-
Bug
-
Resolution: Fixed
-
Critical
-
3.1
-
MOODLE_31_STABLE
-
MOODLE_31_STABLE
-
MDL-53271-master -
Solr seems to have some issues with certain unicode characters. I discover that it chokes on:
項目 (U+9805 U+76EE), "item" in Japanese.
It is actually the first character it has a problem with, and many following characters will cause a failure.
That throws a SolrServerException with the message Invalid UTF-8 start byte 0xae
We have also had other server errors, like Exception writing document id mod_page-activity-65414 to the index; possible analysis error
I don't think there is much we can do about that, but, right now in search_solr\engine::add_document we catch SolrClientException errors, and just debug them, but when the server chokes on a doc, it throws a SolrServerException which we don't catch, and kills all the indexing.
- has a non-specific relationship to
-
MDL-31989 Global Search 2
- Closed