[
https://issues.apache.org/jira/browse/IGNITE-28692?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sergey Chugunov updated IGNITE-28692:
-------------------------------------
Description:
MultiDC support enables stretching single Ignite cluster between 2+ data
centers which in turn allows increase data availability by having a copy of
data in each DC.
But spreading copies between DCs requires additional configuration on cache
level, and a user could forget or miss this part of configuration.
To prevent this we need to implement new metrics:
* A boolean cache metric indicating that configuration may be wrong.
* A boolean cache metric indicating that with current configuration actual
partition distribution between data centers does not have the property "one
copy of data in each DC".
was:
MultiDC support enables stretching single Ignite cluster between 2+ data
centers which in turn allows increase data availability by having a copy of
data in each DC.
But spreading copies between DCs requires additional configuration on cache
level, and a user could forget or miss this part of configuration.
To prevent this we need to implement several things:
* A warning on cache start to reflect this lack of configuration in logs.
* A boolean cache metric indicating that configuration may be wrong.
* A boolean cache metric indicating that with current configuration actual
partition distribution between data centers does not have the property "one
copy of data in each DC".
> Validate cache backup filters for caches started in MultiDC environments
> ------------------------------------------------------------------------
>
> Key: IGNITE-28692
> URL: https://issues.apache.org/jira/browse/IGNITE-28692
> Project: Ignite
> Issue Type: Task
> Reporter: Sergey Chugunov
> Assignee: Sergey Chugunov
> Priority: Major
> Labels: IEP-140, ise
> Fix For: 2.19
>
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> MultiDC support enables stretching single Ignite cluster between 2+ data
> centers which in turn allows increase data availability by having a copy of
> data in each DC.
> But spreading copies between DCs requires additional configuration on cache
> level, and a user could forget or miss this part of configuration.
> To prevent this we need to implement new metrics:
> * A boolean cache metric indicating that configuration may be wrong.
> * A boolean cache metric indicating that with current configuration actual
> partition distribution between data centers does not have the property "one
> copy of data in each DC".
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
