Even simpler and more correct is to increase a number of backups for 
AtomicConfiguration that is used for your sequence. This can be done by setting 
AtomicConfiguration.setBackups to value higher than 0. This will make your 
partitioned AtomicSequence fault tolerant unless all primaries and backups 
leave topology at the same time which is a rare situation.

—
Denis

> On Sep 16, 2016, at 4:53 AM, Vladislav Pyatkov <vldpyat...@gmail.com> wrote:
> 
> Hi,
> 
> If you want so the cache will be replicated do this in 
> AtomicConfiguration.setCacheMode()[1].
> But it is not that by default, default is PARTITIONED cache, so you can loose 
> data in instability topology without backups.
> 
> [1]: https://apacheignite.readme.io/docs/atomic-types#atomic-configuration 
> <https://apacheignite.readme.io/docs/atomic-types#atomic-configuration>
> 
> On Fri, Sep 16, 2016 at 2:40 PM, hitansu <hitansu...@gmail.com 
> <mailto:hitansu...@gmail.com>> wrote:
> Is it like replication ?But if the AtomicSequnce is cluster wide, then it
> should work even one of the node is down.Isn't it?Why is it computing the
> AtomicSequnce only in one node ?
> 
> 
> 
> --
> View this message in context: 
> http://apache-ignite-users.70518.x6.nabble.com/AtomicSequence-not-working-when-shutting-down-one-server-node-from-a-cluster-tp7770p7793.html
>  
> <http://apache-ignite-users.70518.x6.nabble.com/AtomicSequence-not-working-when-shutting-down-one-server-node-from-a-cluster-tp7770p7793.html>
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.
> 
> 
> 
> -- 
> Vladislav Pyatkov

Reply via email to