to create small code units which are easier to test.
> >
&
It seems that we could benefit a lot from the "Parameters Parser". Let's
start to
add the dedicated jobmanager/taskmanager config parser classes.
Best,
Yang
Canbin Zheng 于2020年2月24日周一 上午10:39写道：
> Hi, Yang Wang,
>
> Thanks for the feedback.
>
> > Paramete
I think if we could, splitting into as many PRs as possible is good. Maybe
we could
introduce the new designed decorators and parameter parser first, and leave
the existing
decorators as legacy. Once all the new decorators is ready and well tested,
we could
remove the legacy codes and use the new d
;> unit tested.
>>
>> However, it makes no sense we have two codepaths from FlinkKubeClient to
>> decorators
>> since these two version of decorators are not api compatible and there is
>> no reason we keep both
>> of them.
>>
>> Best,
>> ti
;>
>>> Thanks for your clarification Yang! We're on the same page.
>>>
>>> Best,
>>> tison.
>>>
>>>
>>> Yang Wang 于2020年2月25日周二 下午10:07写道：
>>>
>>>> Hi tison,
>>>>
>>>> I do not mean to keep two
w/edit#
>
> Please let me know what you think and please keep the discussion in the ML
> :)
>
> Thanks for starting the discussion and I hope that soon we will be
> able to vote on the FLIP.
>
> Cheers,
> Kostas
>
> On Thu, Jan 16, 2020 at 3:40 AM Yang Wang wrote:
&g
on the client
> >> while in the application mode, the main is executed on the cluster.
> >> I do not think we have to offer "application mode" with running the
> >> main on the client side as this is exactly what the per-job mode does
> >> currently an
unch
> a per job cluster with the main function executed in cluster
> entry-point. +1 for the execution plan.
>
>
>
> Best Regards
> Peter Huang
>
>
>
>
> On Tue, Mar 3, 2020 at 7:11 AM Yang Wang wrote:
>
>> Hi Peter,
>>
>> Having the applica
Hi Andrey,
Thanks for driving this significant FLIP. From the user ML, we could also
know there are
many users running Flink in container environment. Then the docker image
will be the
very basic requirement. Just as you say, we should provide a unified place
for all various
usage(e.g. session,
Hi Kostas,
Thanks a lot for kicking off the vote for FLIP-85. It is really a useful
feature for
platform developers who manage hundreds of Flink jobs on Yarn in production.
Also the native Kubernetes job cluster will benefit a lot from this.
+1 (non-binding) from my side
Best,
Yang
Kostas Klo
gt; Maybe they have been discussed in the ML earlier, but I think they should
>> be part of the FLIP also.
>>
>> Thanks,
>>
>> Jiangjie (Becket) Qin
>>
>> On Thu, Mar 5, 2020 at 10:09 PM Kostas Kloudas
>> wrote:
>>
>>> Also from my side +1 to star
CLI / configuration /
> metrics, etc. Can we update the FLIP to include the conclusion we have here
> in the ML?
>
> Thanks,
>
> Jiangjie (Becket) Qin
>
> On Mon, Mar 9, 2020 at 11:59 AM Yang Wang wrote:
>
>> Hi Becket,
>>
>> Thanks for jumping out and sha
Hi Aljoscha, Kostas,
I would be in favour of something like "bin/flink run-application",
> maybe we should even have "run-job" in the future to differentiate.
I have no preference for the "-R/--remote-deploy" option of "flink run" or
new
introduced "flink run-application". If we always bind the
Thanks for your response.
@Kostas Kloudas Could we update the cli changes and
how to fetch the
user jars to FLIP document? I think other dev or users may have the similar
questions.
Best,
Yang
Aljoscha Krettek 于2020年3月10日周二 下午9:03写道：
> On 10.03.20 03:31, Yang Wang wrote:
> > For
Thanks a lot for moving this FLIP forward. Looking forward to the JIRAs.
Best,
Yang
Flavio Pompermaier 于2020年3月12日周四 下午9:05写道：
> +1 (non-binding).
