You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@activemq.apache.org by Jean-Baptiste Onofre <jb...@nanthrax.net> on 2020/12/01 05:22:24 UTC
[CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Hi everyone,
Due to an issue with the fix for AMQ-8045, I cancel this vote to prepare a new one.
Sorry about that. And thanks Art to find this.
A new vote will start today.
Regards
JB
> Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a écrit :
>
> Hi everyone,
>
> I submit Apache ActiveMQ 5.15.14 to your vote.
> This release includes CVE related updates and bug fixes.
>
> Please take a look on the Release Notes for details:
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294 <https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294>
>
> The Maven staging repository is:
> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/ <https://repository.apache.org/content/repositories/orgapacheactivemq-1220/>
>
> The dist staging repository is:
> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
>
> Git tag:
> activemq-5.15.14
>
> Please vote to approve this release:
>
> [ ] +1 Approve the release
> [ ] -1 Don't approve the release (please provide specific comments)
>
> This vote will be open for at least 72 hours.
>
> Thanks !
> Regards
> JB
Re: [CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Posted by Arthur Naseef <ar...@amlinv.com>.
Makes sense. Glad to hear the tests are run even if we have a few that
fail consistently.
When I did a few official AMQ builds, I learned to use "mvn -fn" to run
through the entire build, then go back and re-run the tests that failed.
At that time, running the tests more than once usually got them to pass,
and I accepted that result.
The more I looked at the ClassCastException in the
InMemeoryJmsSchedulerTest, the more I think that test just doesn't apply.
If that's the case, we could use a simple means to disable the specific
test method when inherited - either by overriding the test method to do
nothing, or by adding a simple "instanceof" check in the base class.
Art
On Wed, Dec 2, 2020 at 2:09 AM Jean-Baptiste Onofre <jb...@nanthrax.net> wrote:
> FYI, I fixed the issues (very quick fix) the InMemory test.
>
> Regards
> JB
>
> > Le 2 déc. 2020 à 05:54, Jean-Baptiste Onofre <jb...@nanthrax.net> a écrit :
> >
> > Hi Art,
> >
> > Thanks for the update !
> >
> > Yeah, those tests are some of I would like to fix (they fail for quite
> long now, since at least 5.15.9).
> >
> > I think some tests should be removed because they don’t make sense
> anymore.
> >
> > I will take a look (I’m also improving the Karaf tests which are flaky).
> >
> > By the way, all is executed on Jenkins with the pipeline I created.
> However:
> > - a full ActiveMQ test is very long (~6h)
> > - the build often fails on Jenkins due to flaky tests
> >
> > So, I started to improve the situation, but it takes sometime and I
> don’t think we should hold releases for non accurate/flaky tests.
> >
> > Thanks again for your feedback, I’m moving forward on release (5.15.14
> take #2) and improving the build/tests.
> >
> > Regards
> > JB
> >
> >> Le 1 déc. 2020 à 22:46, Arthur Naseef <ar...@amlinv.com> a écrit :
> >>
> >> So the ClassCastException happens in 5.15.13 and 5.15.10 when I run that
> >> test specifically.
> >> I suspect we have a case where the test class inheritance is causing
> >> confusion. If so, then the fix here is to bypass the base-class test
> >> method.
> >>
> >> Can someone look at the test method
> >> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded
> >> in JmsSchedulerTest.java and answer the question, "does this test method
> >> make sense for the in-memory scheduler?" I suspect the in-memory
> scheduler
> >> just doesn't have a journal, so the test doesn't apply.
> >>
> >> Still no update on (2).
> >>
> >> Art
> >>
> >>
> >> On Tue, Dec 1, 2020 at 1:48 PM Arthur Naseef <ar...@amlinv.com> wrote:
> >>
> >>> Looks like I'm finding 2 problems with the InMemeoryJmsSchedulerTest:
> >>>
> >>>
> >>> 1. The ClassCastException in one test, and
> >>> 2. "Didn't receive the message" problem in the other.
