Gyula Fora created FLINK-27569:
--
Summary: Terminated Flink job restarted from empty state when
execution.shutdown-on-application-finish is false
Key: FLINK-27569
URL: https://issues.apache.org/jira/browse/FLINK-27569
I have opened a JIRA: https://issues.apache.org/jira/browse/FLINK-27569
with more details, logs and screenshots.
Gyula
On Wed, May 11, 2022 at 7:03 AM Gyula Fóra wrote:
> Hi Devs!
>
> I ran into a concerning situation and would like to hear your thoughts on
> this.
>
> I am running Flink 1.15 o
Hi Sebastian,
Thanks for the reply and patient discussion. I agree this is a tricky
decision.
> Nevertheless, Piotr has valid concerns about Option c) which I see as
> follows:
> (1) An interface with default NOOP implementation makes the implementation
> optional. And in my opinion, a default i
Really excited to see the very first release of the flink-table-store!
Kudos to everyone who helped with this effort!
Cheers,
Jiangjie (Becket) Qin
On Wed, May 11, 2022 at 1:55 PM Jingsong Lee
wrote:
> The Apache Flink community is very happy to announce the release of Apache
> Flink Table S
I assume this is the responsibility of job result store[1]. However, it
seems that it does not work as expected.
[1].
https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=195726435
Best,
Yang
Gyula Fóra 于2022年5月11日周三 12:55写道：
> Sorry I messed up the email, I meant false .
>
> So wh
Thanks for your warm welcome. It is my pleasure to work in such a nice
community.
Best,
Yang
Thomas Weise 于2022年5月11日周三 00:10写道：
> Congratulations, Yang!
>
> On Tue, May 10, 2022 at 3:15 AM Márton Balassi
> wrote:
> >
> > Congrats, Yang. Well deserved :-)
> >
> > On Tue, May 10, 2022 at 9:1
Underwood created FLINK-27570:
-
Summary: A checkpoint path error does not cause the job to stop
Key: FLINK-27570
URL: https://issues.apache.org/jira/browse/FLINK-27570
Project: Flink
Issue Type:
Roman Khachatryan created FLINK-27571:
-
Summary: Recognize "less is better" benchmarks in regression
detection script
Key: FLINK-27571
URL: https://issues.apache.org/jira/browse/FLINK-27571
Projec
Gyula Fora created FLINK-27572:
--
Summary: Verify HA Metadata present before performint last-state
restore
Key: FLINK-27572
URL: https://issues.apache.org/jira/browse/FLINK-27572
Project: Flink
Thanks, Thomas.
I noticed this issue and asked Jiang for details.
If he doesn't finish this work yet, I will draft the initial FLIP,
otherwise we can wait for Jiang's implementation.
Thomas Weise 于2022年5月11日周三 02:29写道：
> fyi there is a related ticket here:
> https://issues.apache.org/jira/brow
I don't think we can maintain two additional channels. Some people have
already concerns about covering one additional channel.
I think, a forum provides a better user experience than a mailing list.
Information is structured better, you can edit messages, sign up and search
is easier.
To make so
Matyas Orhidi created FLINK-27573:
-
Summary: Configuring a new random job result store directory
Key: FLINK-27573
URL: https://issues.apache.org/jira/browse/FLINK-27573
Project: Flink
Issue T
tanjialiang created FLINK-27574:
---
Summary: [QUESTION] In Flink k8s Application mode with HA can not
using History Server for history backend
Key: FLINK-27574
URL: https://issues.apache.org/jira/browse/FLINK-27574
Hi Jark,
Thanks a lot for your opinions and suggestions! Please see my replies inline.
> 1) the display of savepoint_path
Agreed. Adding it to the FLIP.
> 2) Please make a decision on multiple options in the FLIP.
Okay. I’ll keep one and move the other to the rejected alternatives section.
+dev
Hi Sebastian,
Thank you for the summary. Please see the detailed replies inline. As a
recap of my suggestions.
