Updating data in a production setup

vijayakumar forgetvijay at gmail.com
Wed Apr 18 09:43:12 EDT 2012


Mark,
     Thanks for your help. I could say the number of keys would be in the
range of millions and total number of buckets is 4.

Regards,
Vijayakumar.

On Tue, Apr 17, 2012 at 9:15 PM, Mark Phillips <mark at basho.com> wrote:

> Hi Vijayakumar,
>
>
> On Sat, Apr 14, 2012 at 2:57 AM, vijayakumar <forgetvijay at gmail.com>wrote:
>
>> Hi,
>>   As our application moves from one release to another, we need to alter
>> the existing records in riak [Say changes like adding new field to the json
>> with a default value]. What's the ideal way to handle such schema changes
>> (if indexing is required for such fields). Is it possible to run a
>> mapreduce on existing buckets and update the records? I am not find any
>> help links for the above mentioned migration.
>> Riak Version:1.1
>>
>>
> The short answer is "yes, that's possible". That said, at the moment I'm
> not aware of any existing code/resources that could walk you through it.
> Anyone have anything they can share?
>
> Keep in mind that running something like this over all your data is going
> to put a lot of load on your cluster and might lead to some timeouts and
> interesting debugging. Out of curiosity, how many keys do you need to
> update?
>
> Mark
>
>
>> Thanks and Regards,
>> Vijayakumar.
>>
>>
>>
>> _______________________________________________
>> 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/20120418/78699974/attachment.html>


More information about the riak-users mailing list