> >>>
> >>>
> >>> Digging into the ClassCastException, all the code appears to be
> unchanged
> >>> for at least 6 months. Could this have been broken all this time?
> >>>
> >>> Haven't dug into the second problem yet.
> >>>
> >>> Art
> >>>
> >>>
> >>> On Tue, Dec 1, 2020 at 1:03 PM Arthur Naseef <ar...@amlinv.com> wrote:
> >>>
> >>>> So it looks like my build only had 2 test failures. The first we
> >>>> discussed already. The second is as follows:
> >>>>
> >>>> [INFO] Running
> >>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
> >>>> [ERROR] Tests run: 8, Failures: 1, Errors: 1, Skipped: 0, Time
> elapsed:
> >>>> 41.993 s <<< FAILURE! - in
> >>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
> >>>> [ERROR]
> >>>>
> testScheduleFullRecoveryRestart(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
> >>>> Time elapsed: 5.514 s <<< FAILURE!
> >>>> java.lang.AssertionError: Didn't receive the message
> >>>>
> >>>> [ERROR]
> >>>>
> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
> >>>> Time elapsed: 0.005 s <<< ERROR!
> >>>> java.lang.ClassCastException:
> >>>> org.apache.activemq.broker.scheduler.memory.InMemoryJobSchedulerStore
> >>>> cannot be cast to
> >>>> org.apache.activemq.store.kahadb.scheduler.JobSchedulerStoreImpl'
> >>>>
> >>>>
> >>>> I'm looking into it - please let me know if this is also a known
> issue.
> >>>>
> >>>> Art
> >>>>
> >>>>
> >>>> On Mon, Nov 30, 2020 at 10:22 PM Jean-Baptiste Onofre <
> jb@nanthrax.net>
> >>>> wrote:
> >>>>
> >>>>> Hi everyone,
> >>>>>
> >>>>> Due to an issue with the fix for AMQ-8045, I cancel this vote to
> prepare
> >>>>> a new one.
> >>>>>
> >>>>> Sorry about that. And thanks Art to find this.
> >>>>>
> >>>>> A new vote will start today.
> >>>>>
> >>>>> Regards
> >>>>> JB
> >>>>>
> >>>>>> Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a
> >>>>> écrit :
> >>>>>>
> >>>>>> Hi everyone,
> >>>>>>
> >>>>>> I submit Apache ActiveMQ 5.15.14 to your vote.
> >>>>>> This release includes CVE related updates and bug fixes.
> >>>>>>
> >>>>>> Please take a look on the Release Notes for details:
> >>>>>>
> >>>>>
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
> >>>>> <
> >>>>>
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
> >>>>>>
> >>>>>>
> >>>>>> The Maven staging repository is:
> >>>>>>
> >>>>>
> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
> >>>>> <
> >>>>>
> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
> >>>>>>
> >>>>>>
> >>>>>> The dist staging repository is:
> >>>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <
> >>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
> >>>>>>
> >>>>>> Git tag:
> >>>>>> activemq-5.15.14
> >>>>>>
> >>>>>> Please vote to approve this release:
> >>>>>>
> >>>>>> [ ] +1 Approve the release
> >>>>>> [ ] -1 Don't approve the release (please provide specific comments)
> >>>>>>
> >>>>>> This vote will be open for at least 72 hours.
> >>>>>>
> >>>>>> Thanks !
> >>>>>> Regards
> >>>>>> JB
> >>>>>
> >>>>>
> >
>
>
Re: [CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Posted by Arthur Naseef <ar...@amlinv.com>.
Oh nice. I missed this earlier. Thanks!
Art
On Wed, Dec 2, 2020 at 2:09 AM Jean-Baptiste Onofre <jb...@nanthrax.net> wrote:
> FYI, I fixed the issues (very quick fix) the InMemory test.
>
> Regards
> JB
>
> > Le 2 déc. 2020 à 05:54, Jean-Baptiste Onofre <jb...@nanthrax.net> a écrit :
> >
> > Hi Art,
> >
> > Thanks for the update !
