[19:22:19] . [19:22:43] dmehus: done [19:23:50] PROBLEM - wiki.yumeka.team - LetsEncrypt on sslhost is CRITICAL: CRITICAL - Socket timeout after 10 seconds [19:31:27] @RhinosF1|NotHere, ty [19:32:01] and lol at your "oh find url yourself" comment :P [19:34:21] @RhinosF1|NotHere What's the process for inviting MH-Discord here? [19:34:30] as that still needs to be done [19:34:32] dmehus: same as I said earlier [19:34:36] oh right [19:34:39] Edit a json file [19:34:48] miraheze/mw-config - paladox the build passed. [19:35:00] can I do a pull request to a private repo? [19:35:06] No [19:35:13] miraheze/mw-config - paladox the build passed. [19:35:19] I said it's on server [19:35:32] Are you mobile, or is your laptop available? Can you do the PR and push it to the private repo? [19:35:42] since Reception123 said any MW engineer can do it [19:35:47] I am mobile [19:35:52] oh ok [19:35:52] There's no private repo [19:36:09] And MacFan,Zppix,me,Reception123 and Void have access [19:36:21] okay [19:36:29] so it's not on GitHub? [19:36:36] No [19:36:41] oh [19:36:42] ok [19:36:50] If you nag me loud enough on Wednesday I'll do the config file [19:36:52] Let me ask MacFan4000 then [19:37:02] To puppet [19:37:10] But it'll still need manual deploy [19:41:34] yeah [19:41:37] that's cool [19:42:55] [mw-config] paladox closed pull request #3720: Test: Do Not merge - https://git.io/JtVTy [19:42:57] [mw-config] paladox deleted branch paladox-patch-2 - https://git.io/vbvb3 [19:42:59] [miraheze/mw-config] paladox deleted branch paladox-patch-2 [19:43:17] miraheze/mw-config - paladox the build passed. [19:43:28] [mw-config] paladox pushed 1 commit to paladox-patch-2 [+0/-0/±1] https://git.io/JtVTD [19:43:30] [mw-config] paladox f1d6354 - Test: Do Not merge [19:43:31] [mw-config] paladox created branch paladox-patch-2 - https://git.io/vbvb3 [19:43:33] [mw-config] paladox reopened pull request #3720: Test: Do Not merge - https://git.io/JtVTy [19:43:34] [mw-config] paladox closed pull request #3720: Test: Do Not merge - https://git.io/JtVTy [19:43:36] [mw-config] paladox deleted branch paladox-patch-2 - https://git.io/vbvb3 [19:43:37] [mw-config] paladox deleted branch paladox-patch-2 [19:43:49] hmm, it sends deleted branch twice [19:44:04] RECOVERY - ping6 on cp3 is OK: PING OK - Packet loss = 0%, RTA = 236.17 ms [19:48:15] PROBLEM - ping6 on cp3 is WARNING: PING WARNING - Packet loss = 0%, RTA = 255.33 ms [19:51:40] RECOVERY - wiki.yumeka.team - LetsEncrypt on sslhost is OK: OK - Certificate 'wiki.yumeka.team' will expire on Mon 07 Feb 2022 23:59:59 GMT +0000. [20:03:53] dmehus: does the relay work? [20:05:44] I just added MirahezeBot [20:05:58] It should probably get flags [20:07:58] MacFan4000: rss so it'll disappear on reboot? [20:08:12] RhinosF1|NotHere relay doesn't work [20:08:24] MacFan4000: ^ [20:08:33] I used .join [20:08:49] It auto updates the file [20:09:08] MacFan4000: doesn't persist iirc because rss messes that up. [20:11:18] As for the relay - I added it to the map and to the irc channels so not sure [20:12:01] RhinosF1|NotHere: ^ (restarted without an issue) [20:12:07] Ok [20:13:25] I trust: Reception123!.*@miraheze/Reception123 (2admin), .*@miraheze/John (2admin), .*@miraheze/RhinosF1 (2admin), .*@wikimedia/paladox (2admin), .*@wikipedia/Southparkfan (2admin), .