> There's also a related issue that I opened a long time ago
> https://issues.apache.org/jira/browse/FLINK-10879 that could be closed
> once
> imple
For K8s deployment(including standalone session/perjob, native
integration), Flink could
not support port range for rest options. You need to set the
`rest.bind-port` exactly same
with `rest.port`.
Hi LakeShen
I am trying to understand your problem and do not think it is about the
port configurati
docker image.
>> >> To customise this in the current mainstream image is difficult, this
>> would
>> >> require to ship it w/o Java at all.
>> >> Or this is a separate discussion whether we want to distribute docker
>> hub
>> >> images wit
+1 (non-binding).
This FLIP will make jobmanager memory configuration keep up with
taskmanager.
Best,
Yang
Xintong Song 于2020年3月19日周四 上午11:23写道：
> +1 (non-binding) from my side.
>
> Thank you~
>
> Xintong Song
>
>
>
> On Wed, Mar 18, 2020 at 10:16 PM Andrey Zagrebin
> wrote:
>
> > Hi All,
>
Hi Konstantin,
I think it is a good idea. Currently, our users also report a similar issue
with
resourceId of standalone cluster. When we start a standalone cluster now,
the `TaskManagerRunner` always generates a uuid for the resourceId. It will
be used to register to the jobmanager and not conven
Congratulations all.
Best,
Yang
Leonard Xu 于2020年4月1日周三 下午10:15写道：
> Congratulations Konstantin, Dawid and Zhijiang! Well deserved!
>
> Best,
> Leonard Xu
> > 在 2020年4月1日，21:22，Jark Wu 写道：
> >
> > Congratulations to you all!
> >
> > Best,
> > Jark
> >
> > On Wed, 1 Apr 2020 at 20:33, Kurt You
Hi Ufuk,
Thanks for make the conclusion and directly point out what need to be done
in
FLIP-111. I agree with you that we should narrow down the scope and focus
the
most important and basic part about docker image unification.
(1) Extend the entrypoint script in apache/flink-docker to start the j
Thanks Yangze for the efforts to support GPU extended resources.
+1 for this FLIP
Best,
Yang
Till Rohrmann 于2020年4月2日周四 下午11:10写道：
> Thanks for driving this effort Yangze.
>
> +1
>
> Cheers,
> Till
>
> On Wed, Apr 1, 2020 at 12:41 PM Canbin Zheng
> wrote:
>
> > Thanks Yangze for driving the
Thanks Andrey for the efforts of docker image unification.
+1 (non-binding)
Best,
Yang
Till Rohrmann 于2020年4月7日周二 下午11:04写道：
> Thanks for driving this effort Andrey.
>
> +1 (binding)
>
> Cheers,
> Till
>
> On Tue, Apr 7, 2020 at 4:48 PM Ufuk Celebi wrote:
>
> > Thanks for starting this FLIP.
Hi Niels,
Thanks a lot for starting this discussion. Although the latter option is
more stable,
i think it is not acceptable for all the developers to execute `mvn
generate-sources`
first. Otherwise, the Flink project is just broken and could not run tests,
Flink jobs
in IDE.
So i think the versi
Thanks Robert for starting this significant discussion.
Since hadoop3 has been released for long time and many companies have
already
put it in production. No matter you are using flink-shaded-hadoop2 or not,
currently
Flink could already run in yarn3(not sure about HDFS). Since the yarn api
is al
Hi jinhai,
I think it is a good improvement to also add groupingKey to
PrometheusReporter.And making
the PrometheusPushGatewayReporter and PrometheusReporter always share the
same logics
is important. AFAIK, they should not have differences except for the
metrics exposing mechanism
pulling/pushing
faster.
Best,
Yang
[1].
https://docs.google.com/document/d/1-jNzqGF6NfZuwVaFICoFQ5HFFXzF5NVIagUZByFMfBY/edit?usp=sharing
[2].
https://www.alibabacloud.com/zh/apsara-conference-2019?spm=a2c4e.11165380.1395221.13
Yang Wang 于2019年8月30日周五 下午2:23写道：
> Hi Zhenghua,
>
> You are right.
+1 to Rong’s approach.
Using java option and log4j, we could save the user logs to different file.
Best
Yang
Gyula Fóra 于2019年10月18日周五 下午4:41写道：
> Hi all!