> >
> > Yeah, those tests are some of I would like to fix (they fail for quite
> long now, since at least 5.15.9).
> >
> > I think some tests should be removed because they don’t make sense
> anymore.
> >
> > I will take a look (I’m also improving the Karaf tests which are flaky).
> >
> > By the way, all is executed on Jenkins with the pipeline I created.
> However:
> > - a full ActiveMQ test is very long (~6h)
> > - the build often fails on Jenkins due to flaky tests
> >
> > So, I started to improve the situation, but it takes sometime and I
> don’t think we should hold releases for non accurate/flaky tests.
> >
> > Thanks again for your feedback, I’m moving forward on release (5.15.14
> take #2) and improving the build/tests.
> >
> > Regards
> > JB
> >
> >> Le 1 déc. 2020 à 22:46, Arthur Naseef <ar...@amlinv.com> a écrit :
> >>
> >> So the ClassCastException happens in 5.15.13 and 5.15.10 when I run that
> >> test specifically.
> >> I suspect we have a case where the test class inheritance is causing
> >> confusion. If so, then the fix here is to bypass the base-class test
> >> method.
> >>
> >> Can someone look at the test method
> >> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded
> >> in JmsSchedulerTest.java and answer the question, "does this test method
> >> make sense for the in-memory scheduler?" I suspect the in-memory
> scheduler
> >> just doesn't have a journal, so the test doesn't apply.
> >>
> >> Still no update on (2).
> >>
> >> Art
> >>
> >>
> >> On Tue, Dec 1, 2020 at 1:48 PM Arthur Naseef <ar...@amlinv.com> wrote:
> >>
> >>> Looks like I'm finding 2 problems with the InMemeoryJmsSchedulerTest:
> >>>
> >>>
> >>> 1. The ClassCastException in one test, and
> >>> 2. "Didn't receive the message" problem in the other.
> >>>
> >>>
> >>> Digging into the ClassCastException, all the code appears to be
> unchanged
> >>> for at least 6 months. Could this have been broken all this time?
> >>>
> >>> Haven't dug into the second problem yet.
> >>>
> >>> Art
> >>>
> >>>
> >>> On Tue, Dec 1, 2020 at 1:03 PM Arthur Naseef <ar...@amlinv.com> wrote:
> >>>
> >>>> So it looks like my build only had 2 test failures. The first we
> >>>> discussed already. The second is as follows:
> >>>>
> >>>> [INFO] Running
> >>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
> >>>> [ERROR] Tests run: 8, Failures: 1, Errors: 1, Skipped: 0, Time
> elapsed:
> >>>> 41.993 s <<< FAILURE! - in
> >>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
> >>>> [ERROR]
> >>>>
> testScheduleFullRecoveryRestart(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
> >>>> Time elapsed: 5.514 s <<< FAILURE!
> >>>> java.lang.AssertionError: Didn't receive the message
> >>>>
> >>>> [ERROR]
> >>>>
> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
> >>>> Time elapsed: 0.005 s <<< ERROR!
> >>>> java.lang.ClassCastException:
> >>>> org.apache.activemq.broker.scheduler.memory.InMemoryJobSchedulerStore
> >>>> cannot be cast to
> >>>> org.apache.activemq.store.kahadb.scheduler.JobSchedulerStoreImpl'
> >>>>
> >>>>
> >>>> I'm looking into it - please let me know if this is also a known
> issue.
> >>>>
> >>>> Art
> >>>>
> >>>>
> >>>> On Mon, Nov 30, 2020 at 10:22 PM Jean-Baptiste Onofre <
> jb@nanthrax.net>
> >>>> wrote:
> >>>>
> >>>>> Hi everyone,
> >>>>>
> >>>>> Due to an issue with the fix for AMQ-8045, I cancel this vote to
> prepare
> >>>>> a new one.
> >>>>>
> >>>>> Sorry about that. And thanks Art to find this.
> >>>>>
> >>>>> A new vote will start today.
