[00:18:52] 10DBA, 10Jade, 10Operations, 10TechCom-RFC, and 3 others: Introduce a new namespace for collaborative judgements about wiki entities - https://phabricator.wikimedia.org/T200297 (10Harej) [00:19:22] 10DBA, 10Jade, 10Operations, 10TechCom-RFC, and 3 others: Introduce a new namespace for collaborative judgements about wiki entities - https://phabricator.wikimedia.org/T200297 (10Harej) a:05awight→03None [00:20:01] 10DBA, 10Jade, 10Operations, 10TechCom-RFC, and 3 others: Introduce a new namespace for collaborative judgements about wiki entities - https://phabricator.wikimedia.org/T200297 (10Harej) I have not seen any updates for this RFC in a few months. My understanding is that most of the issues are addressed. Is... [05:00:40] 10DBA, 10MediaWiki-Cache, 10Patch-For-Review, 10Performance-Team (Radar), 10User-Marostegui: Replace parsercache keys to something more meaningful on db-XXXX.php - https://phabricator.wikimedia.org/T210725 (10Marostegui) pc1007 is now back to 56% after the optimization [05:33:32] 10DBA, 10MediaWiki-API: Slow query "ApiQueryLogEvents::execute" after actor rollout - https://phabricator.wikimedia.org/T220999 (10Marostegui) I guess we need to FORCE index? :( [06:27:22] 10DBA, 10MediaWiki-Cache, 10Patch-For-Review, 10Performance-Team (Radar), 10User-Marostegui: Replace parsercache keys to something more meaningful on db-XXXX.php - https://phabricator.wikimedia.org/T210725 (10Marostegui) ` pc[2007-2010].codfw.wmnet,pc[1007-1010].eqiad.wmnet Confirm to continue [y/n]? y =... [08:32:18] 10DBA, 10MediaWiki-Database, 10Core Platform Team (Multi-DC (TEC1)), 10Core Platform Team Backlog (Watching / External), and 4 others: Fix mediawiki heartbeat model, change pt-heartbeat model to not use super-user, avoid SPOF and switch automatically to the real ma... - https://phabricator.wikimedia.org/T172497 [08:35:21] 10DBA, 10Jade, 10Operations, 10TechCom-RFC, and 3 others: Introduce a new namespace for collaborative judgements about wiki entities - https://phabricator.wikimedia.org/T200297 (10jcrespo) Could you summarize the tables added and the resource commitment, just to make sure we all understand it well? [10:43:44] 10DBA, 10cloud-services-team (Kanban): DB planning: include a writeable (?) misc DB cluster in codfw for WMCS - https://phabricator.wikimedia.org/T218570 (10mark) [12:40:27] hi team [12:40:41] there is an item in the roadmap called "misc clusters in codfw" [12:40:51] is that referring to mN databases? [12:40:55] yep [12:41:11] ack [12:41:12] thanks :) [12:41:20] And that also implies proxies (which already have their own task) [15:56:33] 10DBA, 10MediaWiki-API: Slow query "ApiQueryLogEvents::execute" after actor rollout - https://phabricator.wikimedia.org/T220999 (10Anomie) We could force a STRAIGHT_JOIN instead. I'm not sure which would be better. STRAIGHT_JOIN could be validly applied for almost all the code paths,[1] while the index forcin... [16:17:04] 10DBA, 10MediaWiki-API: Slow query "ApiQueryLogEvents::execute" after actor rollout - https://phabricator.wikimedia.org/T220999 (10jcrespo) If you want a practical (but general) reason of `STRAIGHT_JOIN` vs `FORCE`, I would vote for `STRAIGHT_JOIN` (both can create bad query plans, but S_J cannot create a har... [16:18:58] paravoid, in case it is useful, there is a quick reference at https://wikitech.wikimedia.org/wiki/MariaDB#Sections_and_shards [16:19:56] (but yeah, not of yet-unexistent ones :-D) [17:36:40] jynus are you planning using the ssds for the os? [17:36:59] dbprov? should that be the first vd? [17:37:13] no, let me link you the full info [17:37:31] we prepared it for papaul, it is the same for you :-D [17:37:36] one sec [17:38:08] i see in the task [17:38:17] Raid setup: Raid6 HDDs ( first virtual drive), Raid0 SSDs (second virtual drive). Both same options as the dbs- write back with 256K of stripe [17:38:19] Raid setup: Raid6 HDDs ( first virtual drive), Raid0 SSDs (second virtual drive). Both same options as the dbs- write back with 256K of stripe [17:38:22] he he [17:38:24] :-) [17:38:26] hah [17:38:36] one more thing [17:38:41] disable sd card reader [17:38:44] or it won't install [17:38:49] ok [17:38:51] thx [17:39:01] so first drive is sda [17:39:04] like on dbs [17:39:45] okay...and the ssd's are together as raid0 not separate raid 0's [17:40:19] it says together but just want to make sure [17:41:02] ywp [17:41:05] ywp [17:41:07] *yep [17:41:20] the ssds will be scratch data [17:41:23] for performance [17:41:35] so not part of the main installation, we set it up