sleep before updating and deleting test documents - otherwise their index entries may not be fully written and this causes errors (update and delete first attempt to remove index references, but if they are not in the DB yet then they are not removed but then they do get added and the FK constraints make the delete fail). Since we know indexing occurs in a separate thread with a configured delay, we just use this same delay in our testing.
Use a final static string to replace the hard code.Search document title rather than id in testReplicateEML_AtoB method.
use CN session when testing getLogRecords() and getOperationStatistics() becuase they are now protecting "sensitive" information
test for running concurrent Metacat queries to mimic Kepler data search.http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5518
include testSynchronizationFailed() and call as the CN subject so that it is authorized.
use MN (self) as the Session.subject so that the MN.delete() call is successful.
comment out testDelete because it requires acting as the MN comment out testSynchronizationFailed because it requires acting as the CN
uncomment the MN tests (I bet this was an oversight during local testing)
no not record EML access rules that use the "denyFirst" permOrder.https://redmine.dataone.org/issues/2614
change ordering of getLogRecords() parameter -- pidFilter is in the middle now
View revisions
Also available in: Atom