> >>>>>
> >>>>> Regards
> >>>>> JB
> >>>>>
> >>>>>> Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a
> >>>>> écrit :
> >>>>>>
> >>>>>> Hi everyone,
> >>>>>>
> >>>>>> I submit Apache ActiveMQ 5.15.14 to your vote.
> >>>>>> This release includes CVE related updates and bug fixes.
> >>>>>>
> >>>>>> Please take a look on the Release Notes for details:
> >>>>>>
> >>>>>
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
> >>>>> <
> >>>>>
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
> >>>>>>
> >>>>>>
> >>>>>> The Maven staging repository is:
> >>>>>>
> >>>>>
> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
> >>>>> <
> >>>>>
> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
> >>>>>>
> >>>>>>
> >>>>>> The dist staging repository is:
> >>>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <
> >>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
> >>>>>>
> >>>>>> Git tag:
> >>>>>> activemq-5.15.14
> >>>>>>
> >>>>>> Please vote to approve this release:
> >>>>>>
> >>>>>> [ ] +1 Approve the release
> >>>>>> [ ] -1 Don't approve the release (please provide specific comments)
> >>>>>>
> >>>>>> This vote will be open for at least 72 hours.
> >>>>>>
> >>>>>> Thanks !
> >>>>>> Regards
> >>>>>> JB
> >>>>>
> >>>>>
> >
>
>
Re: [CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Posted by Jean-Baptiste Onofre <jb...@nanthrax.net>.
FYI, I fixed the issues (very quick fix) the InMemory test.
Regards
JB
> Le 2 déc. 2020 à 05:54, Jean-Baptiste Onofre <jb...@nanthrax.net> a écrit :
>
> Hi Art,
>
> Thanks for the update !
>
> Yeah, those tests are some of I would like to fix (they fail for quite long now, since at least 5.15.9).
>
> I think some tests should be removed because they don’t make sense anymore.
>
> I will take a look (I’m also improving the Karaf tests which are flaky).
>
> By the way, all is executed on Jenkins with the pipeline I created. However:
> - a full ActiveMQ test is very long (~6h)
> - the build often fails on Jenkins due to flaky tests
>
> So, I started to improve the situation, but it takes sometime and I don’t think we should hold releases for non accurate/flaky tests.
>
> Thanks again for your feedback, I’m moving forward on release (5.15.14 take #2) and improving the build/tests.
>
> Regards
> JB
>
>> Le 1 déc. 2020 à 22:46, Arthur Naseef <ar...@amlinv.com> a écrit :
>>
>> So the ClassCastException happens in 5.15.13 and 5.15.10 when I run that
>> test specifically.
>> I suspect we have a case where the test class inheritance is causing
>> confusion. If so, then the fix here is to bypass the base-class test
>> method.
>>
>> Can someone look at the test method
>> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded
>> in JmsSchedulerTest.java and answer the question, "does this test method
>> make sense for the in-memory scheduler?" I suspect the in-memory scheduler
>> just doesn't have a journal, so the test doesn't apply.
>>
>> Still no update on (2).
>>
>> Art
>>
>>
>> On Tue, Dec 1, 2020 at 1:48 PM Arthur Naseef <ar...@amlinv.com> wrote:
>>
>>> Looks like I'm finding 2 problems with the InMemeoryJmsSchedulerTest:
>>>
>>>
>>> 1. The ClassCastException in one test, and
>>> 2. "Didn't receive the message" problem in the other.
>>>
>>>
>>> Digging into the ClassCastException, all the code appears to be unchanged
>>> for at least 6 months. Could this have been broken all this time?
>>>
>>> Haven't dug into the second problem yet.
>>>
>>> Art
>>>
>>>
>>> On Tue, Dec 1, 2020 at 1:03 PM Arthur Naseef <ar...@amlinv.com> wrote:
>>>
>>>> So it looks like my build only had 2 test failures. The first we
>>>> discussed already. The second is as follows:
>>>>
>>>> [INFO] Running
>>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
>>>> [ERROR] Tests run: 8, Failures: 1, Errors: 1, Skipped: 0, Time elapsed:
>>>> 41.993 s <<< FAILURE! - in
>>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
>>>> [ERROR]
>>>> testScheduleFullRecoveryRestart(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
>>>> Time elapsed: 5.514 s <<< FAILURE!
