[
https://issues.apache.org/jira/browse/COUCHDB-2594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14395953#comment-14395953
]
Robert Newson commented on COUCHDB-2594:
----------------------------------------
I think the 2.0 blocker list included a button or screen in Fauxton to allow
the user to choose "single node" versus "add more nodes" and only after that
would database creation succeed.
Joan is right that we have not addressed the behavior for nodes added *after*
initial cluster setup is complete.
However, I don't think the post facto log warning has really helped anyone
figure out their mistake (and the only way to fix it is to delete and recreate).
A strong mechanism would be to remove the silent reduction of N when creating a
database, to instead return a 400 or 500 error message (we'll have to bikeshed
a little on whether the user is in error for asking for more replicas than the
cluster can create or the server is in error for not having enough nodes to
satisfy the user).
When choosing "single node" in setup, the default N value should be set to 1,
so that database creations do not return the error.
Should the administrator subsequently add a node, we should increase the
default N for all nodes, up to some threshold (3 would be a good choice).
Alternately, the act of adding the node through Fauxton could ask the
administrator for the new N value (or at least confirm our new suggestion of
min(3, number_of_nodes_in_the_new_cluster))
All of this is a UI nicety over the raw mechanics of fabric and mem3, which
would not be substantially altered. For expert users like Cloudant, all cluster
operations would be automated by some other system which ultimately alters both
the contents of the INI files and the runtime state of the config application.
> Single node mode: remove warning
> --------------------------------
>
> Key: COUCHDB-2594
> URL: https://issues.apache.org/jira/browse/COUCHDB-2594
> Project: CouchDB
> Issue Type: Task
> Security Level: public(Regular issues)
> Components: Database Core
> Reporter: Robert Kowalski
> Priority: Blocker
> Fix For: 2.0.0
>
>
> we have to remove a warning that is sent as response if the node is not
> joined into a multi-node cluster and has no membership
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
