Updating bucket search schema

Rusty Klophaus rusty at basho.com
Tue Apr 12 10:07:35 EDT 2011


Hi Gordon,

You should be able to run "search-cmd clear-schema-cache" as step 3, instead
of restarting the cluster.

This clears the schema on all connected nodes. (I stress "connected nodes"
here because if a network partition results in an unreachable--but still
running--node, then that node will continue to use the old cached version of
the schema.)

Best,
Rusty

On Wed, Apr 6, 2011 at 3:39 PM, Gordon Tillman <gtillman at mezeo.com> wrote:

> Howdy Gang,
>
> During our various development iterations I have had to alter the schema
> that riaksearch uses for our application's bucket.  I have noticed that when
> I do that -- when I assign a new schema to a bucket -- even if I completely
> purge all data and then repopulate the bucket -- the new schema is not used.
>  For a new schema to be used I have to perform the following steps:
>
> 1. purge bucket
> 2. install  new schema
> 3. stop/start Riak cluster
> 4. repopulate
>
> So question - is there some kind of run-time attribute I can set or
> whatever that will force Riaksearch to use the new schema when indexing data
> that is stored -- without having to restart the cluster?
>
> Many thanks,
>
> --gordon
> _______________________________________________
> riak-users mailing list
> riak-users at lists.basho.com
> http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.basho.com/pipermail/riak-users_lists.basho.com/attachments/20110412/4d4048d0/attachment.html>


More information about the riak-users mailing list