Thanks everyone! I'll then go ahead and cut the branch early next week. This will give us enough time to set up all the necessary jenkins builders before the voting starts. We can then start the formal release process a week after (8th January) once everybody gets back from vacation.


On Thu, Dec 21, 2017 at 8:48 PM, Kazuaki Ishizaki <> wrote:
+1 for cutting a branch earlier.
In some Asian countries, 1st, 2nd, and 3rd January are off.
How about 4th or 5th?

Kazuaki Ishizaki

From:        Felix Cheung <>
To:        Michael Armbrust <>, Holden Karau <>
Cc:        Sameer Agarwal <>, Erik Erlandson <>, dev <>
Date:        2017/12/21 04:48
Subject:        Re: Timeline for Spark 2.3

I think the earlier we cut a branch the better.

From: Michael Armbrust <>
Tuesday, December 19, 2017 4:41:44 PM
Holden Karau
Sameer Agarwal; Erik Erlandson; dev
Re: Timeline for Spark 2.3

Do people really need to be around for the branch cut (modulo the person cutting the branch)?

1st or 2nd doesn't really matter to me, but I am +1 kicking this off as soon as we enter the new year :)


On Tue, Dec 19, 2017 at 4:39 PM, Holden Karau <> wrote:
Sounds reasonable, although I'd choose the 2nd perhaps just since lots of folks are off on the 1st?

On Tue, Dec 19, 2017 at 4:36 PM, Sameer Agarwal <> wrote:
Let's aim for the 2.3 branch cut on 1st Jan and RC1 a week after that (i.e., week of 8th Jan)?

On Fri, Dec 15, 2017 at 12:54 AM, Holden Karau <> wrote:
So personally I’d be in favour or pushing to early January, doing a release over the holidays is a little rough with herding all of people to vote.

On Thu, Dec 14, 2017 at 11:49 PM Erik Erlandson <> wrote:
I wanted to check in on the state of the 2.3 freeze schedule.  Original proposal was "late Dec", which is a bit open to interpretation.

We are working to get some refactoring done on the integration testing for the Kubernetes back-end in preparation for testing upcoming release candidates, however holiday vacation time is about to begin taking its toll both on upstream reviewing and on the "downstream" spark-on-kube fork.

If the freeze pushed into January, that would take some of the pressure off the kube back-end upstreaming. However, regardless, I was wondering if the dates could be clarified.

On Mon, Nov 13, 2017 at 5:13 PM,<> wrote:

What is the process to request an issue/fix to be included in the next
release? Is there a place to vote for features?
I am interested in, to see
if we can get Spark upgrade parquet to 1.9.0, which addresses the
Can we include the fix in Spark 2.3 release?



Sent from:

To unsubscribe e-mail:


Sameer Agarwal
Software Engineer | Databricks Inc.


Sameer Agarwal
Software Engineer | Databricks Inc.