This vote will remain open for 96 hours (4 days, because I'm getting this out so late today, I doubt anybody will look at it tonight), until Sun, April 20, 02:00 UTC 2014. (That's Saturday, April 19, 10pm EDT.)
[ ] +1 - I have verified and accept... [ ] +0 - I have reservations, but not strong enough to vote against... [ ] -1 - Because..., I do not accept... ... these artifacts as the 1.6.0 release of Apache Accumulo.
The removal of .gitignore is something that is done upstream, by the Apache source release assembly and related plugins.
I think it's an acceptable and expected change from the behavior of previous behavior, as the .gitignore file has no meaning in the source tarball, and is only useful inside the git repo itself. In other words, it's not part of our source, so much as part of our SCM. I, for one, prefer it gone, as it is now. I couldn't find the specific upstream change/justification for removing it from the assembly, though.
Also, FYI, if you want to run a minimal set of ITs with the unit tests, you can do: "mvn verify -P sunny"
* Built and tested (unit and integration) src tarball against Apache Hadoop 2.2.0, 2.3.0 and 2.4.0. * Verified src and bin tarballs hash/sigs * Verified performance fix in place after RC1 (ACCUMULO-2668) * Ran Pig jobs * Checked over user manual
I was getting ready to +1 before Christopher found 2700
I did the following checks * sigs @ hashes of bin.tgz are ok * ran through a few examples and found ACCUMULO-2691 * upgrade 1.5.1 to 1.6.0 RC2 went ok * built native lib ok
Ran the following test on 20 node cluster * 24 hr ci w/o ag RC1 + ACCUMULO_2668 patch (~31B k/v verified clean, ran into ACCUMULO-2677) * 24 hr RW conditional w/o agitation RC2 (all 20 walkers ran until I killed them) * 24 hr RW conditional w/ agitation RC2 and found ACCUMULO-2695 (7 of 17 walkers hung) On Tue, Apr 15, 2014 at 9:57 PM, Christopher <[EMAIL PROTECTED]> wrote: