[05:18:37] 10serviceops, 10Operations, 10ops-eqiad: mw1286.mgmt is down - https://phabricator.wikimedia.org/T234009 (10jijiki) [06:31:23] 10serviceops, 10Operations: Update component/php72 to 7.2.22 - https://phabricator.wikimedia.org/T230024 (10jijiki) @Dzahn Is it ok if you upgrade on phab*? [06:34:41] 10serviceops, 10Operations, 10HHVM, 10Performance-Team (Radar): Remove HHVM from production - https://phabricator.wikimedia.org/T229792 (10jijiki) [16:43:58] 10serviceops: Free up space on mwdebug* - https://phabricator.wikimedia.org/T234063 (10jijiki) [17:30:06] 10serviceops, 10Operations, 10ops-eqiad: mw1286.mgmt is down - https://phabricator.wikimedia.org/T234009 (10Dzahn) Just tried to ssh to it now and it works for me. I get to login. [17:51:44] 10serviceops, 10Operations, 10ops-eqiad: mw1286.mgmt is down - https://phabricator.wikimedia.org/T234009 (10Dzahn) p:05Triage→03Normal @Cmjohnson This seems to be flapping. Sometimes it works and sometimes it doesn't. Could you check for loose cable and/or switch port, maybe just reconnecting it will do it. [18:23:13] 10serviceops: Free up space on mwdebug* - https://phabricator.wikimedia.org/T234063 (10Dzahn) 05Open→03Resolved [20:52:06] 10serviceops, 10Operations, 10Traffic, 10Puppet: Puppet systemd::mask is an anti pattern that has unwanted side effect - https://phabricator.wikimedia.org/T233839 (10Dzahn) [21:05:12] 10serviceops, 10Operations, 10Wikimedia-Logstash, 10observability: Errors managed by wmf-errors (like OOMs) lack normalized_message on logstash - https://phabricator.wikimedia.org/T233828 (10herron) These logs appear to be nesting the message field inside the exception field, and the message field at the r... [21:17:45] 10serviceops, 10Operations, 10Wikimedia-Logstash, 10observability, 10Patch-For-Review: Errors managed by wmf-errors (like OOMs) lack normalized_message on logstash - https://phabricator.wikimedia.org/T233828 (10herron) A few ideas to address this: Parse the nested exception field into the root of the lo... [21:21:18] 10serviceops, 10Operations, 10Wikimedia-Logstash, 10observability, 10Patch-For-Review: Errors managed by php-wmerrors (like OOMs) lack normalized_message on logstash - https://phabricator.wikimedia.org/T233828 (10Krinkle) [21:24:41] 10serviceops, 10Operations, 10Wikimedia-Logstash, 10observability, 10Patch-For-Review: Errors managed by php-wmerrors (like OOMs) lack normalized_message on logstash - https://phabricator.wikimedia.org/T233828 (10Krinkle) I think this should be fixed at the source in [puppet: php7-fatal-error.php](https:... [22:10:15] 10serviceops, 10Operations, 10Traffic, 10Patch-For-Review: Applayer services without TLS - https://phabricator.wikimedia.org/T210411 (10Dzahn) [22:13:04] 10serviceops, 10Operations, 10Traffic, 10Patch-For-Review: Applayer services without TLS - https://phabricator.wikimedia.org/T210411 (10Dzahn) >>! In T210411#5496180, @Vgutierrez wrote: > Please note that the docker-registry certificate is missing the public hostname: `docker-registry.wikimedia.org` Per I... [22:13:28] 10serviceops, 10Operations, 10Traffic, 10Patch-For-Review: Applayer services without TLS - https://phabricator.wikimedia.org/T210411 (10Dzahn) https://performance.wikimedia.org switch to https://performance.discovery.wmnet as backend. [22:40:35] 10serviceops, 10Operations: Update component/php72 to 7.2.22 - https://phabricator.wikimedia.org/T230024 (10Dzahn) @jijiki Any examples how it was done on the other servers? Did you keep the locally modified php.ini and fpm/php.ini files? I let the package overwrite but then let puppet revert that.