Understood. Thanks! Regards Kamal C
On Tue, Aug 16, 2016 at 3:47 PM, Taras Ledkov <[email protected]> wrote: > Hi, > > Affinity job doesn't block rebalance. We reserve partition to prevent > remove data from the node while job is running. > So, topology is changed, partitions are moved to another nodes but data is > available locally during affinity job. Magic! > Please modify your test like below and be sure that data is available for > local query. > > @IgniteInstanceResource > private Ignite ig; > ... > public void run() { > ... > while (true) { > assert ignite.cache(CACHE_NAME).localPeek('a', null).equals('a'); > > On 16.08.2016 11:32, Kamal C wrote: > > Vladislav, > > Partitions are moved to another node while executing the job. > > Val, > > I've tried with the latest nighty build. Still, it has the same behavior. > To reproduce the issue, I've used `while(true)` inside the IgniteRunnable > task. You can reproduce it with the below gist[1]. > > [1]: https://gist.github.com/Kamal15/0a4066de152b8ebc856fc264f7b4037d > > Regards, > Kamal C > > On Sat, Aug 13, 2016 at 12:15 AM, vkulichenko < > [email protected]> wrote: > >> Note that this was changed recently [1] and the change was not released in >> 1.7. Kamal, can you try the nightly build [2] and check if it works as you >> expect there? >> >> [1] https://issues.apache.org/jira/browse/IGNITE-2310 >> [2] >> https://builds.apache.org/view/H-L/view/Ignite/job/Ignite- >> nightly/lastSuccessfulBuild/ >> >> >> >> -- >> View this message in context: http://apache-ignite-users.705 >> 18.x6.nabble.com/Ignite-Compute-Rebalance-Scenario-tp7004p7021.html >> Sent from the Apache Ignite Users mailing list archive at Nabble.com. >> > > > -- > Taras Ledkov > Mail-To: [email protected] > >
