[ 
https://issues.apache.org/jira/browse/IGNITE-10135?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16675697#comment-16675697
 ] 

Prachi Garg commented on IGNITE-10135:
--------------------------------------

[~syssoftsol],

1. The edits you suggested in the documentation ( in "Crash-safe Affinity" 
callout on 
[https://apacheignite.readme.io/docs/affinity-collocation|https://apacheignite.readme.io/docs/affinity-collocation_]
 page)

"`ClusterNodeAttributeAffinityBackupFilter ` can be used to create an 
`AffinityBackupFilter` which separates primary and backup based on the nodes' 
values for some environment variable, e.g., AVAILABILITY_ZONE.",

is this applicable for Ignite version 2.7 and later, or is it applicable for 
current version too? 

2. Does the environment variable "AVAILABILITY_ZONE" need to be set to 
configure ClusterNodeAttributeAffinityBackupFilter?

3. You have mentioned - "some environment variable, e.g., AVAILABILITY_ZONE". 
Are there other environment variable too that need to be set to use this 
functionality?

 

> Documentation link to ClusterNodeAttributeAffinityBackupFilter
> --------------------------------------------------------------
>
>                 Key: IGNITE-10135
>                 URL: https://issues.apache.org/jira/browse/IGNITE-10135
>             Project: Ignite
>          Issue Type: Improvement
>          Components: documentation
>            Reporter: David Harvey
>            Priority: Major
>             Fix For: 2.7
>
>
> IGNITE-9365 adds ClusterNodeAttributeAffinityBackupFilter to allow 
> "Crash-safe Affinity"  
> ([https://apacheignite.readme.io/docs/affinity-collocation|https://apacheignite.readme.io/docs/affinity-collocation_])
>  to be configured from spring.  
> The class should have an adequate description of how to set this up, but the 
> above section should link to or otherwise flag that such a procedure exist.
> [https://github.com/apache/ignite/blob/master/modules/core/src/main/java/org/apache/ignite/cache/affinity/rendezvous/ClusterNodeAttributeAffinityBackupFilter.java]
>  
> Note: the implementation is generic, allowing any node attribute (or 
> environment variable) to be configured so that primary and backup's are 
> always forced to not be on the same node.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to