Index field data is not updated in case of Full-import with clean=false and commit=true

[Dharmendra Jaiswal <dh...@gmail.com>]

I am using SOLR-3.3.0 version with multi-core setup.
We have already perform Full-import to index data from database.
Let assume that we have updated some value in table for set of column
and some how delta-import has missed these update in index due to xyz
reason.

In that case we have run Full-import with clean=false and commit=true.
Indexing was successful but we are still getting old data in search
response.
Updated column data is not reflected in search response.

Please let me know is something i have missed in above steps.
or do i need to run full-import with clean=true.

Any pointer will be helpful in this case.







--
View this message in context: http://lucene.472066.n3.nabble.com/Index-field-data-is-not-updated-in-case-of-Full-import-with-clean-false-and-commit-true-tp4158146.html
Sent from the Solr - User mailing list archive at Nabble.com.

Re: Index field data is not updated in case of Full-import with clean=false and commit=true

[Shawn Heisey <so...@elyograg.org>]

On 9/11/2014 2:30 AM, Dharmendra Jaiswal wrote:
> I am using SOLR-3.3.0 version with multi-core setup.
> We have already perform Full-import to index data from database.
> Let assume that we have updated some value in table for set of column
> and some how delta-import has missed these update in index due to xyz
> reason.
> 
> In that case we have run Full-import with clean=false and commit=true.
> Indexing was successful but we are still getting old data in search
> response.
> Updated column data is not reflected in search response.
> 
> Please let me know is something i have missed in above steps.
> or do i need to run full-import with clean=true.
> 
> Any pointer will be helpful in this case.

Do you have a uniqueKey field in your schema that is actually unique to
each document?  If you don't, then you cannot use "clean=false" with
your full-import.  You'll need to go with clean=true, to wipe the index
before importing.  The uniqueKey allows new documents with the same
value in the uniqueKey field to overwrite the existing document
automatically -- Solr deletes the old document before inserting the new one.

Usually it's better to have clean=true anytime you do a complete reindex
with the dataimport handler, even when you DO have a uniqueKey.

For various reasons, upgrading to the latest release of Solr would be
advisable.  Solr 3.3.0 is over three years old now, there have been MANY
bugfixes and new features in the meantime.

Thanks,
Shawn