[09:52:53] RC down [16:31:22] yes that's known [16:31:38] petan: Matthew: ^ [16:32:55] Yeah, feed stuff is the one area I'm not super familiar with. [18:33:29] Huh—does wm-bot use the same xmlrcsd instance as Huggle? [18:33:36] If so, that's the problem, as that's down on Huggle too. [18:36:25] Well, it's been rebooted. /shrug [18:39:17] yeah that's related [18:39:19] @reboot [18:39:23] System is shutting down, requested by petan from #wm-bot [18:39:48] it needs to be restarted after restart of xmlrcs [18:39:48] systemd all the things! [18:40:25] Only tangentially related, but despite all the hate systemd gets, I really haven't had a bad experience with it on the smaller side of things. [18:40:38] I did [18:42:18] main problem is that it's all part of one big blob, and when it crashes there is no other cure than rebooting the OS, which isn't always trivial, when you run some business critical servers, I know in modern days you should containerize your workload and have it distributed across multiple nodes, but it's not always possible (DB server for example) [18:42:25] systemd is essentially against all UNIX principles [18:42:42] "it's one big blob" sums its issues up pretty well, honestly :P [18:42:58] each program should be designated to do only its one own job and do it well, systemd is trying to do million jobs at once and suck at it in same time [18:43:29] Also, doesn't its logging suck? I feel like I remember that being a pain. [18:43:34] most of criticism it gets is well deserved [18:43:58] I don't even know why it needs to be in charge of logging in a first place [18:44:05] there are standardized ways to deal with logs [18:45:53] I think one argument for it though is that "systemd isn't an init system", but instead more of a collection of utilities for running a system which includes an init program [18:46:15] (Kinda reminds of the Emacs joke: "Emacs is a great operating system; it's just missing a decent text editor".) [18:47:40] yes, one day systemd will probably replace even kernel, it's gonna all in one, like busybox on steroids [18:48:17] Lol [18:50:04] But yeah—systemd kinda is just like a lot of mildly small, composable bits (à la the UNIX philosophy) that ends up as one big bundled piece of burning metal. [18:50:49] While I do think a lot of the criticism it receives is legitimate, I do think the actual "usefulness"/"usability" or it is not that bad. [18:51:45] There's also issues of scope creep and whatnot, but when used as a lightweight (in heavy quotes) process manager it does work. [18:57:13] 10WM-Bot, 10User-MacFan4000: wm-bot not reporting on recent changes (5) - https://phabricator.wikimedia.org/T275966 (10Perryprog) 05Open→03Resolved [18:57:26] 10WM-Bot, 10User-MacFan4000: wm-bot not reporting on recent changes (5) - https://phabricator.wikimedia.org/T275966 (10Perryprog) See also T228121. [18:57:33] pingy mc pign [18:57:35] ping* [18:58:58] Eh, that original task should probably be made into one of those tracking issues or something. That's not my job, though. [19:18:37] http://wm-bot.wmflabs.org/dump/%23wm-bot.htm [19:18:37] @info