*@miraheze/Universal-Omega (2admin), [20:13:25] @trusted [20:16:07] "#sre":"miraheze-sre" [20:16:17] That’s the entry in the map [20:16:56] PROBLEM - wiki.yumeka.team - LetsEncrypt on sslhost is CRITICAL: CRITICAL - Socket timeout after 10 seconds [20:17:45] @RhinosF1|NotHere Just saw...doesn't seem to be yet, no [20:18:10] [02miraheze/mediawiki] 07paladox pushed 031 commit to 03REL1_35 [+0/-0/±1] 13https://git.io/JtVI8 [20:18:11] [02miraheze/mediawiki] 07paladox 03a6c9a53 - Update Medik [20:18:36] I’m not sure why it isn’t working [20:19:02] Oh wait [20:19:04] MacFan4000: maybe try restarting MH-Discord? can't hurt, can it? [20:19:04] The only thing I can think of would be if it’s not in #sre on discord [20:19:05] I know [20:19:07] I think [20:19:19] I think the Discord bot needs send messages permissions in the channel [20:19:21] Let me look [20:19:37] dmehus: it should have htem [20:19:38] *them [20:19:51] as the "relay bot" role gives permissions out across the server [20:20:31] @Reception123, wasn't sure if the channel permissions were specific but yeah [20:20:32] dmehus: I just saw you tick the "Send messages", but the / means that it keeps the server-wide permission [20:20:34] that wasn't it [20:20:45] unfortunately yeah, it already has send permission :( [20:20:48] [02miraheze/ManageWiki] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVIu [20:20:50] [02miraheze/ManageWiki] 07Universal-Omega 03303bd2a - Use #miraheze-sre [20:20:50] yeah :( [20:20:56] I think a restart couldn't hurt, it does fix things sometimes [20:21:04] yeah [20:21:07] we could try that [20:21:13] unless it was Omega's commit [20:21:18] that fixed it [20:21:32] dmehus: that's for github [20:21:33] [02miraheze/CreateWiki] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVIg [20:21:34] [02miraheze/CreateWiki] 07Universal-Omega 03824556e - Use #miraheze-sre [20:21:39] Restarted [20:21:46] miraheze/ManageWiki - Universal-Omega the build passed. [20:21:50] doesn't seem to have fixed it :( oh well it was worth a try [20:22:09] ah [20:22:11] MacFan4000: is there maybe something extra that needs to be ran for it to take effect? [20:22:21] [02miraheze/landing] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVI2 [20:22:22] [02miraheze/landing] 07Universal-Omega 03d4d9976 - Use #miraheze-sre [20:22:35] I’ve never tried to add a channel before so no idea [20:22:41] miraheze/CreateWiki - Universal-Omega the build passed. [20:22:55] too bad there's not any docs [20:23:26] miraheze/landing - Universal-Omega the build passed. [20:23:38] [02miraheze/DataDump] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVIw [20:23:40] [02miraheze/DataDump] 07Universal-Omega 0312bd021 - Use #miraheze-sre [20:23:57] Zppix may know - it’s his bot [20:24:09] yeah, he probably does but he doesn't seem around [20:24:18] [02miraheze/MatomoAnalytics] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVIK [20:24:19] [02miraheze/MatomoAnalytics] 07Universal-Omega 035605b5f - Use #miraheze-sre [20:24:41] miraheze/DataDump - Universal-Omega the build passed. [20:24:58] [02miraheze/MirahezeMagic] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVIP [20:24:59] [02miraheze/MirahezeMagic] 07Universal-Omega 03089aa63 - Use #miraheze-sre [20:25:19] MacFan4000: do you have any idea where the bot "comes from"? [20:25:21] miraheze/MatomoAnalytics - Universal-Omega the build passed. [20:25:27] since Zppix didn't make it himself I don't think [20:25:32] so maybe there would be docs there [20:25:33] Reception123: they are docs [20:25:42] And litterally google discord-irc [20:25:42] [02phabricator-extensions] 07Universal-Omega opened pull request 03#10: Use #miraheze-sre - 13https://git.io/JtVIM [20:26:00] https://github.com/reactiflux/discord-irc [20:26:01] [ GitHub - reactiflux/discord-irc: Connects Discord and IRC channels by sending messages back and forth. ] - github.com [20:26:01] miraheze/MirahezeMagic - Universal-Omega the build passed. [20:26:21] [02miraheze/WikiDiscover] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVI9 [20:26:22] [02miraheze/WikiDiscover] 07Universal-Omega 0351eccd1 - Use #miraheze-sre [20:26:49] miraheze/phabricator-extensions - Universal-Omega the build passed. [20:26:54] it