+1
- checked the PR of the announcement blog post.
- compiled the source code (without test)
- ran table/sql test cases
Thanks, vino.
Till Rohrmann 于2018年10月26日周五 下午11:33写道：
> +1 (binding)
>
> - Built Flink 1.5.5 with Scala 2.11.11 and Hadoop 2.8.5 and ran all tests
> - Ran all end-to-end tes
+1 (binding)
- Built Flink 1.5.5 with Scala 2.11.11 and Hadoop 2.8.5 and ran all tests
- Ran all end-to-end tests
- Executed the Jepsen test suite with the Hadoop 2.8 binaries
Cheers,
Till
On Fri, Oct 26, 2018 at 8:18 AM Tzu-Li (Gordon) Tai
wrote:
> +1
>
> - Built locally, with Scala 2.11, Had
+1
- Built locally, with Scala 2.11, Hadoop free. Build passes without errors
(failed on HBaseConnectorITCase but passed after restarting the build from the
Hbase connector).
- Verified signature and hashes
- Ran e2e tests locally overnight in a loop, no attempts were failing
- No missing artifa
+1
- ran a local cluster
- submitted multiple jobs through WebUI
- verified availability of metrics
- verified that latency marks are still enabled by default but
configurable via configuration
On 18.10.2018 13:17, Chesnay Schepler wrote:
Hi everyone,
Please review and vote on the release can
Thanks Chesnay for the investigation. Your explanations work in my case
as well. Therefore I give +1.
On 19/10/2018 13:19, Chesnay Schepler wrote:
> See the JIRA for details.
>
> On 19.10.2018 13:18, Chesnay Schepler wrote:
>> This is likely not a new issue and existed since FLINK-9234 was
>> merg
Thanks for the investigation Chesnay!
Withdrawing my -1.
Fabian
Am Fr., 19. Okt. 2018 um 15:15 Uhr schrieb Chesnay Schepler <
ches...@apache.org>:
> To summarize:
>
> The Hbase issue exists since 1.5.0 and stems from flink-tabls's
> jcl-over-slf4j dependency. When running both the flink-table a
To summarize:
The Hbase issue exists since 1.5.0 and stems from flink-tabls's
jcl-over-slf4j dependency. When running both the flink-table and
flink-hbase tests in one run the jc-over-slf4j bridge is loaded, causing
the Logger casts in the HBaseConnectorITCase to fail since it assumes
that Lo
Yes, exactly
Am Fr., 19. Okt. 2018 um 13:49 Uhr schrieb Chesnay Schepler <
ches...@apache.org>:
> When you went past hbase, which was the other module you failed? I would
> guess you also failed in the ES5 module here like for 1.6.2?
>
> On 18.10.2018 23:55, Fabian Hueske wrote:
> > Hi,
> >
> > I
When you went past hbase, which was the other module you failed? I would
guess you also failed in the ES5 module here like for 1.6.2?
On 18.10.2018 23:55, Fabian Hueske wrote:
Hi,
I checked the following things:
* no dependencies added or changed since Flink 1.5.4
* compiling the source distr
See the JIRA for details.
On 19.10.2018 13:18, Chesnay Schepler wrote:
This is likely not a new issue and existed since FLINK-9234 was
merged. (1.4.3/1.5.0)
On 19.10.2018 11:16, Chesnay Schepler wrote:
For the test failure: https://issues.apache.org/jira/browse/FLINK-10609
On 19.10.2018 11:0
This is likely not a new issue and existed since FLINK-9234 was merged.
(1.4.3/1.5.0)
On 19.10.2018 11:16, Chesnay Schepler wrote:
For the test failure: https://issues.apache.org/jira/browse/FLINK-10609
On 19.10.2018 11:00, Dawid Wysakowicz wrote:
Hi,
-1
I've checked:
+ compiling sources wo
For the test failure: https://issues.apache.org/jira/browse/FLINK-10609
On 19.10.2018 11:00, Dawid Wysakowicz wrote:
Hi,
-1
I've checked:
+ compiling sources works
+ verified signatures - OK
- the release notes are missing two issues: FLINK-10259, FLINK-10247,
and are including an issue that
FLINK-10247 was reverted for 1.5.5 and 1.6.2.
https://github.com/apache/flink/commit/1feb1b4c1da8cb5006a1fc2a9eb8fba67d1bc696
https://github.com/apache/flink/commit/61330e47afe8960c804b8cd2edf2ba589b594d8b
I've corrected the fixVersion for FLINK-10259 and FLINK-4052.
On 19.10.2018 11:00, Dawid W
Hi,
-1
I've checked:
+ compiling sources works
+ verified signatures - OK
- the release notes are missing two issues: FLINK-10259, FLINK-10247,
and are including an issue that was actually not merged for this branch:
FLINK-4052
- I get the same test failures as Fabian did:
Java HotSpot(TM) 64-B
Hi,
I checked the following things:
* no dependencies added or changed since Flink 1.5.4
* compiling the source distribution without tests succeeds
* compiling the source distribution with tests fails (see exception
appended below). When I restart the compilation, it goes past flink-hbase
but fai
15 matches
Mail list logo