>
> Thanks for the answers, this has been very helpful and we could set up a
> similar scheme using the Env variables.
>
> Cheers,
> Gyula
We could benefit a lot from unifying the cli options and config options.
+1(non-binding)
Best,
Yang
Aljoscha Krettek 于2019年10月23日周三 下午5:24写道：
> +1 (binding)
>
> Aljoscha
>
> > On 23. Oct 2019, at 10:41, Zili Chen wrote:
> >
> > Thanks for starting this voting process. I have looked at the FLI
Congratulations Becket :)
Best,
Yang
Vijay Bhaskar 于2019年10月29日周二 下午4:31写道：
> Congratulations Becket
>
> Regards
> Bhaskar
>
> On Tue, Oct 29, 2019 at 1:53 PM Danny Chan wrote:
>
> > Congratulations :)
> >
> > Best,
> > Danny Chan
> > 在 2019年10月29日 +0800 PM4:14，dev@flink.apache.org，写道：
> > >
>
Thanks for tison starting this exciting discussion. We also suffer a lot
from the per job mode.
I think the per-job cluster is a dedicated cluster for only one job and
will not accept more other
jobs. It has the advantage of one-step submission, do not need to start
dispatcher first and
then submit
Thanks Yu for starting this discussion.
I'm in favor of adding a e2e performance testing framework. Currently the
e2e tests are mainly focused
on functionality and written in shell. We need a better e2e framework for
performance and functionality tests.
Best,
Yang
Biao Liu 于2019年11月5日周二 上午10:1
Hi Jincheng, Hequn
I am now working on FLINK-13184 and already attach a PR to fix this issue.
I think it is
important for large scale deployment on yarn(1000+ containers). It
will accelerate TaskExecutor
launch and reduce the pressure of hdfs. I hope it could be merge to 1.8.3.
Best,
Yang
Yu Li
+1 (non-binding)
It is great to have a new end-to-end test framework, even it is only for
performance tests now.
Best,
Yang
Jingsong Li 于2019年11月19日周二 上午9:54写道：
> +1 (non-binding)
>
> Best,
> Jingsong Lee
>
> On Mon, Nov 18, 2019 at 7:59 PM Becket Qin wrote:
>
> > +1 (binding) on having the t
limitation is eliminated. We also make
> use of YARN locality so that
> reducing uploading overhead, instead, asking YARN to localize the jar on
> AM container started.
>
> Besides, it possibly has overlap with FLINK-13938. I'd like to put the
> discussion on our
> mailing list
Thanks Yangze for starting this discussion.
Just share my thoughts.
If the mesos official docker image could not meet our requirement, i
suggest to build the image locally.
We have done the same things for yarn e2e tests. This way is more flexible
and easy to maintain. However,
i have no idea how
Hi Peter,
Thanks a lot for starting this discussion. I think this is a very useful
feature.
Not only for Yarn, i am focused on flink on Kubernetes integration and come
across the same
problem. I do not want the job graph generated on client side. Instead, the
user jars are built in
a user-defined
Hi Hequn,
+1 (non-binding)
- verified checksums and hashes
- built from sources (Scala 2.11 and Scala 2.12)
- running flink per-job/session cluster on Yarn with more that 1000
containers, good
Danny Chan 于2019年12月10日周二 上午9:39写道：
> Hi Hequn,
>
> +1 (non-binding)
>
> - verified checksums and has
Graph. It can also used on yarn or mesos,not only k8s
> 2. Also It can resolve by user ‘initcontainers’ to download jar
> from remote, But It only can use in k8s .
> @Yang Wang What do think about this？ It can add a jira for this？
> ouywl
> ou...@139.com
>
> <https://m
t; On Wed, Dec 11, 2019 at 3:30 AM Hequn Cheng wrote:
>
> > Hi everyone,
> >
> > I'm happy to announce that we have unanimously approved this release.
> >
> > There are 10 approving votes, 3 of which are binding:
> > * Jincheng (binding)
> > *
Congratulations, Zhu Zhu!
wenlong.lwl 于2019年12月16日周一 上午9:56写道：
> Congratulations, Zhu Zhu!