doesn't seem to indicate anything else that needs to be done other than the JSON [20:26:56] [02miraheze/IncidentReporting] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVI7 [20:26:57] [02miraheze/IncidentReporting] 07Universal-Omega 03005a636 - Use #miraheze-sre [20:27:30] [02miraheze/CustomHeader] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVIb [20:27:31] miraheze/WikiDiscover - Universal-Omega the build passed. [20:27:32] [02miraheze/CustomHeader] 07Universal-Omega 03a538469 - Use #miraheze-sre [20:27:56] miraheze/IncidentReporting - Universal-Omega the build passed. [20:28:14] [02miraheze/PDFEmbed] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVIA [20:28:16] [02miraheze/PDFEmbed] 07Universal-Omega 035c1efbf - Use #miraheze-sre [20:28:26] miraheze/CustomHeader - Universal-Omega the build passed. [20:29:03] [02miraheze/RottenLinks] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVLv [20:29:04] [02miraheze/RottenLinks] 07Universal-Omega 03d0e3845 - Use #miraheze-sre [20:29:11] miraheze/PDFEmbed - Universal-Omega the build passed. [20:29:36] [02miraheze/SpriteSheet] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVLU [20:29:38] [02miraheze/SpriteSheet] 07Universal-Omega 03594d6da - Use #miraheze-sre [20:30:01] miraheze/RottenLinks - Universal-Omega the build passed. [20:30:06] [02miraheze/YouTube] 07Universal-Omega pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVLT [20:30:07] [02miraheze/YouTube] 07Universal-Omega 030941937 - Use #miraheze-sre [20:30:36] miraheze/SpriteSheet - Universal-Omega the build passed. [20:31:11] miraheze/YouTube - Universal-Omega the build passed. [20:31:18] [02phabricator-extensions] 07paladox closed pull request 03#10: Use #miraheze-sre - 13https://git.io/JtVIM [20:31:20] [02miraheze/phabricator-extensions] 07paladox pushed 031 commit to 03master [+0/-0/±1] 13https://git.io/JtVLt [20:31:21] [02miraheze/phabricator-extensions] 07Universal-Omega 030a1b0a8 - Use #miraheze-sre (#10) [20:32:17] miraheze/phabricator-extensions - paladox the build passed. [20:34:31] Reception123: they're isn't [20:34:37] It's litteraly a json file [20:34:48] 2 extra lines [20:34:59] well something clearly isn't working [20:35:02] see -bots, maybe Void will fix it [20:35:03] Someone send me the file with passwords redacted [20:36:29] [02miraheze/mediawiki] 07paladox pushed 031 commit to 03REL1_35 [+0/-0/±1] 13https://git.io/JtVLo [20:36:31] [02miraheze/mediawiki] 07paladox 03737804c - Update CSS [20:36:40] Ping [20:36:43] Ping [20:37:39] Relay [20:37:43] is up [20:37:44] heh of course Voidwalker fixed it :) [20:37:45] RECOVERY - wiki.yumeka.team - LetsEncrypt on sslhost is OK: OK - Certificate 'wiki.yumeka.team' will expire on Mon 07 Feb 2022 23:59:59 GMT +0000. [20:37:55] * Reception123 gives Voidwalker sysadmin of the year award even though he's not a sysadmin [20:38:11] so how did you fix? :) [20:38:26] "#sre":"miraheze-sre" -> "#sre":"#miraheze-sre" [20:38:33] just a single missing character :P [20:38:54] ah [20:39:21] it's always that kind of thing that does it [20:43:53] I just set the MirahezeBot chanops list [20:44:28] If somebody could please give it at least +o you’ll be able to use .op and such [20:44:50] Nice. 