afterwards [17:41:52] ok [17:42:15] and it won't contain os or important data, can be lost [18:06:04] 10DBA, 10Jade, 10Operations, 10TechCom-RFC, and 3 others: Introduce a new namespace for collaborative judgements about wiki entities - https://phabricator.wikimedia.org/T200297 (10colewhite) p:05Triage→03Normal [18:10:36] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: rack/setup/deploy eqiad dedicated backup recovery/provisioning hosts - https://phabricator.wikimedia.org/T219399 (10Cmjohnson) [18:13:59] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: rack/setup/deploy eqiad dedicated backup recovery/provisioning hosts - https://phabricator.wikimedia.org/T219399 (10Cmjohnson) @robh, @Marostegui or @jcrespo These servers are finished with the on-site specific tasks. All yours to finish [18:15:01] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: rack/setup/deploy eqiad dedicated backup recovery/provisioning hosts - https://phabricator.wikimedia.org/T219399 (10jcrespo) a:03jcrespo Thanks! [18:16:56] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: db1078 s3 primary DB master BBU pre-failure - https://phabricator.wikimedia.org/T219115 (10jcrespo) a:05Marostegui→03jcrespo This is fixed, but not closing because I cannot repool the server yet (Deployment schedule conflic). I will repool it tomorr... [18:56:45] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: rack/setup/deploy eqiad dedicated backup recovery/provisioning hosts - https://phabricator.wikimedia.org/T219399 (10Marostegui) Yaaay! [20:06:14] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: rack/setup/install db1139|db1140.eqiad.wmnet (2 dump slaves) - https://phabricator.wikimedia.org/T218985 (10Cmjohnson) [20:07:01] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: rack/setup/install db1139|db1140.eqiad.wmnet (2 dump slaves) - https://phabricator.wikimedia.org/T218985 (10Cmjohnson) a:05Cmjohnson→03RobH @Marostegui @robh these are racked and all on-site work is completed. [20:19:57] 10DBA, 10Operations, 10ops-eqiad, 10Patch-For-Review: rack/setup/install db1139|db1140.eqiad.wmnet (2 dump slaves) - https://phabricator.wikimedia.org/T218985 (10Marostegui) So from DC Ops side only missing the production DNS entries? Thanks Chris! [23:49:55] 10DBA, 10MediaWiki-Database, 10Performance-Team, 10Core Platform Team (Multi-DC (TEC1)), and 2 others: FY18/19 TEC1.6 Q4: Improve or replace GTID + pt-heartbeat logic for cross-DC - https://phabricator.wikimedia.org/T221159 (10mobrovac) p:05Triage→03High [23:50:36] 10DBA, 10MediaWiki-Database, 10Performance-Team, 10Core Platform Team (Multi-DC (TEC1)), and 2 others: FY18/19 TEC1.6 Q4: Improve or replace GTID + pt-heartbeat logic for cross-DC - https://phabricator.wikimedia.org/T221159 (10mobrovac) @aaron @jcrespo @Marostegui the task desc is a draft, please feel free... [23:50:49] 10DBA, 10MediaWiki-Database, 10Performance-Team, 10Core Platform Team (Multi-DC (TEC1)), and 2 others: FY18/19 TEC1.6 Q4: Improve or replace GTID + pt-heartbeat logic for cross-DC - https://phabricator.wikimedia.org/T221159 (10mobrovac) [23:51:22] 10DBA, 10MediaWiki-Database, 10Performance-Team, 10Core Platform Team (Multi-DC (TEC1)), and 4 others: FY18/19 TEC1.6 Q4: Improve or replace GTID + pt-heartbeat logic for cross-DC - https://phabricator.wikimedia.org/T221159 (10mobrovac) [23:51:42] 10DBA, 10MediaWiki-Database, 10Core Platform Team (Multi-DC (TEC1)), 10Core Platform Team Backlog (Watching / External), and 4 others: Fix mediawiki heartbeat model, change pt-heartbeat model to not use super-user, avoid SPOF and switch automatically to the real ma... - https://phabricator.wikimedia.org/T172497 [23:52:42] 10DBA, 10MediaWiki-Database, 10Core Platform Team Backlog (Watching / External), 10Performance-Team (Radar), and 2 others: Fix mediawiki heartbeat model, change pt-heartbeat model to not use super-user, avoid SPOF and switch automatically to the real master without... - https://phabricator.wikimedia.org/T172497 [23:55:56] 10DBA, 10MediaWiki-Database, 10Performance-Team, 10Core Platform Team (Multi-DC (TEC1)), and 4 others: FY18/19 TEC1.6 Q4: Improve or replace GTID + pt-heartbeat logic for cross-DC - https://phabricator.wikimedia.org/T221159 (10Krinkle) [23:56:24] 10DBA, 10MediaWiki-Database, 10Performance-Team, 10Core Platform Team (Multi-DC (TEC1)), and 4 others: FY18/19 TEC1.6 Q4: Improve or replace GTID + pt-heartbeat logic for cross-DC - https://phabricator.wikimedia.org/T221159 (10Krinkle) (unexpected T ref in task description as it overflows with scroll, the...