Thanks a lot Mikhail, moving parent filters to fq solved the problem.
Thanks for the wt=csv technique that also was very helpful.
Thanks.
Faisal
On Mon, Oct 13, 2014 at 11:08 PM, Mikhail Khludnev <
mkhlud...@griddynamics.com> wrote:
> Hello Fasial,
>
> It's convenient to use wt=csv to verify blo
Hello Fasial,
It's convenient to use wt=csv to verify block layout (make sue that rows=
is high enough). Note, every span of children should be followed by parent.
see
http://blog.griddynamics.com/2013/12/grandchildren-and-siblings-with-block.html
Putting BRAND_s: clause into the {! which= } is ab
There's the CheckIndex tool, but I don't know whether it checks this case.
Is there any way you could check this against 4.10.1? I know there
were a couple of edge cases that were fixed in that release, but have
no idea whether block joins were affected.
But other than that, I'm clueless.
Erick
Thanks Erick,
I am using the 4.10.0 solr. To update any field, I first delete all
existing documents matching _root_:id before inserting the updated version.
So, I don't think that's causing the problem.
Is there a way to validate the sanity of the index, are there any tools
which can analyze the
What version of Solr? For Block Join you need to index all the docs
(parent/child) docs at once, and when they're updated they _all_ need
to be updated AFAIK.
Best,
Erick
On Mon, Oct 13, 2014 at 3:51 PM, Faisal Mansoor
wrote:
> Hi,
>
> I am having problems running Parent Block Join Query in Solr
