00:17:55 derpy sent me this message: it appears the /meta/milestones are huge since 0.20. Also, the rcfiles dir, can you delete any default .rc files and replace them with a symlink to a default.rc 00:19:00 is it even possible to delete default.rc files and replace with a symlink to a default? I don't like making any non-standard changes to the setup so I prefer to not do anything different for fear of unintended consequences that will require hours of fixing 00:19:31 additionally, and probably more importantly, I don't think the rc files are more than a drop in the ocean on the filesize 00:19:40 the issue is the tons and tons of trunk installs 00:26:29 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-508-gf08a75f (34) 00:41:33 johnstein: derpy is just confused as to what milestones/logfile is I think 00:41:39 those files are always going to be huge 00:42:06 I doubt 0.20 is some crazy order of magnitude bigger or anything 00:42:15 well, unless there's some weird bug with milestones I guess 00:42:41 yeah I would also guess that rc files are not the cause of space issues 00:47:01 ^status 00:47:01 31 Crawlers. CBRO disk usage=96% (135GB) | RAM usage=26% (4GB)| uptime/CPU= 00:47:01 up 270 days, 6:25, 2 users, load average: 0.50, 0.50, 0.55 (4 Cores) http://status.berotato.org 00:51:50 little do users know that cbro has the Powered by Pain mutation 00:52:00 as its disk becomes more full, their RNG improves 00:52:26 if I actually had free time I'd work on migrating to something bigger and set up a group of trusted admins 00:58:48 what size are the hard drives of the other crawl servers? 01:44:28 I think cbro has a pretty small partition for a server 01:52:28 this was my impression as well 03:21:08 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-508-gf08a75f (34) 04:27:14 -!- amalloy_ is now known as amalloy 05:16:24 -!- amalloy is now known as amalloy_ 08:22:30 -!- Amnesiac_ is now known as Amnesiac 08:49:11 03advil02 07* 0.21-a0-509-gecc5317: Do a better job setting cursor size in fontbuf_line_reader (aidanh) 10(81 seconds ago, 2 files, 16+ 10-) 13https://github.com/crawl/crawl/commit/ecc531740210 08:56:15 on the one hand, I'm very glad someone has already written all this sdl/gl code, because I doubt I could do it myself, but on the other hand, I wish it had just a tiny bit less levels of indirection / unnecessary abstraction 09:22:07 advil: that fixes the issue for me, thanks 09:23:26 seperately there seems to be a bug with the font rendering: text has a black shadow 09:24:02 looks like the antialiased rendering is done without an alpha channel or something 10:11:50 aidanh: yes, I was trying to figure that out; it appears that only one of the two ways of drawing text does blending 10:15:01 so the text on the actual texture atlas is rendered ok, but when rendering some geometry it's using the wrong blending mode? 10:25:31 ^status 10:25:31 30 Crawlers. CBRO disk usage=91% (135GB) | RAM usage=29% (4GB)| uptime/CPU= 10:25:31 up 270 days, 16:03, 3 users, load average: 0.65, 0.45, 0.38 (4 Cores) http://status.berotato.org 10:25:53 ^status 10:25:53 30 Crawlers. CBRO disk usage=91% (135GB) | RAM usage=30% (4GB)| uptime/CPU= 10:25:53 up 270 days, 16:04, 3 users, load average: 0.55, 0.45, 0.38 (4 Cores) http://status.berotato.org 10:26:09 that's pretty much the most space I can clear up 10:26:37 well, actually I can start throwing out all ttyrecs except for the past 2 weeks (as opposed to the last 30 days like I'm doing now) 10:26:49 thowing out = rsync 10:27:31 the only issue would be a game that's actively running. I need to skip those. I figure it's safe to say no one would be playing for 2 weeks in the same session 10:32:00 -!- wwwn is now known as somethingasometh 10:33:06 -!- somethingasometh is now known as antikek 11:26:33 -!- antikek is now known as kekdisciple 11:30:34 -!- kekdisciple is now known as kooky 11:52:43 Good evening :) 12:12:19 Is there an easy way to make webtiles reload the html without restarting it? 12:12:52 I kind of want to warn people about upcomming instability during this weekend as I'll be trying to tweak the server to get faster at startups 12:22:39 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-509-gecc5317 (34) 12:43:24 <|amethyst> TZer0: the banner you mean? 12:43:59 |amethyst: the one called banner.html 12:44:03 The top part on webtiles. 12:45:59 <|amethyst> TZer0: If just updating banner.html doesn't do it, try touch $chroot/crawl-master/webserver/templates/client.html 12:47:33 TZer0: btw, are you aware that footv isn't working for cue still? 