Hi all,
Earlier today I updated my development server, which uses SQLite, and ran the update schema script (now at v118). With hindsight there could have been a warning that I missed - because afterwards despite updating to the galaxy-central tip, trying to use any of the tools failed with an SQL error:
OperationalError: (OperationalError) no such column: history_dataset_association.extended_metadata_id u'SELECT history_dataset_association.id AS history_dataset_association_id, ...
$ ./manage_db.sh db_version 118
$ ./manage_db.sh version 118
Attempting a downgrade/upgrade suggested I had a stale migration_tmp table blocking this,
Since this is only a development instance, I removed the SQLite database (and the old data files) to allow a clean regeneration:
$ ./run.sh ... galaxy.model.migrate.check INFO 2014-02-18 13:25:52,248 Migrating 117 -> 118... galaxy.model.migrate.check INFO 2014-02-18 13:25:54,176 galaxy.model.migrate.check INFO 2014-02-18 13:25:54,177 Add link from history_dataset_association to the extended_metadata table galaxy.model.migrate.check INFO 2014-02-18 13:25:54,177 galaxy.model.migrate.check INFO 2014-02-18 13:25:54,177 migrate.versioning.repository DEBUG 2014-02-18 13:25:54,206 Loading repository lib/tool_shed/galaxy_install/migrate... ...
This fixed the history_dataset_association.extended_metadata_id error - so is the most likely explanation a failed schema update? Might a stale migration_tmp table have been to blame?
Thanks,
Peter