>>>> java.lang.AssertionError: Didn't receive the message
>>>>
>>>> [ERROR]
>>>> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
>>>> Time elapsed: 0.005 s <<< ERROR!
>>>> java.lang.ClassCastException:
>>>> org.apache.activemq.broker.scheduler.memory.InMemoryJobSchedulerStore
>>>> cannot be cast to
>>>> org.apache.activemq.store.kahadb.scheduler.JobSchedulerStoreImpl'
>>>>
>>>>
>>>> I'm looking into it - please let me know if this is also a known issue.
>>>>
>>>> Art
>>>>
>>>>
>>>> On Mon, Nov 30, 2020 at 10:22 PM Jean-Baptiste Onofre <jb...@nanthrax.net>
>>>> wrote:
>>>>
>>>>> Hi everyone,
>>>>>
>>>>> Due to an issue with the fix for AMQ-8045, I cancel this vote to prepare
>>>>> a new one.
>>>>>
>>>>> Sorry about that. And thanks Art to find this.
>>>>>
>>>>> A new vote will start today.
>>>>>
>>>>> Regards
>>>>> JB
>>>>>
>>>>>> Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a
>>>>> écrit :
>>>>>>
>>>>>> Hi everyone,
>>>>>>
>>>>>> I submit Apache ActiveMQ 5.15.14 to your vote.
>>>>>> This release includes CVE related updates and bug fixes.
>>>>>>
>>>>>> Please take a look on the Release Notes for details:
>>>>>>
>>>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>>>>> <
>>>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>>>>>>
>>>>>>
>>>>>> The Maven staging repository is:
>>>>>>
>>>>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>>>>> <
>>>>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>>>>>>
>>>>>>
>>>>>> The dist staging repository is:
>>>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <
>>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
>>>>>>
>>>>>> Git tag:
>>>>>> activemq-5.15.14
>>>>>>
>>>>>> Please vote to approve this release:
>>>>>>
>>>>>> [ ] +1 Approve the release
>>>>>> [ ] -1 Don't approve the release (please provide specific comments)
>>>>>>
>>>>>> This vote will be open for at least 72 hours.
>>>>>>
>>>>>> Thanks !
>>>>>> Regards
>>>>>> JB
>>>>>
>>>>>
>
Re: [CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Posted by Jean-Baptiste Onofre <jb...@nanthrax.net>.
Hi Art,
Thanks for the update !
Yeah, those tests are some of I would like to fix (they fail for quite long now, since at least 5.15.9).
I think some tests should be removed because they don’t make sense anymore.
I will take a look (I’m also improving the Karaf tests which are flaky).
By the way, all is executed on Jenkins with the pipeline I created. However:
- a full ActiveMQ test is very long (~6h)
- the build often fails on Jenkins due to flaky tests
So, I started to improve the situation, but it takes sometime and I don’t think we should hold releases for non accurate/flaky tests.
Thanks again for your feedback, I’m moving forward on release (5.15.14 take #2) and improving the build/tests.
Regards
JB
> Le 1 déc. 2020 à 22:46, Arthur Naseef <ar...@amlinv.com> a écrit :
>
> So the ClassCastException happens in 5.15.13 and 5.15.10 when I run that
> test specifically.
> I suspect we have a case where the test class inheritance is causing
> confusion. If so, then the fix here is to bypass the base-class test
> method.
>
> Can someone look at the test method
> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded
> in JmsSchedulerTest.java and answer the question, "does this test method
> make sense for the in-memory scheduler?" I suspect the in-memory scheduler
> just doesn't have a journal, so the test doesn't apply.
>
> Still no update on (2).
