tinkerpop-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stephen Mallette <spmalle...@gmail.com>
Subject Re: Code Freeze 3.1.8/3.2.6/3.3.0
Date Fri, 25 Aug 2017 18:08:59 GMT
Hi all, Ted and I are finishing up release stuff now that all votes are
closed. There were some hassles with pypi - apparently some APIs became
deprecated and i had to fix some things in various configurations. Anyway,
they finally published but I'll have some updates to make to docs/poms
after this. Python ecosystem is so weird to me still.

It will take some time to get all the mirrors to propagate all the
artifacts we're hanging out there this afternoon so I think we'll just wait
until monday to make an official announcement for release. I'll wait until
monday too to update the web site, then it can all happen at once.

Release branches should all be re-opened shortly. I just have Gremlin
DotNet release candidate to do and then i'll bump to snapshot. I'll post
back here when that's done

On Mon, Aug 21, 2017 at 10:31 AM, Stephen Mallette <spmallette@gmail.com>

> Just a heads up that we're in the process of getting release artifacts
> available.
> On Wed, Aug 16, 2017 at 10:44 AM, Stephen Mallette <spmallette@gmail.com>
> wrote:
>> Just a quick update, I've managed to publish snapshot docs for 3.3.0 and
>> 3.2.6. Things are still a little mysterious as to why things were broken,
>> but here's my best guess of things that caused problems:
>> 1. there was a changed required to the doc post processor for python - a
>> recent renaming of a python class caused a break
>> 2. giraph 1.2.0 seems to hang hadoop. so i reverted. it not only broke
>> for docs, but i also couldn't get even a job to run manually from the
>> console with that 1.2.0 version
>> 3. the hang seemed to end hadoop completely (i.e. no other job would run
>> after that failure) which led me to believe there were problems on tp32
>> when i went to generate docs over there. i managed to notice that a restart
>> of hadoop between failures seemed to clear things up.
>> none of the above explains the hangs we've also seen in docker.......of
>> course that seemed to only happen in some environments. some of you have
>> mentioned to me privately that docker works on tp32. i've not heard of a
>> working run on master yet though. i'm trying tp32 with docker now..
>> going forward i hope to run doc builds more often. it stinks to got his
>> deep into code freeze focused getting this working. would have liked to
>> have spend more of this week on testing and documentation review.
>> On Mon, Aug 14, 2017 at 12:05 PM, Ted Wilmes <twilmes@gmail.com> wrote:
>>> Good morning,
>>> The 3.1.8 SNAPSHOT docs and jars have been deployed.
>>> Thanks,
>>> Ted
>>> On Sat, Aug 12, 2017 at 5:37 AM, Stephen Mallette <spmallette@gmail.com>
>>> wrote:
>>> > For the next week we're in code freeze on tp31, tp32 and master. Let's
>>> > spend the next week testing and reviewing documentation. I've deployed
>>> the
>>> > latest SNAPSHOTs for 3.2.6 and 3.3.0. As Ted is handling 3.1.8, I
>>> assume
>>> > he'll post a reply here when he does the same for that version.
>>> >
>>> > I know there are a couple of PRs still open that Robert Dale was
>>> trying to
>>> > see through:
>>> >
>>> > https://github.com/apache/tinkerpop/pull/692 (netty version bump)
>>> > https://github.com/apache/tinkerpop/pull/680 (neo4j version bump)
>>> >
>>> > Personally, I feel reasonably comfortable with getting the netty one
>>> in.
>>> > Not so sure how the neo4j one will work.
>>> >
>>> > Anyway, as usual, let's use this thread to keep track of issues that
>>> pop up
>>> > during code freeze week.
>>> >

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message