Re: Surge of flaky tests

2016-06-16 Thread Apekshit Sharma
Also, I see some tests failing with UnkownScannerException and
ScannerTimeoutExcpetion. They weren't their 10-15 days before. Is someone
aware of code changes around that?
Otherwise i'll have to dig.

On Thu, Jun 16, 2016 at 6:44 AM, Ted Yu  wrote:

> I got the following in branch-1.3 as well:
>
> testCallQueueTooBigException(org.apache.hadoop.hbase.client.TestFastFail)
>  Time elapsed: 2.402 sec  <<< FAILURE!
> java.lang.AssertionError: There should have been 1 hit expected:<1> but
> was:<0>
> at org.junit.Assert.fail(Assert.java:88)
> at org.junit.Assert.failNotEquals(Assert.java:834)
> at org.junit.Assert.assertEquals(Assert.java:645)
> at
>
> org.apache.hadoop.hbase.client.TestFastFail.testCallQueueTooBigException(TestFastFail.java:340)
>
> FYI
>
> On Wed, Jun 15, 2016 at 9:09 PM, Mikhail Antonov 
> wrote:
>
> > I suspected this failure was caused by me reverting the split/merge
> > switch-related locks, but couldn't reproduce it locally. Let me look
> again
> > too.
> >
> > -Mikhail
> >
> > On Wed, Jun 15, 2016 at 9:04 PM, Ted Yu  wrote:
> >
> > > From the list, some tests, such as the following (reproduced locally),
> > > failed 100%.
> > >
> > >
> testCallQueueTooBigException(org.apache.hadoop.hbase.client.TestFastFail)
> > >  Time elapsed: 1.54 sec  <<< FAILURE!
> > > java.lang.AssertionError: There should have been 1 hit expected:<1> but
> > > was:<0>
> > > at org.junit.Assert.fail(Assert.java:88)
> > > at org.junit.Assert.failNotEquals(Assert.java:834)
> > > at org.junit.Assert.assertEquals(Assert.java:645)
> > > at
> > >
> > >
> >
> org.apache.hadoop.hbase.client.TestFastFail.testCallQueueTooBigException(TestFastFail.java:327)
> > >
> > >
> > > Results :
> > >
> > > Failed tests:
> > >   TestFastFail.testCallQueueTooBigException:327 There should have been
> 1
> > > hit expected:<1> but was:<0>
> > >
> > > Let me try to find out which commit caused the failure.
> > >
> > > Cheers
> > >
> > > On Wed, Jun 15, 2016 at 8:50 PM, Apekshit Sharma 
> > > wrote:
> > >
> > > > Hi
> > > > I am seeing a sudden surge in number of flaky tests.
> > > > http://hbase.x10host.com/flaky-tests/
> > > > I'd request everyone to have a look at the list and see if any test
> > > failure
> > > > correlates with recent changes you made.
> > > > To go to failing runs, click on show/hide next to that test.
> > > >
> > > > -- Appy
> > > >
> > >
> >
> >
> >
> > --
> > Thanks,
> > Michael Antonov
> >
>



-- 

Regards

Apekshit Sharma | Software Engineer, Cloudera | Palo Alto, California |
650-963-6311


Re: Surge of flaky tests

2016-06-16 Thread Ted Yu
I got the following in branch-1.3 as well:

testCallQueueTooBigException(org.apache.hadoop.hbase.client.TestFastFail)
 Time elapsed: 2.402 sec  <<< FAILURE!
java.lang.AssertionError: There should have been 1 hit expected:<1> but
was:<0>
at org.junit.Assert.fail(Assert.java:88)
at org.junit.Assert.failNotEquals(Assert.java:834)
at org.junit.Assert.assertEquals(Assert.java:645)
at
org.apache.hadoop.hbase.client.TestFastFail.testCallQueueTooBigException(TestFastFail.java:340)

FYI

On Wed, Jun 15, 2016 at 9:09 PM, Mikhail Antonov 
wrote:

> I suspected this failure was caused by me reverting the split/merge
> switch-related locks, but couldn't reproduce it locally. Let me look again
> too.
>
> -Mikhail
>
> On Wed, Jun 15, 2016 at 9:04 PM, Ted Yu  wrote:
>
> > From the list, some tests, such as the following (reproduced locally),
> > failed 100%.
> >
> > testCallQueueTooBigException(org.apache.hadoop.hbase.client.TestFastFail)
> >  Time elapsed: 1.54 sec  <<< FAILURE!
> > java.lang.AssertionError: There should have been 1 hit expected:<1> but
> > was:<0>
> > at org.junit.Assert.fail(Assert.java:88)
> > at org.junit.Assert.failNotEquals(Assert.java:834)
> > at org.junit.Assert.assertEquals(Assert.java:645)
> > at
> >
> >
> org.apache.hadoop.hbase.client.TestFastFail.testCallQueueTooBigException(TestFastFail.java:327)
> >
> >
> > Results :
> >
> > Failed tests:
> >   TestFastFail.testCallQueueTooBigException:327 There should have been 1
> > hit expected:<1> but was:<0>
> >
> > Let me try to find out which commit caused the failure.
> >
> > Cheers
> >
> > On Wed, Jun 15, 2016 at 8:50 PM, Apekshit Sharma 
> > wrote:
> >
> > > Hi
> > > I am seeing a sudden surge in number of flaky tests.
> > > http://hbase.x10host.com/flaky-tests/
> > > I'd request everyone to have a look at the list and see if any test
> > failure
> > > correlates with recent changes you made.
> > > To go to failing runs, click on show/hide next to that test.
> > >
> > > -- Appy
> > >
> >
>
>
>
> --
> Thanks,
> Michael Antonov
>


Re: Surge of flaky tests

2016-06-15 Thread Mikhail Antonov
I suspected this failure was caused by me reverting the split/merge
switch-related locks, but couldn't reproduce it locally. Let me look again
too.

-Mikhail

On Wed, Jun 15, 2016 at 9:04 PM, Ted Yu  wrote:

> From the list, some tests, such as the following (reproduced locally),
> failed 100%.
>
> testCallQueueTooBigException(org.apache.hadoop.hbase.client.TestFastFail)
>  Time elapsed: 1.54 sec  <<< FAILURE!
> java.lang.AssertionError: There should have been 1 hit expected:<1> but
> was:<0>
> at org.junit.Assert.fail(Assert.java:88)
> at org.junit.Assert.failNotEquals(Assert.java:834)
> at org.junit.Assert.assertEquals(Assert.java:645)
> at
>
> org.apache.hadoop.hbase.client.TestFastFail.testCallQueueTooBigException(TestFastFail.java:327)
>
>
> Results :
>
> Failed tests:
>   TestFastFail.testCallQueueTooBigException:327 There should have been 1
> hit expected:<1> but was:<0>
>
> Let me try to find out which commit caused the failure.
>
> Cheers
>
> On Wed, Jun 15, 2016 at 8:50 PM, Apekshit Sharma 
> wrote:
>
> > Hi
> > I am seeing a sudden surge in number of flaky tests.
> > http://hbase.x10host.com/flaky-tests/
> > I'd request everyone to have a look at the list and see if any test
> failure
> > correlates with recent changes you made.
> > To go to failing runs, click on show/hide next to that test.
> >
> > -- Appy
> >
>



-- 
Thanks,
Michael Antonov


Re: Surge of flaky tests

2016-06-15 Thread Ted Yu
>From the list, some tests, such as the following (reproduced locally),
failed 100%.

testCallQueueTooBigException(org.apache.hadoop.hbase.client.TestFastFail)
 Time elapsed: 1.54 sec  <<< FAILURE!
java.lang.AssertionError: There should have been 1 hit expected:<1> but
was:<0>
at org.junit.Assert.fail(Assert.java:88)
at org.junit.Assert.failNotEquals(Assert.java:834)
at org.junit.Assert.assertEquals(Assert.java:645)
at
org.apache.hadoop.hbase.client.TestFastFail.testCallQueueTooBigException(TestFastFail.java:327)


Results :

Failed tests:
  TestFastFail.testCallQueueTooBigException:327 There should have been 1
hit expected:<1> but was:<0>

Let me try to find out which commit caused the failure.

Cheers

On Wed, Jun 15, 2016 at 8:50 PM, Apekshit Sharma  wrote:

> Hi
> I am seeing a sudden surge in number of flaky tests.
> http://hbase.x10host.com/flaky-tests/
> I'd request everyone to have a look at the list and see if any test failure
> correlates with recent changes you made.
> To go to failing runs, click on show/hide next to that test.
>
> -- Appy
>