>
> Art
>
>
> On Tue, Dec 1, 2020 at 1:48 PM Arthur Naseef <ar...@amlinv.com> wrote:
>
>> Looks like I'm finding 2 problems with the InMemeoryJmsSchedulerTest:
>>
>>
>> 1. The ClassCastException in one test, and
>> 2. "Didn't receive the message" problem in the other.
>>
>>
>> Digging into the ClassCastException, all the code appears to be unchanged
>> for at least 6 months. Could this have been broken all this time?
>>
>> Haven't dug into the second problem yet.
>>
>> Art
>>
>>
>> On Tue, Dec 1, 2020 at 1:03 PM Arthur Naseef <ar...@amlinv.com> wrote:
>>
>>> So it looks like my build only had 2 test failures. The first we
>>> discussed already. The second is as follows:
>>>
>>> [INFO] Running
>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
>>> [ERROR] Tests run: 8, Failures: 1, Errors: 1, Skipped: 0, Time elapsed:
>>> 41.993 s <<< FAILURE! - in
>>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
>>> [ERROR]
>>> testScheduleFullRecoveryRestart(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
>>> Time elapsed: 5.514 s <<< FAILURE!
>>> java.lang.AssertionError: Didn't receive the message
>>>
>>> [ERROR]
>>> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
>>> Time elapsed: 0.005 s <<< ERROR!
>>> java.lang.ClassCastException:
>>> org.apache.activemq.broker.scheduler.memory.InMemoryJobSchedulerStore
>>> cannot be cast to
>>> org.apache.activemq.store.kahadb.scheduler.JobSchedulerStoreImpl'
>>>
>>>
>>> I'm looking into it - please let me know if this is also a known issue.
>>>
>>> Art
>>>
>>>
>>> On Mon, Nov 30, 2020 at 10:22 PM Jean-Baptiste Onofre <jb...@nanthrax.net>
>>> wrote:
>>>
>>>> Hi everyone,
>>>>
>>>> Due to an issue with the fix for AMQ-8045, I cancel this vote to prepare
>>>> a new one.
>>>>
>>>> Sorry about that. And thanks Art to find this.
>>>>
>>>> A new vote will start today.
>>>>
>>>> Regards
>>>> JB
>>>>
>>>>> Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a
>>>> écrit :
>>>>>
>>>>> Hi everyone,
>>>>>
>>>>> I submit Apache ActiveMQ 5.15.14 to your vote.
>>>>> This release includes CVE related updates and bug fixes.
>>>>>
>>>>> Please take a look on the Release Notes for details:
>>>>>
>>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>>>> <
>>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>>>>>
>>>>>
>>>>> The Maven staging repository is:
>>>>>
>>>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>>>> <
>>>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>>>>>
>>>>>
>>>>> The dist staging repository is:
>>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <
>>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
>>>>>
>>>>> Git tag:
>>>>> activemq-5.15.14
>>>>>
>>>>> Please vote to approve this release:
>>>>>
>>>>> [ ] +1 Approve the release
>>>>> [ ] -1 Don't approve the release (please provide specific comments)
>>>>>
>>>>> This vote will be open for at least 72 hours.
>>>>>
>>>>> Thanks !
>>>>> Regards
>>>>> JB
>>>>
>>>>
Re: [CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Posted by Arthur Naseef <ar...@amlinv.com>.
So the ClassCastException happens in 5.15.13 and 5.15.10 when I run that
test specifically.
I suspect we have a case where the test class inheritance is causing
confusion. If so, then the fix here is to bypass the base-class test
method.
Can someone look at the test method
testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded
in JmsSchedulerTest.java and answer the question, "does this test method
make sense for the in-memory scheduler?" I suspect the in-memory scheduler
just doesn't have a journal, so the test doesn't apply.
Still no update on (2).
Art
On Tue, Dec 1, 2020 at 1:48 PM Arthur Naseef <ar...@amlinv.com> wrote:
> Looks like I'm finding 2 problems with the InMemeoryJmsSchedulerTest:
>
>
> 1. The ClassCastException in one test, and
> 2. "Didn't receive the message" problem in the other.