>
> On Mon, 16 Dec 2019 at 09:14, Leonard Xu wrote:
>
> > Congratulations, Zhu Zhu ! !
> >
> > Best,
> > Leonard Xu
> >
> > > On Dec 16, 2019, at 07:53, Becket Qin wrote:
> > >
> > > Congrats, Zhu Zhu!
>
;>
> >> Best Regards
> >> Peter Huang
> >>
> >> On Mon, Dec 9, 2019 at 8:35 PM tison wrote:
> >>
> >> > > 3. What do you mean about the package? Do users need to compile
> their
> >> > jars
> >> > inlcuding flink-
You could have a try the new plugin mechanism.
Create a new directory named "myhdfs" under $FLINK_HOME/plugins, and then
put your filesystem related jars in it.
Different plugins will be loaded by separate classloader to avoid conflict.
Best,
Yang
vino yang 于2019年12月18日周三 下午6:46写道：
> Hi ouywl,
Big +1 for this effort.
It is really exciting we have started this great work. More and more
companies start to
use Flink in container environment(docker, Kubernetes, Mesos, even
Yarn-3.x). So it is
very important that we could have unified official image building and
releasing process.
The ima
>Flink runs user program on the cluster side.
>>
>> Best,
>> tison.
>>
>>
>> Peter Huang 于2019年12月19日周四 上午2:54写道：
>>
>>> Hi Yang,
>>>
>>> Thanks for your input, I can see the master side job graph generation is
>>> a
but in my case , there is no
> > jar to upload but only a sql string .
> > do u had more suggestion to improve for sql mode or it is only a
> > switch for api programme？
> >
> >
> > best
> > bai jj
> >
> >
> > Yang Wang 于2019年12月18日
> only starts the cluster.
> >
> > When {Yarn/Kuberneates}PerJobClusterEntrypoint starts, It will start the
> > dispatch first, then we can use
> > a ClusterEnvironment similar to ContextEnvironment to submit the job with
> > jobName the local
> > d
agree the jar downloading will be a painful thing.
> We can consider about PoC and performance evaluation first. If the end to
> end experience is good enough, then we can consider
> proceeding with the solution.
>
> Looking forward to more opinions from @Yang Wang @Zili
> Chen
Congratulations!
Best,
Yang
Terry Wang 于2020年1月17日周五 下午3:28写道：
> Congratulations!
>
> Best,
> Terry Wang
>
>
>
> 2020年1月17日 14:09，Biao Liu 写道：
>
> Congrats!
>
> Thanks,
> Biao /'bɪ.aʊ/
>
>
>
> On Fri, 17 Jan 2020 at 13:43, Rui Li wrote:
>
>> Congratulations Dian, well deserved!
>>
>> On Thu,
Hi everyone,
Currently Flink supports the resource management system YARN and Mesos.
However, they were not
designed for fast moving cloud native architectures, and they could not
support mixed workloads (e.g. batch,
streaming, deep learning, web services, etc.) relatively well. At the same
time
some ideas about the new features. Join in us,
let's make Flink on Kubernetes
moving forward together.
[1].
https://slack.com/share/ISUKQ2WG5/cXwlw5HZgp3AE5ElE8m580Pk/enQtOTEyNjcwMDk4NTQ5LWZiNTU0ZmJiZTU4MWU1OTk2YjllNmE0OTg2YjIxYjA2YmE1MWFlOTE4NWFhMDBkMzE4NDQzYjk1YmQwMDI2MzU
Yang Wang 于2020
Congratulations, Yu.
Best,
Yang
tison 于2020年1月23日周四 下午7:07写道：
> Congratulations!
>
> Best,
> tison.
>
>
> Dian Fu 于2020年1月23日周四 下午7:06写道：
>
> > Congrats Yu!
> >
> > > 在 2020年1月23日，下午6:47，Hequn Cheng 写道：
> > >
> > > Congratulations Yu!
