Two servers, same files, same catalog properties, diferent size an

[feriber]

Hi.

we have a cluster with two nodes. They have the same software installed.
The sites replicate well from node 1 to node 2.

We have configured index server the same way in both servers. The catalogs
find the same number of files 47431 (most of all PDF's 32793).

The size of the catalogs are very different: node 1 365 Mb., node 2 504 Mb.

All other propertiers (docs to index, deferred for indexing, word list, etc)
are the same.

But the most curious thing is that if you make a query to node 1 (small
catalog) you receive 10 results, while making the same query to node 2 (big
catalog), the results obtained are only 4.

Any ideas ??
I'd apreciate any suggestion.

[Hilary Cotter]

do a merge and see if this fixes the size issue. Is it possible that if you
fail over to the second node and let it process all the docs on the shared
resources you get the same number of hits?
--
Hilary Cotter
Director of Text Mining and Database Strategy
RelevantNOISE.Com - Dedicated to mining blogs for business intelligence.

This posting is my own and doesn't necessarily represent RelevantNoise's
positions, strategies or opinions.

Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html

Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com

Post by feriber
Hi.
we have a cluster with two nodes. They have the same software installed.
The sites replicate well from node 1 to node 2.
We have configured index server the same way in both servers. The catalogs
find the same number of files 47431 (most of all PDF's 32793).
The size of the catalogs are very different: node 1 365 Mb., node 2 504 Mb.
All other propertiers (docs to index, deferred for indexing, word list, etc)
are the same.
But the most curious thing is that if you make a query to node 1 (small
catalog) you receive 10 results, while making the same query to node 2 (big
catalog), the results obtained are only 4.
Any ideas ??
I'd apreciate any suggestion.

[feriber]

First of all, thank you Hilary for your interest, and sorry about my english.

I've tried many things.
1st. Deleted both catalogs and rebuild them with same issues.
2nd. Once created, i merged both, but sizes did not match and querys to the
catalogs obtain differente results.

This issue happens only cataloging this particular site. I've got five more
catalogs defined in both servers (with 1308, 1954, 1930, 19823 and 183726
documents each) and the size are identical, and query results are similar.

Is there anything at IIS that could produce the difference ?.

Thank you.

Post by Hilary Cotter
do a merge and see if this fixes the size issue. Is it possible that if you
fail over to the second node and let it process all the docs on the shared
resources you get the same number of hits?
--
Hilary Cotter
Director of Text Mining and Database Strategy
RelevantNOISE.Com - Dedicated to mining blogs for business intelligence.
This posting is my own and doesn't necessarily represent RelevantNoise's
positions, strategies or opinions.
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com

Post by feriber
Hi.
we have a cluster with two nodes. They have the same software installed.
The sites replicate well from node 1 to node 2.
We have configured index server the same way in both servers. The catalogs
find the same number of files 47431 (most of all PDF's 32793).
The size of the catalogs are very different: node 1 365 Mb., node 2 504 Mb.
All other propertiers (docs to index, deferred for indexing, word list, etc)
are the same.
But the most curious thing is that if you make a query to node 1 (small
catalog) you receive 10 results, while making the same query to node 2 (big
catalog), the results obtained are only 4.
Any ideas ??
I'd apreciate any suggestion.

[feriber]

Windows 2000 Server

I've updated Adobe PDF ifilter to version 6, and now my catalogs are same
size.

I'm still having another trouble. I have PDF files with same name but stored
in different paths.

When i try to see an abstract of one of them, index server replies me the
one corresponding to the oldest file.

If i merge the catalog, the results are the same.

If i rebuild the catalog the resuls are the expected ones.

Any ideas ?