[
https://issues.apache.org/jira/browse/IGNITE-5229?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16042606#comment-16042606
]
Roman Shtykh commented on IGNITE-5229:
--------------------------------------
[~anovikov] Thank you for the review!
Sorry, I am not sure I understand the problem you found.
Do you think that registering caches other than "redis-ignite-internal-cache-0"
(for instance, "redis-ignite-internal-cache-1",
"redis-ignite-internal-cache-2") explicitly is the problem, and this has to be
prevented? We end up with several caches with different configurations then,
right?
To me, it is ok -- if users really want it, there is no strong reason to stop
them. Does it break anything I cannot see?
Perhaps a Redis user won't expect this behavior, but probably we don't need to
follow Redis rules very strictly and give users more freedom with caches. What
do you think?
Your understanding on how it works is correct.
> Specify caches when using Redis protocol
> ----------------------------------------
>
> Key: IGNITE-5229
> URL: https://issues.apache.org/jira/browse/IGNITE-5229
> Project: Ignite
> Issue Type: Improvement
> Affects Versions: 2.0
> Reporter: Roman Shtykh
> Assignee: Roman Shtykh
> Labels: redis
> Fix For: 2.1
>
>
> Currently there's no way to switch caches -- all requests go to 'default'.
> _Note that this is the switch needed only for a subset of Redis data
> structures (currently only STRINGs) -- for SETs and HASHTABLEs caches will be
> specified as keys (see IGNITE-5241)_
> The solution to be implemented:
> 1. A user specifies the cache configuration (template) with predefined name
> ‘redis-ignite-internal-cache’
> 2. Then issues ‘SELECT n’, and uses ‘redis-ignite-internal-cache-n’.
> Caches are configurable by providing a template.
> http://apache-ignite-developers.2346864.n4.nabble.com/Changing-cache-name-when-using-Redis-protocol-td17727.html
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
