> On March 9, 2015, 10:33 p.m., Raghav Gautam wrote: > > falcon-regression/merlin/src/test/java/org/apache/falcon/regression/InstanceParamTest.java, > > line 112 > > <https://reviews.apache.org/r/31133/diff/1/?file=866515#file866515line112> > > > > I am not sure if it works like that. My understanding is that, out of > > the three clusters that you have specified - one will be used to schedule > > the oozie job, and that cluster will be used to run all the instances. Is > > the test passing ?
As per my understanding, the instances of each server are scheduled separately on that server, and hence the prism will return all instances. Yes, this test is passing. However, in the latest falcon code, a change was made in the behaviour of start time. For instance -params, the start time is now being treated as nominal time. So, instead of 15 in prism and 5 in server, we'll get 3 instances from prism and 1 from server, for a given start time. So, will change the patch as per this behaviour. - Karishma ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/31133/#review75800 ----------------------------------------------------------- On Feb. 17, 2015, 8:30 p.m., Karishma Gulati wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/31133/ > ----------------------------------------------------------- > > (Updated Feb. 17, 2015, 8:30 p.m.) > > > Review request for Falcon. > > > Repository: falcon-git > > > Description > ------- > > InstanceParamsTest currently just hits the api and logs the response. Adding > validations to the test to validate the response > > > Diffs > ----- > > > falcon-regression/merlin/src/test/java/org/apache/falcon/regression/InstanceParamTest.java > c8b76e9 > > Diff: https://reviews.apache.org/r/31133/diff/ > > > Testing > ------- > > Made changes to InstanceParamsTest > > > Thanks, > > Karishma Gulati > >
