On Thu, Jan 25, 2018 at 2:05 PM, Mikael Ståldal wrote:
> I the delay is there for the "testClose" test cast, but it slows down the
> other test methods as well, which is unintended.
>
> I have fixed it now by breaking out the "testClose" test case to its own
> class.
Great! And thank you.
Gary
I the delay is there for the "testClose" test cast, but it slows down
the other test methods as well, which is unintended.
I have fixed it now by breaking out the "testClose" test case to its own
class.
On 2018-01-23 14:39, Gary Gregory wrote:
It would be nice to heard from Mike his thought
It would be nice to heard from Mike his thoughts on my change since he
added the delay in the first place. I am worried about any 'bite you later
' factor :-p
On Jan 22, 2018 11:31 PM, "Remko Popma" wrote:
> Nice!!
>
> (Shameless plug) Every java main() method deserves http://picocli.info
>
> >
Nice!!
(Shameless plug) Every java main() method deserves http://picocli.info
> On Jan 23, 2018, at 14:25, Gary Gregory wrote:
>
> On Mon, Jan 22, 2018 at 9:45 PM, Gary Gregory
> wrote:
>
>> Hm, it already uses the mock stuff!
>>
>> I reduced test delays in the MockProducer introduced in com
On Mon, Jan 22, 2018 at 9:45 PM, Gary Gregory
wrote:
> Hm, it already uses the mock stuff!
>
> I reduced test delays in the MockProducer introduced in commit
> 96436fb958ce1f1a3d4f0c951f556f0709c91b15 (by Mike) from 3 seconds to 50
> milliseconds. This reduces running this test case from 43 to 3
Hm, it already uses the mock stuff!
I reduced test delays in the MockProducer introduced in commit
96436fb958ce1f1a3d4f0c951f556f0709c91b15 (by Mike) from 3 seconds to 50
milliseconds. This reduces running this test case from 43 to 3 seconds.
Let's watch this test in Jenkins to make sure it still
On Mon, Jan 22, 2018 at 1:11 PM, Matt Sicker wrote:
> The Kafka test could probably be rewritten to use the
> MockProducer/MockConsumer classes instead of presumably embedding Kafka.
>
For my money, using mocks here does not guarantee anything :-(
Gary
>
> On 22 January 2018 at 14:08, Gary Gr
Maybe, but is that as reliable as what is currently done.
FWIW, Kafka is another one that should be moved. The issue with a lot of these
tests is that the smallest granularity that can be counted on for a file last
modified time is 1 second. That means the unit test often has to wait for
longer
Lame formatting, I put these results here as well:
https://pastebin.com/grKtxPTq
Gary
On Mon, Jan 22, 2018 at 1:08 PM, Gary Gregory
wrote:
> Hi All:
>
> Here are some number based on https://builds.apache.org/
> user/ggregory/my-views/view/Logging/job/Log4j 2.x/3315. There are some
> obvious lo
The Kafka test could probably be rewritten to use the
MockProducer/MockConsumer classes instead of presumably embedding Kafka.
On 22 January 2018 at 14:08, Gary Gregory wrote:
> Hi All:
>
> Here are some number based on
> https://builds.apache.org/user/ggregory/my-views/view/Logging/job/Log4j
>
Hi All:
Here are some number based on
https://builds.apache.org/user/ggregory/my-views/view/Logging/job/Log4j
2.x/3315. There are some obvious low-hanging fruits.
43.078 org.apache.logging.log4j.core.appender.mom.kafka.KafkaAppenderTest
33.799
org.apache.logging.log4j.core.appender.routing.Rout
11 matches
Mail list logo
