[05:53:43] 10Traffic, 10Discovery-Search, 10Elasticsearch, 10Operations: Enable nginx prometheus metrics for all elastic nodes - https://phabricator.wikimedia.org/T216681 (10Aklapper) a:05Mathew.onipe→03None Removing assignee @mathew.onipe as the user does not seem to be active anymore. [08:26:32] 10Traffic, 10Core Platform Team, 10Operations: Move all purge traffic to kafka - https://phabricator.wikimedia.org/T250781 (10ema) [10:23:00] 10netops, 10Operations, 10observability: Investigate Juniper structured logs - https://phabricator.wikimedia.org/T250703 (10ayounsi) Looks like it's a mixed bag, not all logs are structured, eg: ` <78>1 2020-04-21T10:00:00.049Z re0.cr2-esams /usr/sbin/cron 59827 - - (root) CMD ( /usr/libexec/atrun) <35>1 2... [10:55:29] vgutierrez: https://netflixtechblog.com/how-netflix-brings-safer-and-faster-streaming-experience-to-the-living-room-on-crowded-networks-78b8de7f758c [11:01:14] gilles: there are some performances data too you might be interested in ^ [11:03:44] well, good for Netflix, I'm guessing they're not using ATS 🙃 [11:22:51] 10Traffic, 10Operations, 10serviceops, 10Patch-For-Review: Applayer services without TLS - https://phabricator.wikimedia.org/T210411 (10Dzahn) [12:16:49] 10Traffic, 10Operations, 10serviceops, 10Patch-For-Review: Applayer services without TLS - https://phabricator.wikimedia.org/T210411 (10Dzahn) [12:18:23] 10Traffic, 10Operations, 10serviceops, 10Patch-For-Review: Applayer services without TLS - https://phabricator.wikimedia.org/T210411 (10Dzahn) The change for contint (integration.wikimedia.org, integration.mediawiki.org) had to be reverted because https://integration.wikimedia.org/ci/ returned 502s. [13:36:48] 10Traffic, 10Analytics, 10Operations: Create replacement for Varnishkafka - https://phabricator.wikimedia.org/T237993 (10elukey) Me and Ema had several chats on IRC, reporting in here the summary: - the HTTP status `000` seems to be used for clients that have some trouble doing a HTTP request to ats-tls, wi... [16:40:54] 10Traffic, 10Cloud-Services, 10Operations, 10Wikimedia-Incident, 10cloud-services-team (Kanban): Requests to production are sometimes timing out or giving empty response - https://phabricator.wikimedia.org/T249035 (10JHedden) p:05Triage→03Medium [19:36:24] 10Wikimedia-Apache-configuration, 10MediaWiki-Configuration: noc.wikimedia.org incorrect raw file urls - https://phabricator.wikimedia.org/T250852 (10Dvorapa) [19:36:44] 10Wikimedia-Apache-configuration, 10MediaWiki-Configuration: noc.wikimedia.org incorrect raw file urls - https://phabricator.wikimedia.org/T250852 (10Dvorapa) [19:39:34] 10Wikimedia-Apache-configuration, 10MediaWiki-Configuration: noc.wikimedia.org incorrect raw file urls - https://phabricator.wikimedia.org/T250852 (10Dvorapa) It seems in some cases .txt is needed and in some cases .txt is not needed: T227606 [20:20:06] 10Traffic, 10Core Platform Team, 10Operations: Move all purge traffic to kafka - https://phabricator.wikimedia.org/T250781 (10Pchelolo) AFAIK cdnPurgeJob is only involved if the delayed purge is required if reboundDelay option is set. For every rebound purge job there's an instant purge multicast. Is the req... [20:26:03] 10Traffic, 10Core Platform Team, 10Operations: Move all purge traffic to kafka - https://phabricator.wikimedia.org/T250781 (10holger.knust) p:05Triage→03Medium [20:42:52] 10Wikimedia-Apache-configuration, 10MediaWiki-Configuration, 10Patch-For-Review, 10User-Urbanecm: noc.wikimedia.org incorrect raw file urls - https://phabricator.wikimedia.org/T250852 (10Urbanecm) a:03Urbanecm Makes sense.