Re: [VOTE] Apache Helix 0.8.3 Release

[Olivier Lamy]

+1

On Tue, 5 Feb 2019 at 08:23, Xue Junkai  wrote:

> Hi,
>
>
> This is to call for a vote on releasing the following candidate as Apache
> Helix 0.8.3. This is the 15th release of Helix as an Apache project, as
> well as the 11th release as a top-level Apache project.
>
>
> Apache Helix is a generic cluster management framework that makes it easy
> to build partitioned and replicated, fault-tolerant and scalable
> distributed systems.
>
>
> Release notes:
>
> *https://helix.apache.org/0.8.3-docs/releasenotes/release-0.8.3.html
> *
>
>
> Release artifacts:
>
> https://repository.apache.org/content/repositories/orgapachehelix-1022
>
>
> Distribution:
>
> * binaries:
>
> *https://dist.apache.org/repos/dist/dev/helix/0.8.3/binaries/
> *
>
> * sources:
>
> https://dist.apache.org/repos/dist/dev/helix/0.8.3/src/
>
>
> The 0.8.3 release tag:
>
>
> https://git-wip-us.apache.org/repos/asf?p=helix.git;a=tag;h=refs/tags/helix-
> 0.8.3
>
>
> KEYS file available here:
>
> https://dist.apache.org/repos/dist/dev/helix/KEYS
>
>
> Please vote on the release. The vote will be open for at least 72 hours.
>
>
> [+1] -- "YES, release"
>
> [0] -- "No opinion"
>
> [-1] -- "NO, do not release"
>
>
> Thanks,
>
> The Apache Helix Team
>


-- 
Olivier Lamy
http://twitter.com/olamy | http://linkedin.com/in/olamy

[GitHub] helix issue #300: Daemonize SubscribeChangeEventProcessor thread in Callback...

[jiajunwang]

Github user jiajunwang commented on the issue:

https://github.com/apache/helix/pull/300
  
I don't think we can simply make this thread daemon. Actually, it only 
makes sense when all the handlers are finalized. If one or more handlers are 
still active, the re-subscribe function is required (and shall not be treated 
as a daemon).


---

[GitHub] helix pull request #299: Remove unused _forceRebalanceTimer from GenericHeli...

[mebe]

GitHub user mebe opened a pull request:

https://github.com/apache/helix/pull/299

Remove unused _forceRebalanceTimer from GenericHelixController

`GenericHelixController. _forceRebalanceTimer` is a non-daemon `Timer` 
thread. It is started in the initializer of `GenericHelixController` and never 
shut down.

This causes process JVM shutdown to never finish as this thread is hanging.

However, ` _forceRebalanceTimer` is never actually used for anything. I 
propose fixing the issue by removing the timer thread entirely.

You can merge this pull request into a Git repository by running:

$ git pull https://github.com/mebe/helix remove-_forceRebalanceTimer

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/helix/pull/299.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #299


commit 8752c65fdb71247413d44b70fafcf39b2b078f84
Author: Iikka Niinivaara 
Date:   2019-02-05T18:22:22Z

Remove unused _forceRebalanceTimer from GenericHelixController




---

[GitHub] helix pull request #300: Daemonize SubscribeChangeEventProcessor thread in C...

[mebe]

GitHub user mebe opened a pull request:

https://github.com/apache/helix/pull/300

Daemonize SubscribeChangeEventProcessor thread in CallbackProcessor

`CallbackProcessor.SubscribeChangeEventProcessor` is a `DedupEventProcessor 
extends Thread`. It is started in a static initializer in `CallbackProcessor` 
and never shut down.

This causes process JVM shutdown to never finish as this thread is hanging.

There is no obvious place to shut down the `SubscribeChangeEventProcessor` 
thread. I propose fixing this by making it a daemon thread. This causes the 
thread to be abandoned at shutdown. But this might have some unwanted side 
effects?

You can merge this pull request into a Git repository by running:

$ git pull https://github.com/mebe/helix 
daemonize-SubscribeChangeEventProcessor

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/helix/pull/300.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #300


commit fd7b28fece994e7b22ae0428faae9da203b24ee4
Author: Iikka Niinivaara 
Date:   2019-02-05T18:12:22Z

Daemonize SubscribeChangeEventProcessor thread in CallbackProcessor




---