[jira] [Created] (FLINK-10078) Dispatcher should only block job submission during recovery

Joey Echeverria created FLINK-10078: --- Summary: Dispatcher should only block job submission during recovery Key: FLINK-10078 URL: https://issues.apache.org/jira/browse/FLINK-10078 Project: Flink

[jira] [Created] (FLINK-10077) Web UI should respond when there is no elected Dispatcher

Joey Echeverria created FLINK-10077: --- Summary: Web UI should respond when there is no elected Dispatcher Key: FLINK-10077 URL: https://issues.apache.org/jira/browse/FLINK-10077 Project: Flink

[jira] [Created] (FLINK-10076) Upgrade Calcite dependency to 1.18

Shuyi Chen created FLINK-10076: -- Summary: Upgrade Calcite dependency to 1.18 Key: FLINK-10076 URL: https://issues.apache.org/jira/browse/FLINK-10076 Project: Flink Issue Type: Task

Re: Setting an allowable number of checkpoint failures

Thanks Thomas for opening the issue. This is indeed a useful feature to make the checkpointing more controllable. On Mon, Aug 6, 2018 at 6:36 PM Thomas Weise wrote: > Hi vino, > > Yes, I believe we are on the same page. I created > https://issues.apache.org/jira/browse/FLINK-10074 to track it.

Re: [VOTE] Release 1.6.0, release candidate #2

I think so far, we did not do much manual testing with respect to TTL. There is also an end-to-end test covering this area. We actually intended to include in the second RC but it was merged just after starting the creation procedure was started. Therefore I would like to include it if there is no

[jira] [Created] (FLINK-10075) HTTP connections to a secured REST endpoint flood the log

Stephan Ewen created FLINK-10075: Summary: HTTP connections to a secured REST endpoint flood the log Key: FLINK-10075 URL: https://issues.apache.org/jira/browse/FLINK-10075 Project: Flink

Re: Setting an allowable number of checkpoint failures

Hi vino, Yes, I believe we are on the same page. I created https://issues.apache.org/jira/browse/FLINK-10074 to track it. Thanks, Thomas On Mon, Aug 6, 2018 at 8:42 AM vino yang wrote: > Hi Thomas, > > What I am saying is what you mean, maybe I am not very accurate. > > Thanks, vino. > >

[jira] [Created] (FLINK-10074) Allowable number of checkpoint failures

Thomas Weise created FLINK-10074: Summary: Allowable number of checkpoint failures Key: FLINK-10074 URL: https://issues.apache.org/jira/browse/FLINK-10074 Project: Flink Issue Type:

Re: Setting an allowable number of checkpoint failures

Hi Thomas, What I am saying is what you mean, maybe I am not very accurate. Thanks, vino. 2018-08-06 21:22 GMT+08:00 Thomas Weise : > Hi, > > What we are looking for is that the job does *not* restart on transient > checkpoint failures and we would like to cap the number of allowable >

Re: [VOTE] Release 1.6.0, release candidate #2

TBH I would prefer if we'd only include fixes for release blocking issues in follow-up RCs. If you now include state TTL improvemements we effectively invalidate any testing done so far in this area. On 06.08.2018 17:17, Till Rohrmann wrote: Yes, the next RC will also include FLINK-9938.

Re: [VOTE] Release 1.6.0, release candidate #2

Yes, the next RC will also include FLINK-9938. On Mon, Aug 6, 2018 at 5:03 PM Aljoscha Krettek wrote: > Does this mean https://issues.apache.org/jira/browse/FLINK-9938 < > https://issues.apache.org/jira/browse/FLINK-9938> can move to 1.6.0 from > 1.6.1 as well? > > > On 6. Aug 2018, at 17:00,

Re: [VOTE] Release 1.6.0, release candidate #2

Does this mean https://issues.apache.org/jira/browse/FLINK-9938 can move to 1.6.0 from 1.6.1 as well? > On 6. Aug 2018, at 17:00, Till Rohrmann wrote: > > Officially cancelling this RC in order to fix FLINK-10070. I will publish > the next RC

Re: [VOTE] Release 1.6.0, release candidate #2

Officially cancelling this RC in order to fix FLINK-10070. I will publish the next RC later today. Cheers, Till On Mon, Aug 6, 2018 at 3:50 PM Till Rohrmann wrote: > I agree with Chesnay and think that we should fix FLINK-10070 for 1.6. A > PR fixing this issue is already open and can be

Re: [VOTE] Release 1.6.0, release candidate #2

I agree with Chesnay and think that we should fix FLINK-10070 for 1.6. A PR fixing this issue is already open and can be merged in half an hour. Since this change only affects the build, I would like to create a new RC with a shortened voting period until Wednesday evening (CET) which would end

[jira] [Created] (FLINK-10073) Allow setting a restart strategy in SQL Client

Timo Walther created FLINK-10073: Summary: Allow setting a restart strategy in SQL Client Key: FLINK-10073 URL: https://issues.apache.org/jira/browse/FLINK-10073 Project: Flink Issue Type:

Re: Setting an allowable number of checkpoint failures

Hi, What we are looking for is that the job does *not* restart on transient checkpoint failures and we would like to cap the number of allowable subsequent failures until a restart occurs. The reason is that every restart is a service interruption that is potentially very expensive. Thanks,