1. Pausable splits API.
a) Add default implementations to methods "pauseOrResumeSplits" in both
SourceReader and SplitReader where both default implementations throw
Unsupported
Huang Xingbo created FLINK-27575:
Summary: Support PyFlink in Python3.9
Key: FLINK-27575
URL: https://issues.apache.org/jira/browse/FLINK-27575
Project: Flink
Issue Type: New Feature
In Flink k8s application mode with high-availability, it's job id always
00, but in history server, it make job's id for the key. Can I modify
the job id in HA? Maybe the History Server must keep the ClusterID and the
JobID for it's key?
Best,
tanjialiang.
| |
tanjialiang
|
|
tanjl
zhisheng created FLINK-27576:
Summary: Flink will request new pod when jm pod is delete, but
will remove when TaskExecutor exceeded the idle timeout
Key: FLINK-27576
URL: https://issues.apache.org/jira/browse/FLINK-2
Hi Becket,
Thanks a lot for your fast and detailed response. For me, it converges and
dropping the supportsX method sounds very reasonable to me. (Side note:
With "pausable splits" enabled as "default" I think we misunderstood. As
you described now "default" I understand as that it should be the n
Hi,
Actually previously I thought about having a decorative interface and
whenever watermark alignment is enabled, checking that the source
implements the decorative interface. If not, throwing an exception.
The option with default methods in the source interfaces throwing
`UnsupportedOperationEx
Sorry, I found this is not a the right place to ask some question. I'll ask
question in the u...@flink.apache.org.
| |
谭家良
|
|
tanjl_w...@126.com
|
On 5/11/2022 19:52，谭家良 wrote：
In Flink k8s application mode with high-availability, it's job id always
00, but in history server, it make j
Hi everyone,
As mentioned in the previous update [1] we were working on the final step
for moving out the Elasticsearch connector, which was related to the
documentation integration. This has now been completed: the documentation
you see on
https://nightlies.apache.org/flink/flink-docs-master/docs
Hi Martijn,
+1 to do a release which is compatible with Flink 1.5.x. With the release,
we should add something like a compatibility matrix to the documentation,
but I am sure that's already on your list.
Cheers,
Konstantin
Am Mi., 11. Mai 2022 um 20:10 Uhr schrieb Martijn Visser <
martijnvis
Hi Folks,
Getting this error when sinking data to a firehosesink, would really
appreciate some help !
DataStream inputStream = env.addSource(new
FlinkKafkaConsumer<>("xxx", new SimpleStringSchema(), properties));
Properties sinkProperties = new Properties();
sinkProperties.put(AWSCon
Hi all,
If there are no more comments, I'm going to start a vote.
Best,
Jingsong
On Tue, May 10, 2022 at 10:37 AM Jingsong Li wrote:
> Hi Jark,
>
> Thanks for your feedback.
>
> > 1) Does table-store support evolve schemas multiple times during a
> checkpoint?
>
> In this case this checkpoint
Thanks for the clarification, Piotr and Sebastian.
It looks like the key problem is still whether the implementation of
pausable splits in the Sources should be optional or not.
I think it might be helpful to agree on the definition of optional in our
case. To me:
Optional = "You CAN leave the me
>
> To make some progress, maybe we decide on chat vs forum vs none and then
> go into a deeper discussion on the implementation or is there anything
> about Slack that would be complete blocker for the implementation?
>
Sure, then I'd be +1 for chat. From my side, the initiative is more about
mak
Hi all,
Regarding using ASF slack. I share the problems I saw in the Apache Druid
community. [1]
> As you may have heard, it’s become increasingly difficult for new users
without an @apache.org email address to join the ASF #druid Slack channel.
ASF Infra disabled the option to publicly provide a
Hi all,
I think forum might be a good choice for search and maintain. However, unlike
slack workspace, it seems no existing popular product could be leveraged easily.
Thus, I am +1 to create an Apache Flink slack channel. If the ASF slack cannot
be joined easily for most of users, I prefer to s
Hi Jark!
Sorry for the late response. I would like to make some comments and
clarify my points.
1) I agree with your first statement. I think we can achieve both
advantages this way: put the Cache interface in flink-table-common,
but have implementations of it in flink-table-runtime. Therefore if
30 matches
Mail list logo
