Good: We are back to green builds on GitHub
https://github.com/apache/commons-net/actions and Jenkins.

Gary

On Sat, Jan 1, 2022 at 7:46 AM Gary Gregory <[email protected]> wrote:

> Wow, so there is some odd set up in the test that does not account for
> being run near the start of a year!
>
> Gary
>
> On Sat, Jan 1, 2022, 07:32 Gilles Sadowski <[email protected]> wrote:
>
>> Le sam. 1 janv. 2022 à 03:24, Gilles Sadowski <[email protected]> a
>> écrit :
>> >
>> > Hello.
>> >
>> > Le sam. 1 janv. 2022 à 03:04, Gary Gregory <[email protected]> a
>> écrit :
>> > >
>> > > Hi All & Sebb,
>> > >
>> > > Git bisect tells me the commit below broke the build and it's been
>> broken
>> > > ever since. Please run a local build before you commit.
>> > >
>> > > 991b775f2052e150c4c16d1c1cd2073021d40c2e is the first bad commit
>> > > commit 991b775f2052e150c4c16d1c1cd2073021d40c2e
>> > > Author: Sebb <[email protected]>
>> > > Date:   Wed Dec 15 00:18:48 2021 +0000
>> > >
>> > >     Prevent serialization
>> > >
>> > >     It is not useful and is unlikely to work properly.
>> > >
>> > >  src/changes/changes.xml                              |  4 ++++
>> > >  .../apache/commons/net/ProtocolCommandSupport.java   | 18
>> > > +++++++++++++++++-
>> > >  .../java/org/apache/commons/net/ftp/FTPFile.java     | 20
>> > > +++++++++++++++++++-
>> > >  .../java/org/apache/commons/net/ntp/TimeStamp.java   | 16
>> ++++++++++++++++
>> > >  .../org/apache/commons/net/util/ListenerList.java    | 15
>> +++++++++++++++
>> > >  5 files changed, 71 insertions(+), 2 deletions(-)
>> > >
>> > > Please revert or fix.
>> > >
>> > > [ERROR] Failures:
>> > > [ERROR]   FTPTimestampParserImplTest.testParseTimestampWithSlop:474
>> > > test.slop.roll.back.year expected:<1> but was:<0>
>> >
>> > Here I don't get that error, but another one:
>> > ---CUT---
>> > [INFO] Results:
>> > [INFO]
>> > [ERROR] Failures:
>> > [ERROR]
>>  FTPTimestampParserImplTest.testParseTimestampAcrossTimeZones:437
>> > rollback.even.with.time.zones expected:<1> but was:<2>
>> > ---CUT---
>> >
>> > This one is also reported by Jenkins:
>> >     https://builds.apache.org/job/Commons/job/commons-net/205/console
>> >
>> > It seems that several Jenkins builds were successful *after*
>> > the "Prevent serialization" commit.
>>
>> This failure has now vanished both from the local build and from
>> Jenkins.[1]
>> Hence, I suspect a fragile unit test that fails at the turn of the year...
>>
>> Gilles
>>
>> [1] https://ci-builds.apache.org/job/Commons/job/commons-net/209
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [email protected]
>> For additional commands, e-mail: [email protected]
>>
>>

Reply via email to