🙂 [20:45:00] (It’s inheriting from #miraheze) [20:45:05] I've blacklisted embed rights for this channel and #irc-relay for the relay bot. [20:45:21] Oh, nice. [20:45:22] i don't think we need it to embed links (makes things harder to read as it just clutters) [20:45:29] yeah [20:45:39] this channel shouldn't need embeds really by anyone [20:46:02] yeah, it's for sre business [20:46:09] https://git.io/JtVLo [20:46:10] [ Comparing a6c9a53a46dd...737804c46a11 · miraheze/mediawiki · GitHub ] - git.io [20:46:11] embeds aren't really useful for that [20:46:11] testing [20:46:28] yeah [20:46:37] Seems to be working [20:46:46] I didn't see any embed for that link [20:46:48] Reception123: can you please give MirahezeBot +o in here [20:47:05] done [20:47:05] ty [20:47:10] no problem :) [20:47:10] ty @Reception123 [20:47:24] .op [20:47:24] MacFan4000: Access Denied. If in error, please contact the channel founder. [20:47:42] That working as it should :) [20:47:52] yeah [20:48:09] .op [20:48:09] Attempting to OP... [20:48:13] heh [20:50:03] heh music troll doesn't know about this channel, so that's a 👍 [20:55:31] [02miraheze/mediawiki] 07paladox pushed 031 commit to 03REL1_35 [+0/-0/±1] 13https://git.io/JtVtP [20:55:33] [02miraheze/mediawiki] 07paladox 03d1b8f12 - Update CommentStreams [20:56:35] dmehus: heh, don't jinx it, it has been mentioned in #miraheze [20:56:45] let's just cross our fingers that the troll won't find out about it [20:57:44] just to be sure, we should have this channel "mirror" #miraheze bans like the others [20:57:51] I forgot how to do that though [20:58:39] mode +b $j:#miraheze [20:59:20] thanks! [21:00:01] Seen is now enabled in the channel [21:00:01] @seen-on [21:00:09] @statistics-on [21:00:10] Statistics were now enabled [21:00:20] http://wm-bot.wmflabs.org/dump/%23miraheze-sre.htm [21:00:20] @info [21:00:22] [ #miraheze-sre ] - wm-bot.wmflabs.org [21:00:57] MacFan4000: I'm sure we can do that ourselves [21:01:11] thanks for doing it though :) [21:01:47] MacFa4000, yes, thanks, and apologies for RhinosF1's unnecessary directness :( [21:01:58] s/MacFan4000/MacFan4000 [21:02:04] wait [21:02:09] why doesn't that work? [21:02:18] what bot controls that? [21:03:52] probably didn't work because it didn't actually match your message [21:04:15] lol yes, you just wrote MacFan twice :P [21:12:43] s/MacFa4000/MacFan4000 [21:12:43] dmehus meant to say: MacFan4000, yes, thanks, and apologies for RhinosF1's unnecessary directness :( [21:12:52] @Voidwalker lol I did too [21:12:57] and thanks also @Reception123 [21:18:40] https://grafana.miraheze.org/d/W9MIkA7iz/miraheze-cluster?orgId=1&from=now-1h&to=now-1m&var-job=node&var-node=dbbackup1.miraheze.org&var-port=9100 pity [21:18:40] [ Grafana ] - grafana.miraheze.org [21:19:15] even with reduced load the disks aren't fast enough to cope well with the replication load [21:20:27] SPF|Cloud, ah, interesting...yeah, we're still seeing a fair bit of slowness right now [21:20:57] I don't think this is the cause of any of your performance issues [21:21:38] Well, I mean querying user lists and page searching would be a check of the database, so it might be. [21:21:51] Would increasing the number of course from 2 to 4 help? [21:22:00] s/course/cores [21:22:00] dmehus meant to say: Would increasing the number of cores from 2 to 4 help? [21:22:51] no MediaWiki process accesses a dbbackup server [21:23:16] oh [21:23:45] so what is the downside you're seeing with the dbbackup1 server then? [21:23:55] where does that manifest itself? [21:24:18] it's a backup server, not a production replica. the first is to help us create proper database backups, the second helps to increase service availability and reduce load on the master [21:24:50] ability of the backup server to stay in sync with the latest wiki edits, speed of backup creation [21:25:02] yeah, and ah that makes sense [21:27:00] db12 mariadb has an uptime of almost 17 weeks :o [21:28:38] wow [21:43:27] dmehus: heh you’re already top of wm-bots statistics list [21:46:35] MacFan4000, lol yeah. Just in the Miraheze channels list though, I think, or is there statistics for all channels? [21:46:50] http://wm-bot.wmflabs.org/dump/%23miraheze-sre.htm [21:46:50] @info [21:46:51] [ #miraheze-sre ] - wm-bot.wmflabs.org [21:46:55] This channel [21:47:07] oh heh [21:47:43] heh we'll actually have a chance of topping the bots in #miraheze now :P [21:48:05] In there it’s currently still icinga [21:48:12] https://wm-bot.wmflabs.org/dump/%23miraheze.htm [21:48:13] [ #miraheze ] - wm-bot.wmflabs.org [21:48:30] I won't last for long once MH-Discord, icinga-miraheze, and Not-7024 produce more notifications [21:48:47] yeah [21:48:59] even paladox is a distant second at 34,000-ish messages [21:49:28] I almost have more than NDKilla though :P [21:49:47] Icinga is at 98935 which is still way more [21:50:25] yeah [21:59:43] !log depool/repool mw10 and rebuild lc [21:59:47] Logged the message at https://meta.miraheze.org/wiki/Tech:Server_admin_log [21:59:54] paladox: why? [22:00:30] I'm trying to see if this will help with https://phabricator.miraheze.org/T6826. I rebuilt mw9 yesturday which doesn't seem to be experencing the issue anymore. [22:00:31] [ ⚓ T6826 Core Namespace translations having no effect ] - phabricator.miraheze.org [22:00:39] it won't [22:00:42] We've been over it [22:00:43] https://phabricator.miraheze.org/T6826 [22:00:44] [ ⚓ T6826 Core Namespace translations having no effect ] - phabricator.miraheze.org [22:01:00] oh [22:01:07] when running var_dump( Title::newFromText( 'Plantilla:AD' ) );, you can see differences between mw9 and mw11 [22:01:20] If you want to see if i18n would, run MediaWikiServices::getInstance()->getLocalisationCache()->getItem( $code, 'namespaceNames' ) [22:01:33] on mw9, Title->mNamespace is 10, but on mw11 Title-mNamespace is 0 [22:01:34] PROBLEM - cp11 Varnish Backends on cp11 is CRITICAL: 1 backends are down. mw10 [22:01:51] If that returns none-English, LC isn't the problem. And so far, I've ran that on every server and it's worked fine [22:02:22] PROBLEM - cp12 Varnish Backends on cp12 is CRITICAL: 1 backends are down. mw10 [22:02:40] PROBLEM - cp3 Varnish Backends on cp3 is CRITICAL: 1 backends are down. mw10 [22:02:57] PROBLEM - cp10 Varnish Backends on cp10 is CRITICAL: 1 backends are down. mw10 [22:03:49] > var_dump(MediaWiki\MediaWikiServices::getInstance()->getLocalisationCache()->getItem( 'es', 'Plantilla' )); [22:03:49] NULL [22:04:49] uh, 'Plantilla' should be 'namespaceNames' [22:05:00] https://es.nocyclo.tk/wiki/Plantilla:AD works on mw10 now [22:05:02] [ Plantilla:AD - Risapedia, la enciclopedia seria ] - es.nocyclo.tk [22:05:04] after rebuilding lc [22:05:09] by wiping it [22:06:37] But that’s not the fix [22:06:53] If it was, it would have worked the first 3 times you did it :) [22:06:56] RECOVERY - cp10 Varnish Backends on cp10 is OK: All 7 backends are healthy [22:07:11] JohnLewis i mean it did [22:07:17] I also don't think constantly clearing things and reinstalling them is sustainable [22:07:17] The fix it to clear CreateWiki cache *correctly* [22:07:19] mw9 is working without issue since i did that yesturday [22:07:35] RECOVERY - cp11 Varnish Backends on cp11 is OK: All 7 backends are healthy [22:07:44] So why isn’t mw10 or mw11 which you did yesterday and the day before? [22:07:56] I didn't wipe lc on mw 10/11 though [22:08:05] You reinstalled them [22:08:06] i just rebuilt it by running the command, i wiped mw9 lc [22:08:13] no, i reinstalled mw9 [22:08:22] RECOVERY - cp12 Varnish Backends on cp12 is OK: All 7 backends are healthy [22:08:27] But you ran that command that proves it’s working [22:08:31] aka not LC [22:08:41] RECOVERY - cp3 Varnish Backends on cp3 is OK: All 7 backends are healthy [22:11:48] please don't rebuild on mw11 [22:12:41] SPF|Cloud; if you’re debugging it, I can tell you problem now because I already debugged earlier and said what the problem was [22:12:42] sure i'm not :) [22:12:57] what was the problem? [22:13:12] English namespace names are cached in CreateWiki cache [22:13:57] oh [22:14:06] what would the fix be to prevent that happening JohnLewis ? [22:14:29] Clearing the cache via resetWiki [22:14:51] ok, but i mean prevent it happening again [22:15:49] a) support English only wikis b) drop support for changing language via CreateWiki c) drop i18n support for namespaces on Miraheze [22:16:42] or d) revert CreateWiki/ManageWiki back about a year to using all.dblist and remove pre-setup configuration [22:16:53] oh [22:18:42] SPF|Cloud can i clear CW cache? [22:18:49] or do you want me to hold off [22:18:51] *still [22:18:55] go for it [22:19:07] thanks! [22:19:07] JohnLewis: why is it only happening now and wasn't before the past couple days? [22:19:54] !log root@jobrunner3:/home/paladox# /usr/local/bin/foreachwikiindblist /srv/mediawiki/w/cache/databases.json /home/paladox/resetWikiCaches.php [22:19:57] Logged the message at https://meta.miraheze.org/wiki/Tech:Server_admin_log [22:21:51] Universal_Omega; because before we had existing cache that provided the language for namespaces to use. Right now, wikis are assumed English when cache generation happens first time [22:22:24] Oh [22:22:31] Thanks [22:24:52] JohnLewis: couldn't it get that from database when creating the cache [22:25:49] > English namespace names are cached in CreateWiki cache @JohnLewis Oh interesting, is that the cause of #T6826? [22:29:38] RhinosF1; it will, the second time. Otherwise you’d need to provision MediaWiki without MediaWiki knowing what wiki to is, what the database server is, nothing [22:30:04] [02puppet] 07Universal-Omega edited pull request 03#1638: Remove old - 13https://git.io/JtauB [22:30:35] JohnLewis: how long does it take to generate the cache a second time? [22:30:38] [02puppet] 07Universal-Omega edited pull request 03#1638: Remove old infrastructure - 13https://git.io/JtauB [22:30:48] An action that causes the cache to regen [22:31:07] There’s no real clean fix and it’s a 1 in a 10000 action [22:31:31] we don’t do fresh server installs hourly, we don’t harshly wipe cache hourly, we don’t even do it monthly [22:31:56] so investing in a real fix when the solution is so easy and needs to be done so rarely makes it unrealistic [22:33:40] Ah [22:34:00] So it just needs cache to warm up on all servers and be cleared once [22:34:37] Yeah [22:34:59] Ah [22:35:16] But then once it’s done, it’s done. Unless you remove all cache files entirely, it’s extremely unlikely to reoccur. And if it does, it’s a bug [22:36:05] If we really wanted to I suppose we could force send a request to every wiki on every app server and then do a proper reset [22:36:27] But I'm too lazy to write that and it would take a while with the rate limit [22:36:51] I suppose only non English wikis would matter [22:37:30] [02puppet] 07dmehus commented on pull request 03#1638: Remove old infrastructure - 13https://git.io/JtV33 [22:37:36] It’s not even per server [22:37:50] cache expiration is centralised in Redis [22:37:57] Oh that's cool [22:52:00] JohnLewis, ah, that's helpful to know. Do we have a default cache expiry for MediaWiki (i.e., things like "page information using `action=info`)? If so, I'm wondering if we can lower the cache expiry to, say, 24 hours [22:55:38] CreateWiki cache is in theory infinite [23:04:07] JohnLewis, oh this is just CreateWiki cache we're talking about, right [23:05:42] Yes [23:12:23] ah, ok [23:22:49] PROBLEM - test3 APT on test3 is CRITICAL: CHECK_NRPE STATE CRITICAL: Socket timeout after 10 seconds. [23:24:47] RECOVERY - test3 APT on test3 is OK: APT OK: 28 packages available for upgrade (0 critical updates). [23:40:13] PROBLEM - ping6 on dbbackup2 is CRITICAL: PING CRITICAL - Packet loss = 100% [23:42:15] PROBLEM - ping6 on dbbackup2 is WARNING: PING WARNING - Packet loss = 0%, RTA = 102.75 ms