12:47:40 I saw someone mention that yesterday 12:48:05 I think it can't fetch the timestamp file when it tries to play a ttyrec, but the ttyrec link -ttyrec gives works fine 12:48:14 !lm * cue -tv 12:48:19 2794493. MummyMadness, XL13 MuEn, T:33565 (milestone) requested for FooTV: telnet://termcast.develz.org or http://termcast.develz.org. 12:49:10 well you can't see the full error, but the ttyrec url it gives does work ok 12:49:13 !lm * cue -ttyrec 12:49:15 2794493. MummyMadness, XL13 MuEn, T:33565 (milestone): https://underhound.eu/crawl/ttyrec/MummyMadness/2017-11-28.17:22:26.ttyrec 12:55:07 Okay, I'll look at it in a sec. 13:16:07 |amethyst: didn't work. 13:16:20 <|amethyst> hm 13:16:26 !lm * cue -tv 13:16:27 2794519. Smoofers, XL5 GrBe, T:1511 (milestone) requested for FooTV: telnet://termcast.develz.org or http://termcast.develz.org. 13:16:38 <|amethyst> I know I've done that before 13:17:06 <|amethyst> I can't think of anything special I did to make it care about the date 13:17:08 gammafunk: how does it parse the listing? 13:17:25 Also, the :81 should be removed and http should be changed to https 13:17:39 <|amethyst> ah 13:17:42 I have an automatic redirect, so it may be getting confused. 13:30:45 oh that would be it, yeah 13:31:23 The redirect? 13:31:29 Or the fact that I run nginx, not apache 13:31:34 and thus the file listings may be different 13:31:45 hrm 13:31:55 the config doesn't have :81 and is using https 13:31:57 Try fixing the URL first and we'll see. 13:32:03 Hmm, that's what it showed in the GUI 13:32:24 !lm * cue -tv 13:32:25 2794537. Smoofers, XL5 GrBe, T:2795 (milestone) requested for FooTV: telnet://termcast.develz.org or http://termcast.develz.org. 13:32:41 "Fetching ttyrec listing from http://underhound.eu:81/crawl/ttyrec/Smoofers..." 13:32:50 yeah it's trying to go to :81 but I'm not sure why 13:32:52 that's not in the config 13:33:54 is that just an old config that's installed for some reason? 13:34:06 I'll have to contact snark about that one 13:34:31 Hmm 13:34:52 https://github.com/crawl/sequell/blob/master/config/sources.yml#L124 13:34:52 has the right base URL 13:34:54 hrm 13:35:00 !lm * cue -ttyrec 13:35:01 2794537. Smoofers, XL5 GrBe, T:2795 (milestone): https://underhound.eu/crawl/ttyrec/Smoofers/2017-11-28.18:23:58.ttyrec 13:35:06 ah, and it's correct there 13:35:16 It is. 13:35:50 maybe it's some stale config; I'll ask snark since it looks like it's on shallot's side of things, thanks 13:35:50 And the URL works... 13:35:57 Okay, no worries 13:36:11 unless you have some kind of redirect happening on timestamp files or something 13:36:16 I don't actually know where those are stored 13:36:16 Nope 13:36:30 It is a pure nginx file listing on those folders 13:36:35 nothing special magic going on there. 13:37:40 location /crawl/ { root /var/www; autoindex on; } 13:37:42 That's the entire thing. 13:42:19 <|amethyst> I get a 400 with wget 13:42:22 <|amethyst> it works in chromium 13:42:23 <|amethyst> on 13:42:30 <|amethyst> http://underhound.eu:81/crawl/ttyrec/Smoofers/2017-11-28.18:23:58.ttyrec 13:43:24 <|amethyst> maybe wget doesn't like http to https requests, but that doesn't explain getting a 400 13:44:12 <|amethyst> s/ques/direc/ 13:47:17 Ah, yeah, because the request has to be https:// 13:47:20 So you get a redirect to it 13:47:39 Hmm 13:47:40 Ahh... 13:47:43 I see. 13:48:15 <|amethyst> is it HSTS? 13:48:29 <|amethyst> so the browser rewrites the http: to https: automatically, but wget doesn't? 13:48:32 Yeah 13:49:33 <|amethyst> hm, wget is supposed to do HSTS by default though 13:49:43 Try now? 13:49:46 <|amethyst> maybe not the versions on the servers, but the version I just tested with should 13:49:53 <|amethyst> works now 13:49:58 <|amethyst> !lm * cue -tv 13:49:59 2794540. Ge0ff, XL25 HuSu, T:91400 (milestone) requested for FooTV: telnet://termcast.develz.org or http://termcast.develz.org. 13:50:09 <|amethyst> nice 13:50:41 <|amethyst> footv works now too it seems 13:50:57 Yeah, but we should get it over to https anyway 13:51:42 <|amethyst> yeah 13:51:54 <|amethyst> gammafunk: is footv in one of our repos? 13:56:28 <|amethyst> gammafunk: if not, we should get snark to put it there 14:06:16 |amethyst: I wasn't even aware that it had its own config/codebase 14:06:23 thought it was part of the sequell codebase 14:06:40 <|amethyst> pretty sure it's not in sequell 14:07:06 <|amethyst> where by "pretty sure" I mean "this would totally collapse my worldview" 14:10:10 <|amethyst> /nick hyperbolic # that nick seems to be taken! 14:11:16 <|amethyst> so, "hyperbole" and "hyperbola" are both words—why isn't "parabole" a word 14:11:18 <|amethyst> "parable" doesn't count, unless you make "hyperble" a word too 14:11:40 -!- |amethyst is now known as hyperble 14:12:00 lol I'm totally a conic section 14:12:20 -!- hyperble is now known as |amethyst 14:12:26 -!- |amethyst is now known as |amethyts 14:12:34 <|amethyts> lol I'm mispelt 14:45:08 -!- |amethyts is now known as |amethyst 17:02:07 -!- MadCoyote is now known as FunkyBomb 17:13:42 iirc parabole *is* a word --- just not in the same context 17:13:54 (but, weirdly, related --- distantly) 18:18:09 Unstable branch on underhound.eu updated to: 0.21-a0-509-gecc5317402 (34) 22:57:23 -!- n1 is now known as caveman1000 22:58:01 -!- caveman1000 is now known as n1 23:59:31 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-509-gecc5317 (34)