ES 1.1.1
when i use the synonyms filter with a synonyms_path file i have to create
whole index again to take new synonyms in effect.
It is not enough to update the index data, in need to remove, create and
update whole index.
Is that expected behaviour? If so what is ur strategy to get this done on a
productive system?
Anything happening according this or am i monitoring the wrong path?
Am Freitag, 16. Mai 2014 13:42:19 UTC+2 schrieb Tom:
Hi,
ES 1.1.1
when i use the synonyms filter with a synonyms_path file i have to create
whole index again to take new synonyms in effect.
It is not enough to update the index data, in need to remove, create and
update whole index.
Is that expected behaviour? If so what is ur strategy to get this done on
a productive system?
The synonym file will be read whenever a new index is created or if an
index is re-opened. Closing and opening an index is not the cleanest
solution, so I created a plugin to help solve the problem: https://github.com/brusic/refresh-token-filters
However, this version of the plugin does not seem to be working 100%.
Unfortunately, I do not have time to test. I have a different version of
the plugin working at my company for over a year without any issues. It has
more bells and whistles (connects to a database for the term lists, timed
refreshes), but I wanted my public version to be simpler. The most
important change is that I moved from tracking the filter Java objects
directly to using the IndicesService. However, I found that some analyzers
can be cached and therefore there internal filters do not change. Like I
said, I don't have time to test, but I am assuming that is the issue.
Anything happening according this or am i monitoring the wrong path?
Am Freitag, 16. Mai 2014 13:42:19 UTC+2 schrieb Tom:
Hi,
ES 1.1.1
when i use the synonyms filter with a synonyms_path file i have to create
whole index again to take new synonyms in effect.
It is not enough to update the index data, in need to remove, create and
update whole index.
Is that expected behaviour? If so what is ur strategy to get this done on
a productive system?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.