cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alex Huang <Alex.Hu...@citrix.com>
Subject RE: Master broken
Date Wed, 03 Apr 2013 23:45:27 GMT
Prasanna,

This is actually a very interesting case.  In our check-in test, is there any way for us to
disable db access when running the unit tests?  I can see that this would be one of the things
that the developer can easily forget but hits on the build server and breaks the Jenkins build
often.

Thanks.

--Alex

> -----Original Message-----
> From: Kelven Yang [mailto:kelven.yang@citrix.com]
> Sent: Wednesday, April 3, 2013 11:23 AM
> To: Hugo Trippaers; Alena Prokharchyk; dev@cloudstack.apache.org;
> cloudstack-dev@incubator.apache.org
> Subject: Re: Master broken
> 
> 
> 
> On 4/3/13 12:42 AM, "Hugo Trippaers" <HTrippaers@schubergphilis.com>
> wrote:
> 
> >Kelven,
> >
> >That commit (5782abf8f80fce929e3d6e20068bc165f2360426 ) was actually
> >the commit that broke the tests. In that commit you add the call to
> >'ComponentContext.initComponentsLifeCycle();' which in turn initializes
> >other modules and some of them make database calls. Those database call
> >will fail on our Jenkins systems as they have no database running.
> >
> >Can you try the following:
> > * shutdown any local mysql databases
> > * git checkout 5782abf8f80fce929e3d6e20068bc165f2360426
> > * mvn clean test
> >
> >I expect that you will get the same error, if not something strange is
> >happening. You have to make sure that you are not running a database on
> >the localhost though. Otherwise the test will succeed
> 
> 
> Now it makes sense and that's why I couldn't see this problem happen in my
> local environment. I'll give it a way without local mysql DB
> 
> -Kelven
> 
> 
> >
> >I'll add a negative testcase that will fail if a local database is
> >present to prevent this, I ran into the same problem a while back.
> >
> >Cheers,
> >
> >Hugo
> >
> >
> >
> >
> >> -----Original Message-----
> >> From: Kelven Yang [mailto:kelven.yang@citrix.com]
> >> Sent: Tuesday, April 02, 2013 7:41 PM
> >> To: Alena Prokharchyk; dev@cloudstack.apache.org; cloudstack-
> >> dev@incubator.apache.org; Hugo Trippaers
> >> Subject: Re: Master broken
> >>
> >> I've applied a commit 5782abf8f80fce929e3d6e20068bc165f2360426 to
> >>address these test configuration fix already. This was done at last
> >>Friday.
> >>
> >> Kelven
> >>
> >>
> >>
> >> On 4/2/13 9:39 AM, "Alena Prokharchyk" <Alena.Prokharchyk@citrix.com>
> >> wrote:
> >>
> >> >The test itself is written by me. It doesn't touch the DB - I mock
> >> >"persist" method of the corresponding Dao. The test was added about
> >> >2 weeks ago to master branch, and it was running fine the last time
> >> >I've checked - last Friday.
> >> >
> >> >As Pranav stated, the test got broken with commit
> >> >5782abf8f80fce929e3d6e20068bc165f2360426, and related to
> UserContext
> >> >initializing.
> >> >
> >> >-Alena.
> >> >
> >> >On 4/2/13 1:32 AM, "Pranav Saxena" <pranav.saxena@citrix.com> wrote:
> >> >
> >> >>Yeah I saw that . Thanks Hugo . I hope master is stable now though
> >> >>I haven't got a change to build it again.
> >> >>
> >> >>Regards,
> >> >>Pranav
> >> >>
> >> >>-----Original Message-----
> >> >>From: Hugo Trippaers [mailto:HTrippaers@schubergphilis.com]
> >> >>Sent: Tuesday, April 02, 2013 1:37 PM
> >> >>To: dev@cloudstack.apache.org; cloudstack-
> dev@incubator.apache.org
> >> >>Subject: RE: Master broken
> >> >>
> >> >>Thanks Pranav,
> >> >>
> >> >>I couldn't find an easy way to fix the test so I indeed disabled it.
> >> >>See commit df6b97c194caa8b34fa14bf5164eb2fe1f26b2b1.
> >> >>
> >> >>Raised a ticket for Kelven (CLOUDSTACK-1884) to have a look at it.
> >> >>
> >> >>Cheers,
> >> >>
> >> >>Hugo
> >> >>
> >> >>> -----Original Message-----
> >> >>> From: Pranav Saxena [mailto:pranav.saxena@citrix.com]
> >> >>> Sent: dinsdag 2 april 2013 10:04
> >> >>> To: dev@cloudstack.apache.org;
> >> >>> cloudstack-dev@incubator.apache.org
> >> >>> Subject: RE: Master broken
> >> >>>
> >> >>> This commit deals with the tests for create network offering -
> >>Commit:
> >> >>> 5782abf8f80fce929e3d6e20068bc165f2360426 and I am presuming
> >> because
> >> >>> of this only the test is failing.
> >> >>>
> >> >>> You might need to revert the commit if required.
> >> >>>
> >> >>> Thanks,
> >> >>> Pranav
> >> >>>
> >> >>> -----Original Message-----
> >> >>> From: Hugo Trippaers [mailto:HTrippaers@schubergphilis.com]
> >> >>> Sent: Tuesday, April 02, 2013 12:32 PM
> >> >>> To: cloudstack-dev@incubator.apache.org
> >> >>> Subject: Master broken
> >> >>>
> >> >>> Heya all,
> >> >>>
> >> >>> Master branch appears to be broken. Is anybody working on this?
> >> >>>
> >> >>> The failed tests are:
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateSharedNtwkOffWithNoVlan
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateSharedNtwkOffWithoutSpecifyIpRanges
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateIsolatedNtwkOffWithSpecifyIpRangesAndSourceNat
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateSharedNtwkOffWithVlan
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateSharedNtwkOffWithSpecifyIpRanges
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateIsolatedNtwkOffWithNoVlan
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateIsolatedNtwkOffWithVlan
> >> >>> >>>
> >> >>>
> >> >>>org.apache.cloudstack.networkoffering.CreateNetworkOfferingTest.c
> >> >>> >>>re
> >> >>> ateIsolatedNtwkOffWithSpecifyIpRangesAndNoSourceNat
> >> >>>
> >> >>>
> >> >>> It appears to have something to do with the
> >> >>> CreateNetworkOfferingTest trying to use the database (which is
> >> >>> never there during test runs)
> >> >>>
> >> >>> I'm looking at it, but if somebody else is doing that as well we
> >> >>> might pool resources.
> >> >>>
> >> >>> Cheers,
> >> >>>
> >> >>> Hugo
> >> >>
> >> >>
> >> >
> >> >
> >


Mime
View raw message