[14:59:56] PROBLEM - High lag on wdqs1002 is CRITICAL: CRITICAL: 31.03% of data above the critical threshold [1800.0] [15:02:56] PROBLEM - High lag on wdqs1002 is CRITICAL: CRITICAL: 37.93% of data above the critical threshold [1800.0] [15:42:19] Does anybody know how to do a SPARQL query to find most commonly used properties? [15:50:11] zboiq: it's not a sparql query, but https://www.wikidata.org/wiki/Wikidata:Database_reports/List_of_properties/Top100 might be useful [15:53:27] Thanks nikki . Specifically I need the top 100 most frequent properties amongst Human entities [16:07:39] zboiq: You can peek in the suggestions [16:08:38] zboiq https://www.wikidata.org/wiki/Wikidata:SPARQL_query_service/queries#More_examples the second example is close to what you're looking for, except ?tgt_class should be wd:Q5 [16:10:32] That's great. Sadly it query times out with Q5 [16:11:17] not unexpected ^^' [16:11:36] we don't have all of them, but there's a lot of Q5 in existence :p [16:11:55] I guess I'll download the json and use mongodb aggregation [16:16:41] hold on, trying a less expensive approach [16:17:07] nope, still timing out ^^' [16:17:19] we should kill all humans. way too many of them. [18:24:00] hey people [18:24:25] i'm curious as to what could possibly cause the (really) high lag on wdqs, reported earlier by icinga-wm [18:24:39] i swear i didn't touch anything ! :p [18:57:30] ACKNOWLEDGEMENT - High lag on wdqs1002 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [1800.0] Gehel restarting updater to catch up on updates [19:01:32] Alphos: I'm absolutely not sure of what the issue is this time (update lag on wdqs) [19:02:20] we've had a few issues lately, related to high edit rate and identified an issue with the updater. But today seems to be different. And only a single server is lagging, not all of them [19:17:42] gehel : yep, noticed it with an increasing lag today [19:18:49] Alphos: https://phabricator.wikimedia.org/T166524 (and it might be related to high CPU temperature, cause or effect, not sure :) [19:19:21] or correlation without one being the cause of the other ? :p [19:19:28] correlation is not causation ! [19:20:24] (could be a zombie process hogging everything because reasons, thus causing high cpu temp *and* lag, but neither would be the cause of the other ; just offering a thought here ;-) ) [19:21:21] yep, I'm just observing symptoms at this point... [19:22:26] still, from the point of view of a guy who really hates doing sysadmin stuff even though he sometimes has to, it's got me curious :D [19:25:48] the only process consuming significant CPU is blazegraph itself... [19:25:55] * gehel is curious too... [19:26:23] eureka ? [19:26:29] it runs on java, doesn't it ? :p [19:27:04] oh come on, what's a teensy weensy troll among good company ? :p [19:34:07] mind if i subscribe ? [19:34:11] please do! [19:34:35] * gehel notes that he is probably the only Java fanboy around... [19:35:25] to each their own, i write bots in php :D [19:35:34] bashing starts… NOW ! [19:36:06] * gehel is trying to find a good reply to that... but I'm crying too much... poor bot... [19:36:41] naughty naughty naughty :D [19:37:06] i even wrote a tool in bash, sooo :p [20:04:02] gehel forgive my lack of knowledge : LVS ? [20:05:14] Alphos: Linux Virtual Server, our load balancer, see https://wikitech.wikimedia.org/wiki/LVS [20:05:20] right [20:07:05] It's getting late here... I'll be back tomorrow... [20:07:20] see ya :) [21:40:22] http://tools.wmflabs.org/wikidata-todo/quick_statements.php is ok? [22:31:00] RECOVERY - High lag on wdqs1002 is OK: OK: Less than 30.00% above the threshold [600.0] [22:31:47] \o/