[00:10] Project parallel-test build #49: STILL FAILING in 1 hr 10 min: https://lpci.wedontsleep.org/job/parallel-test/49/ [00:20] Project windmill-db-devel build #404: STILL FAILING in 1 hr 6 min: https://lpci.wedontsleep.org/job/windmill-db-devel/404/ [00:40] Yippie, build fixed! [00:41] Project db-devel build #646: FIXED in 5 hr 42 min: https://lpci.wedontsleep.org/job/db-devel/646/ [04:46] Project windmill-db-devel build #405: STILL FAILING in 1 hr 4 min: https://lpci.wedontsleep.org/job/windmill-db-devel/405/ [05:10] Project parallel-test build #50: STILL FAILING in 1 hr 14 min: https://lpci.wedontsleep.org/job/parallel-test/50/ [09:05] allenap: oh hai [11:51] Project db-devel build #648: FAILURE in 5 hr 25 min: https://lpci.wedontsleep.org/job/db-devel/648/ [14:23] lifeless: Something I discovered with that query is that it's looking at a batch of 300 DistroSeriesDifferences. If the batch is reduced to the default (for +localpackagediffs) of 75, the query appears to run 10x faster. So, one solution is to split the query into smaller batches - say 50 - and union them back together. [17:28] Project windmill-devel build #251: STILL FAILING in 1 hr 5 min: https://lpci.wedontsleep.org/job/windmill-devel/251/ [21:41] Project devel build #819: FAILURE in 5 hr 30 min: https://lpci.wedontsleep.org/job/devel/819/ [22:03] Project parallel-test build #51: STILL FAILING in 1 hr 20 min: https://lpci.wedontsleep.org/job/parallel-test/51/ [22:12] allenap: thats interesting; looks like the plan goes screwy ? - one cause for that can be large IN clauses [22:13] allenap: another approach there is to make a temp table, put the ids you want into that table, and then join against it. [22:17] Project windmill-db-devel build #406: STILL FAILING in 1 hr 5 min: https://lpci.wedontsleep.org/job/windmill-db-devel/406/