>
>
> Digging into the ClassCastException, all the code appears to be unchanged
> for at least 6 months. Could this have been broken all this time?
>
> Haven't dug into the second problem yet.
>
> Art
>
>
> On Tue, Dec 1, 2020 at 1:03 PM Arthur Naseef <ar...@amlinv.com> wrote:
>
>> So it looks like my build only had 2 test failures. The first we
>> discussed already. The second is as follows:
>>
>> [INFO] Running
>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
>> [ERROR] Tests run: 8, Failures: 1, Errors: 1, Skipped: 0, Time elapsed:
>> 41.993 s <<< FAILURE! - in
>> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
>> [ERROR]
>> testScheduleFullRecoveryRestart(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
>> Time elapsed: 5.514 s <<< FAILURE!
>> java.lang.AssertionError: Didn't receive the message
>>
>> [ERROR]
>> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
>> Time elapsed: 0.005 s <<< ERROR!
>> java.lang.ClassCastException:
>> org.apache.activemq.broker.scheduler.memory.InMemoryJobSchedulerStore
>> cannot be cast to
>> org.apache.activemq.store.kahadb.scheduler.JobSchedulerStoreImpl'
>>
>>
>> I'm looking into it - please let me know if this is also a known issue.
>>
>> Art
>>
>>
>> On Mon, Nov 30, 2020 at 10:22 PM Jean-Baptiste Onofre <jb...@nanthrax.net>
>> wrote:
>>
>>> Hi everyone,
>>>
>>> Due to an issue with the fix for AMQ-8045, I cancel this vote to prepare
>>> a new one.
>>>
>>> Sorry about that. And thanks Art to find this.
>>>
>>> A new vote will start today.
>>>
>>> Regards
>>> JB
>>>
>>> > Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a
>>> écrit :
>>> >
>>> > Hi everyone,
>>> >
>>> > I submit Apache ActiveMQ 5.15.14 to your vote.
>>> > This release includes CVE related updates and bug fixes.
>>> >
>>> > Please take a look on the Release Notes for details:
>>> >
>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>>> <
>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>>> >
>>> >
>>> > The Maven staging repository is:
>>> >
>>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>>> <
>>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>>> >
>>> >
>>> > The dist staging repository is:
>>> > https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <
>>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
>>> >
>>> > Git tag:
>>> > activemq-5.15.14
>>> >
>>> > Please vote to approve this release:
>>> >
>>> > [ ] +1 Approve the release
>>> > [ ] -1 Don't approve the release (please provide specific comments)
>>> >
>>> > This vote will be open for at least 72 hours.
>>> >
>>> > Thanks !
>>> > Regards
>>> > JB
>>>
>>>
Re: [CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Posted by Arthur Naseef <ar...@amlinv.com>.
Looks like I'm finding 2 problems with the InMemeoryJmsSchedulerTest:
1. The ClassCastException in one test, and
2. "Didn't receive the message" problem in the other.
Digging into the ClassCastException, all the code appears to be unchanged
for at least 6 months. Could this have been broken all this time?
Haven't dug into the second problem yet.
Art
On Tue, Dec 1, 2020 at 1:03 PM Arthur Naseef <ar...@amlinv.com> wrote:
> So it looks like my build only had 2 test failures. The first we
> discussed already. The second is as follows:
>
> [INFO] Running
> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
> [ERROR] Tests run: 8, Failures: 1, Errors: 1, Skipped: 0, Time elapsed:
> 41.993 s <<< FAILURE! - in
> org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
> [ERROR]
> testScheduleFullRecoveryRestart(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
> Time elapsed: 5.514 s <<< FAILURE!
> java.lang.AssertionError: Didn't receive the message
>
> [ERROR]
> testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
> Time elapsed: 0.005 s <<< ERROR!
