lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Erick Erickson <erickerick...@gmail.com>
Subject Re: Can Atomic Updates help me to re-indexing w/o crawling external content?
Date Mon, 02 Jun 2014 14:37:45 GMT
Hmmm, when changing the schema there might be issues is you're changing the
definition of an already-existing field. I've seen weirdness when the
fundamental definition of a field changes so I'd be cautious.

You'd only be able to add new fields via copyField I'd guess.

In this situation, since you're storing your data anyway, why not store the
source somewhere else (like the file system) until your schema settles down
and re-index from _there_?

If you can't do that, you could read the docs from one collection and index
it to another. Use the recent (4.8) "deep paging" stuff that Hoss put in to
avoid that problem.

Best,
Erick


On Sun, Jun 1, 2014 at 3:11 PM, Gili Nachum <gilinachum@gmail.com> wrote:

> Hello.
>
> I'm just starting out with my Solr deployment and believe there's a good
> chance I'll want to change how my fields are indexing in the near future, I
> wouldn't want to crawl the original content store again just to re-index.
> I was hoping that Atomic Updates (which keeps all fields as stored) might
> allow me to just update the new schema and "ask" the system to re-index all
> document according to the revised schema. Is that possible? in plan? could
> be done in a different way?
>
> Thanks!
> Gili.
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message