> > > Thanks a lot for being the release manager of the b
+1 (non-binding)
Best,
Yang
Rong Rong 于2020年1月30日周四 上午12:53写道：
> +1
>
> --
> Rong
>
> On Wed, Jan 29, 2020 at 8:51 AM Ismaël Mejía wrote:
>
> > +1 (non-binding)
> >
> > No more maintenance work for us Patrick! Just kidding :), it was mostly
> > done by Patrick, all kudos to him.
> > Just one
>
> Thanks,
> Alexey
>
> ------
> *From:* Yang Wang
> *Sent:* Tuesday, September 1, 2020 1:58 AM
> *To:* dev
> *Cc:* Alexey Trenikhun ; Flink User Mail List <
> u...@flink.apache.org>
> *Subject:* Re: FileSystemHaServices and BlobStore
>
>
Congratulations!
Best,
Yang
Xintong Song 于2020年9月15日周二 上午10:41写道：
> Congrats!
>
> Thank you~
>
> Xintong Song
>
>
>
> On Tue, Sep 15, 2020 at 10:40 AM Zhijiang
> wrote:
>
> > Congrats, Niels!
> >
> > Best,
> > Zhijiang
> >
> >
> > ---
Hi devs and users,
I would like to start the discussion about FLIP-144[1], which will introduce
a new native high availability service for Kubernetes.
Currently, Flink has provided Zookeeper HA service and been widely used
in production environments. It could be integrated in standalone cluster,
+1 (non-binding)
* Build from source and check the signature, checksum
* Check the K8s notice file is correct since the pom has changed
* Run standalone session/application clusters on K8s
* Run native Flink session/application clusters on K8s
Best,
Yang
Dian Fu 于2020年9月15日周二 上午10:11写道：
> +1
Congratulations, Yun!
Best,
Yang
Leonard Xu 于2020年9月15日周二 下午7:11写道：
> Congrats, Yun!
>
> Best,
> Leonard
> > 在 2020年9月15日，19:01，Yangze Guo 写道：
> >
> > Congrats, Yun!
>
>
icitly release the lock so
>> that other peers can write/remove the stored object. What if the previous
>> owner failed to release the lock (e.g., dead before releasing)? Would there
>> be any problem?
>>
>> ## HA storage > HA data clean up
>>
>> If the
confluence/display/FLINK/FLIP-144%3A+Native+Kubernetes+HA+for+Flink
Best,
Yang
Alexey Trenikhun 于2020年9月3日周四 下午11:44写道：
> Hi Yang,
> Yes, I’ve persisted CompletedCheckpointStore, CheckpointIDCounter and
> RunningJobsRegistry
>
> Thanks,
> Alexey
>
> -----
Hi Xu Di,
Sorry for the inconvenience. This is a known issue FLINK-17857[1]. After we
migrate all the e2e tests
from flink-container/docker to apache/flink-docker, the docker and k8s
related e2e tests could not work.
I will revive this ticket and fix it ASAP.
[1]. https://issues.apache.org/jira/b
/blob/6d83d41d50941bf8f2d4e0c859951eb10f617df6/kubernetes-client/src/main/java/io/fabric8/kubernetes/client/extended/leaderelection/LeaderElector.java#L70
Best,
Yang
Till Rohrmann 于2020年9月28日周一 下午9:29写道：
> Thanks for creating this FLIP Yang Wang. I believe that many of our users
> will like a ZooKeeper-less HA setup.
>
g the
>> checkpoints and job blobs like in the ZooKeeper case. The current benefit
>> I
>> see is that we avoid having to implement this multi locking mechanism in
>> the ConfigMaps using the annotations because we can be sure that there is
>> only a single leader at a t
e-stateful-set-pod/#force-deletion
Best,
Yang
Till Rohrmann 于2020年9月30日周三 下午11:57写道：
> Thanks for the clarifications Yang Wang.
>
> 2. Keeping the HA information relevant for a component (Dispatcher,
> JobManager, ResourceManager) in a single ConfigMap sounds good. We should
> check
> avoided
> > the "lock-and-release" in the implementation. So I still insist on using
> > the current Deployment.
> >
> >
> > [1].
> >
> https://kubernetes.io/docs/tasks/run-application/force-delete-stateful-set-pod/#force-deletion
> >
> >
>
1 - 100 of 448 matches
Mail list logo
