Global searches made thru yacysearch.json return results too fast, so fast that results from the local index aren't parsed in time to get included in the returned results set.
The data contained in my local index, specifically crawled for from chosen sites, constitutes the most pertinent pool of results I want to retrieve when searching. On the other hand, results retrieved via p2p are not as context-specific as those in my own index, thus less relevant to my queries. Most often, the results I receive are sourced solely from p2p, due to unpredictable timeout to query my hard disks.
The server load is not constant over time, so I cannot simply re-run the query after a defined timeout to allow enough time for local results to be collected from my hard disks. Besides, the disks serving Yacy may vary in number and quality (IOps) over time.
What would be needed is an adjustable setting —possibly an additional query parameter— to specifically ask Yacy to wait for local results while performing a global query.