Joe,
+1 (binding).
Verified signature, built with mvn -Pcontrib (success), built with -Pmapr
(non-blocking failure on range processors - will raise JIRA), performed a
rolling upgrade into secure test cluster (success with no side effect),
processed a few GB of generated data (using multiple
+1 (binding)
I downloaded all of the artifacts, verified the GPG signature (nice job using
SHA256 for the underlying hash, Joe), verified all checksums, built
successfully (all tests and contrib-check passed on OS X 10.11.6 with Java
1.8.0_101-b13 and Maven 3.3.9), ran the default instance and
regarding the commit hash I meant to indicate that was the has the RC
was initiated off of. I should have used the one that the tag was
made off of to be more consistent. My apologies for the confusion.
Thanks
Joe
On Mon, Nov 28, 2016 at 3:48 PM, Bryan Rosander wrote:
>
+1 (non-binding)
* Signature, hashes are good
* Full contrib-check build with empty .m2 works on macOS, Windows 10
* tls-toolkit secured 3-node cluster comes up fine, can set permissions,
create, start flow after changing relevant paths to 1.1.0 [1]
Can confirm readme issues seen by Jeff and Joe
+1 (non-binding)
Went through the release guide, everything looked good with the exception
of a couple nit-picks:
git commit hash for RC2 is 5536f690a81418955442d52687695f65f0a44cd0 instead
of f61e42c65e1c2387591ee368c2148cd5bda646bd
README.md references version 1.0.0 rather than 1.1.0 in the
+1 (non-binding) but there are README issues
* Signature and hashes checkout
* Source builds and tests (Ubuntu 14.04.5 LTS)
* LICENSE and NOTICE look correct (without a detailed analysis)
* README is incorrect [1]
* binaries run as expected
[1] the built README lists requirements of "JDK 1.7 or
+1
Ran verifications, and a sample data flow.
On Mon, Nov 28, 2016 at 12:25 PM, Joe Percivall <
joeperciv...@yahoo.com.invalid> wrote:
> +1 (binding) Release this as nifi-1.1.0
>
> - - - - - - Joseph Percivalllinkedin.com/in/Percivalle:
> joeperciv...@yahoo.com
>
>
> On Monday, November 28,
+1 (binding) Release this as nifi-1.1.0
- - - - - - Joseph Percivalllinkedin.com/in/Percivalle: joeperciv...@yahoo.com
On Monday, November 28, 2016, 10:25:53 AM EST, Bryan Bende
wrote:+1 (binding) Release this package as nifi-1.1.0
- Verified everything in release helper
-
+1 (binding) Release this package as nifi-1.1.0
- Verified everything in release helper
- Ran the resulting binary in standalone mode secure and unsecure and
verified a couple of test flows
The original email said the git commit was
f61e42c65e1c2387591ee368c2148cd5bda646bd
but the commit that
+1 (non-binding)
- Verified signatures, checksums, LICENSE, NOTICE, and README
- Built using a full clean build with contrib check successfully on OSX,
with Oracle JDK 1.8.0_77-b03, mvn 3.3.9
- Run example data flows on OS X with Chrome Version 54.0.2840.98 (64-bit) on
unsecured instance
- Run
+1 (binding)
Verified md5 and signing.
Verified application started up and functioned normally.
Performed build with contrib-check.
I did run into a test failure in the DBCPConnectionService. I've not seen
anyone else report this yet,
so it may just be a weird environment that I have, perhaps.
+1 Release this package as nifi-1.1.0
Ran through numerous configurations including standalone and clustered in
both secure and unsecured mode and all is working as excepted.
Matt
On Sat, Nov 26, 2016 at 1:11 AM, Joe Witt wrote:
> Hello Apache NiFi Community,
>
> I am
+1 (non-binding)
Verified signatures, checksums, LICENSE, NOTICE, and README
Build and deployment succeeded on CentOS 6.8 (OpenJDK 1.8.0_111-b15)
Successfully tested a complex flow on a 3-node secured cluster with
embedded ZK with the following features:
* Self connections over site-to-site
*
+1 (non-binding)
Full build with contrib-check on OS X and CentOS 7. (Apache Maven 3.3.9 /
Java version: 1.8.0_77)
Classic flows tested on unsecured and secured standalone instance, and on
unsecured cluster with embedded ZK.
Also tested on Windows 10:
mvn -version
Apache Maven 3.3.9
+1 (non-binding)
Verified signatures, checksums, LICENSE and NOTICE
Build was successful on OS X (JDK 1.8.0_111-b14)
Tested basic flows in standalone and three node cluster (using GetTwitter,
JoltTransformJson, PutMongo)
Tested upgrade use case of moving from 1.0.0 to 1.1.0 (See Andre's email
+1 from me. Built without issue for me on Windows 10 and Ubuntu 14.04
(x86_64). My sample flows ran without issue. Did not spot any issues with
licenses. Signature and hashes verified.
On Nov 27, 2016 8:32 PM, "Koji Kawamura" wrote:
> +1 (non-binding)
>
> - Verified
+1 (non-binding)
- Verified checksums
- Successfully ran mvn clean install, on OS X and Windows Server 2016,
with Oracle JDK 1.8.0_111-b14, mvn 3.3.9
- Run example data flows on OS X and Windows
Building and testing on Windows went well without issue. I hope the
InvokeScriptedProcessor test
Aldrin,
I saw the same exception on that test on OS X but it resolved when I
upgraded my version of java (upgraded to 1.8.0_111-b14)
On Sun, Nov 27, 2016 at 5:13 PM, Aldrin Piri wrote:
> +1, binding
>
> Hashes and signatures looked good.
> Contrib passed
> Builds and
Odd, I am still verifying the release, but I was able to build without
issue on Windows 10. Using Oracle java 1.8.0_91 and Apache maven 3.3.3
On Nov 27, 2016 5:14 PM, "Aldrin Piri" wrote:
> +1, binding
>
> Hashes and signatures looked good.
> Contrib passed
> Builds and
+1, binding
Hashes and signatures looked good.
Contrib passed
Builds and functionality were okay on OS X and Linux.
I could not get a successful build on Windows 10 without removing some
tests. I ran into the item covered in
https://issues.apache.org/jira/browse/NIFI-2692 from the 1.0 release
+1 (binding).
Verified the primary finding from RC1 was addressed.
Verified the finding with RAT completion on windows from RC1 was addressed.
Verified the test timeout issue was improved.
Built using a full clean build with contrib check successfully on OSX,
Linux, and Windows 10.
Built using a
21 matches
Mail list logo