Ijon added a comment.
I think postponing //any// solution to after we have a scaling strategy comes down too strongly against user needs. A number of trusted users -- including yours truly -- occasionally need to run queries that would go 20%-50% over the current timeout, and have no reasonable way of doing it. I understand T179879 <https://phabricator.wikimedia.org/T179879> is a significant feature to develop, but some interim measures could be taken, //while// the scaling strategy is devised. Measures such as: 1. increase the timeout to 120 seconds. 2. provide an additional SPARQL endpoint with older data (say, updated only once every 48 hours) and a higher timeout, as these longer queries typically don't need up-to-the-minute replication. These seem to me to be relatively simple to implement, while a longer-term solution is provided, and would unblock some meaningful work in the meantime. TASK DETAIL https://phabricator.wikimedia.org/T187424 EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Ijon Cc: Ijon, Gehel, Iamamz3, Daniel_Mietchen, Addshore, Smalyshev, abian, Lucas_Werkmeister_WMDE, Aklapper, Bugreporter, darthmon_wmde, ET4Eva, Nandana, Lahi, Gq86, Darkminds3113, GoranSMilovanovic, QZanden, EBjune, merbst, LawExplorer, Avner, _jensen, rosalieper, Scott_WUaS, Jonas, FloNight, Xmlizer, jkroll, Wikidata-bugs, Jdouglas, aude, Tobias1984, Dinoguy1000, Manybubbles, Mbch331
_______________________________________________ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