[jira] [Created] (FLINK-10072) Syntax and consistency issues in "The Broadcast State Pattern" documentation

Rick Hofstede created FLINK-10072: - Summary: Syntax and consistency issues in "The Broadcast State Pattern" documentation Key: FLINK-10072 URL: https://issues.apache.org/jira/browse/FLINK-10072

Re: Setting an allowable number of checkpoint failures

Hi Till, I think the way you proposed is a solution. But I think we also can provide a solution to prevent Checkpoint from failing indefinitely, in case the Job does not fail. Instead, a threshold is given to allow the checkpoint to fail a few times. When this threshold is reached, we decide to

[jira] [Created] (FLINK-10071) Document usage of INSERT INTO in SQL Client

Timo Walther created FLINK-10071: Summary: Document usage of INSERT INTO in SQL Client Key: FLINK-10071 URL: https://issues.apache.org/jira/browse/FLINK-10071 Project: Flink Issue Type:

[jira] [Created] (FLINK-10070) Flink cannot be compiled with maven 3.0.x

Chesnay Schepler created FLINK-10070: Summary: Flink cannot be compiled with maven 3.0.x Key: FLINK-10070 URL: https://issues.apache.org/jira/browse/FLINK-10070 Project: Flink Issue

Re: [VOTE] Release 1.6.0, release candidate #2

Potential release blocked: Flink cannot be compiled with maven 3.0.X https://issues.apache.org/jira/browse/FLINK-10070 On 06.08.2018 11:21, Till Rohrmann wrote: Sure, only bugs will be fixed in 1.6.1. On Mon, Aug 6, 2018 at 10:34 AM Aljoscha Krettek > wrote:

[jira] [Created] (FLINK-10069) Add docs for updates SSL model

Stephan Ewen created FLINK-10069: Summary: Add docs for updates SSL model Key: FLINK-10069 URL: https://issues.apache.org/jira/browse/FLINK-10069 Project: Flink Issue Type: New Feature

Re: [VOTE] Release 1.6.0, release candidate #2

Sure, only bugs will be fixed in 1.6.1. On Mon, Aug 6, 2018 at 10:34 AM Aljoscha Krettek wrote: > I don't think the fixVersion of all unresolved issues should be updated to > 1.6.1. If they are not bugs they will not be fixed in 1.6.1 anyways so they > should be updated to 1.7.0 or we can also

[jira] [Created] (FLINK-10068) Add documentation for async/RocksDB-based timers

Stefan Richter created FLINK-10068: -- Summary: Add documentation for async/RocksDB-based timers Key: FLINK-10068 URL: https://issues.apache.org/jira/browse/FLINK-10068 Project: Flink Issue

[jira] [Created] (FLINK-10067) Add Kafka 1.0.0/1.1.0 connectors

Aljoscha Krettek created FLINK-10067: Summary: Add Kafka 1.0.0/1.1.0 connectors Key: FLINK-10067 URL: https://issues.apache.org/jira/browse/FLINK-10067 Project: Flink Issue Type:

Re: [VOTE] Release 1.6.0, release candidate #2

I don't think the fixVersion of all unresolved issues should be updated to 1.6.1. If they are not bugs they will not be fixed in 1.6.1 anyways so they should be updated to 1.7.0 or we can also remove the fixVersion from them entirely. > On 6. Aug 2018, at 10:16, Till Rohrmann wrote: > > I

Re: [VOTE] Release 1.6.0, release candidate #2

I think you're right that it is better to change the fixVersion otherwise some issue might get wrongly attributed when they are merged in the release branch after creating the RC. I will update the fixVersion of all unresolved issues to 1.6.1. Cheers, Till On Mon, Aug 6, 2018 at 9:56 AM Chesnay

[jira] [Created] (FLINK-10066) Keep only archived version of previous executions

Stefan Richter created FLINK-10066: -- Summary: Keep only archived version of previous executions Key: FLINK-10066 URL: https://issues.apache.org/jira/browse/FLINK-10066 Project: Flink Issue

Re: [VOTE] Release 1.6.0, release candidate #2

Actually, this is defined in the release guide. https://cwiki.apache.org/confluence/display/FLINK/Creating+a+Flink+Release#CreatingaFlinkRelease-ReviewReleaseNotesinJIRA On 06.08.2018 09:11, Till Rohrmann wrote: I'm not strictly sure whether there must not be any issues with a fixVersion of

[jira] [Created] (FLINK-10065) InstantiationUtil.deserializeObject(InputStream in, ClassLoader cl, boolean isFailureTolerant) will close the inputStream

Congxian Qiu created FLINK-10065: Summary: InstantiationUtil.deserializeObject(InputStream in, ClassLoader cl, boolean isFailureTolerant) will close the inputStream Key: FLINK-10065 URL:

Re: Setting an allowable number of checkpoint failures

Hi Lakshmi, you could somewhat achieve the described behaviour by setting setFailOnCheckpointintErrors(true) and using the FailureRateRestartStrategy as the restart strategy. That way checkpoint failures will trigger a job restart (this is the downside) which is handled by the restart strategy.

Re: [VOTE] Release 1.6.0, release candidate #2

I'm not strictly sure whether there must not be any issues with a fixVersion of the current RC. At least in the past, we did not do it like this. Moreover, if a RC is canceled some of these issues might still go in the actual release. However, I also see the point that the release notes are