Platform 1.11.5 or above in implementation

@raff in this commit https://github.com/openmrs/openmrs-core/commit/93e5827912517da29478022e4dccb51e88b8bcbb#diff-c0c661e2e2bb32101232a168bcf42521 you switched the patient dao to use lucene but is seems like it was released in 2.1.0 can this be backported to 2.0.5 since at ampath we are still seem to be having this issue with 2.0.5

@achachiez, TRUNK-425 is a new feature with significant behavior changes, so it’s really not appropriate to backport this to a maintenance release.

Is it possible for AMPATH to upgrade to 2.1.x?

Hi @darius is it safe to do that because I think I read somewhere it wa sort of a developer release. And in case we can upgrade is there some sort of of special upgrade path because I tried it and had an issue when running the migrations ??? liquibase-update-to-latest.xml : Table 'amrs.concept_attribute' doesn't exist??? Unable to update the database. See server error logs for the full stacktrace. ??? Caused By: Precondition Error??? ???There was an error while updating the database to the latest. file: liquibase-update-to-latest.xml. Error: Migration failed for change set liquibase-update-to-latest.xml::201609171146-1::vshankar:??? ??? Reason: ??? ???:??? Does this mean we are missing the migration for the concept_attribute table?

When we use 1.9.2 Search war very fast. But I upgraded server and core version. Currently we are using 2.0.6 . Its too slow. Take 30 seconds for patient search. We have 100000+ patients.

Could you please try setting the search.caseSensitiveDatabaseStringComparison global property to false?

Yes, Its already false.

Can you restart tomcat?

Not sure what the migration issues are that were mentioned in this thread, but I would definitely recommend looking into upgrading from 2.1.x. We at PIH upgraded from 1.11 to 2.1 and have been running 2.1 in production now without issue for the past few months… and with the new Lucene search, patient search is very fast.

Take care, Mark