> java.lang.ClassCastException:
> org.apache.activemq.broker.scheduler.memory.InMemoryJobSchedulerStore
> cannot be cast to
> org.apache.activemq.store.kahadb.scheduler.JobSchedulerStoreImpl'
>
>
> I'm looking into it - please let me know if this is also a known issue.
>
> Art
>
>
> On Mon, Nov 30, 2020 at 10:22 PM Jean-Baptiste Onofre <jb...@nanthrax.net>
> wrote:
>
>> Hi everyone,
>>
>> Due to an issue with the fix for AMQ-8045, I cancel this vote to prepare
>> a new one.
>>
>> Sorry about that. And thanks Art to find this.
>>
>> A new vote will start today.
>>
>> Regards
>> JB
>>
>> > Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a
>> écrit :
>> >
>> > Hi everyone,
>> >
>> > I submit Apache ActiveMQ 5.15.14 to your vote.
>> > This release includes CVE related updates and bug fixes.
>> >
>> > Please take a look on the Release Notes for details:
>> >
>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>> <
>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
>> >
>> >
>> > The Maven staging repository is:
>> >
>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>> <
>> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
>> >
>> >
>> > The dist staging repository is:
>> > https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <
>> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
>> >
>> > Git tag:
>> > activemq-5.15.14
>> >
>> > Please vote to approve this release:
>> >
>> > [ ] +1 Approve the release
>> > [ ] -1 Don't approve the release (please provide specific comments)
>> >
>> > This vote will be open for at least 72 hours.
>> >
>> > Thanks !
>> > Regards
>> > JB
>>
>>
Re: [CANCEL][VOTE] Apache ActiveMQ 5.15.14 release
Posted by Arthur Naseef <ar...@amlinv.com>.
So it looks like my build only had 2 test failures. The first we discussed
already. The second is as follows:
[INFO] Running
org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
[ERROR] Tests run: 8, Failures: 1, Errors: 1, Skipped: 0, Time elapsed:
41.993 s <<< FAILURE! - in
org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest
[ERROR]
testScheduleFullRecoveryRestart(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
Time elapsed: 5.514 s <<< FAILURE!
java.lang.AssertionError: Didn't receive the message
[ERROR]
testUpdatesAppliedToIndexBeforeJournalShouldBeDiscarded(org.apache.activemq.broker.scheduler.memory.InMemeoryJmsSchedulerTest)
Time elapsed: 0.005 s <<< ERROR!
java.lang.ClassCastException:
org.apache.activemq.broker.scheduler.memory.InMemoryJobSchedulerStore
cannot be cast to
org.apache.activemq.store.kahadb.scheduler.JobSchedulerStoreImpl'
I'm looking into it - please let me know if this is also a known issue.
Art
On Mon, Nov 30, 2020 at 10:22 PM Jean-Baptiste Onofre <jb...@nanthrax.net>
wrote:
> Hi everyone,
>
> Due to an issue with the fix for AMQ-8045, I cancel this vote to prepare a
> new one.
>
> Sorry about that. And thanks Art to find this.
>
> A new vote will start today.
>
> Regards
> JB
>
> > Le 24 nov. 2020 à 08:00, Jean-Baptiste Onofre <jb...@nanthrax.net> a écrit
> :
> >
> > Hi everyone,
> >
> > I submit Apache ActiveMQ 5.15.14 to your vote.
> > This release includes CVE related updates and bug fixes.
> >
> > Please take a look on the Release Notes for details:
> >
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
> <
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12348294
> >
> >
> > The Maven staging repository is:
> >
> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
> <
> https://repository.apache.org/content/repositories/orgapacheactivemq-1220/
> >
> >
> > The dist staging repository is:
> > https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/ <
> https://dist.apache.org/repos/dist/dev/activemq/activemq/5.15.14/>
> >
> > Git tag:
> > activemq-5.15.14
> >
> > Please vote to approve this release:
> >
> > [ ] +1 Approve the release
> > [ ] -1 Don't approve the release (please provide specific comments)
> >
> > This vote will be open for at least 72 hours.
> >
> > Thanks !
> > Regards
> > JB
>
>