drw
commented
1 year ago I'm having second thoughts about this. The reason we did this in the first place was because the low cut-off caused users to get 32,000 records and think they had everything. But I'm wondering if large API calls might have led to the overloaded database problems we've experienced. Maybe we should think instead about teaching users about navigating paginated results. We could do a better job of documenting how to look for pagination-limited search results.
On the new CKAN instance, this is like 32,000. On the old, one we had it set to 500,000 (if I recall correctly). The variable to set in the site's configuration file is
ckan.datastore.search.rows_max.