←2019-01-31 2019-02-01 2019-02-02→ ↑2019 ↑all
00:08:00 <oerjan> have you tried rotating it 4 times in 24 hours hth
00:13:19 <fizzie> b_jonas: It's just https://twitter.com/fungot
00:13:19 <fungot> fizzie: he fnord unable, and in any event the government were more often the man, and understandably, brexit.
00:13:24 <fizzie> Last tweet from 2012.
00:18:42 <b_jonas> I see
00:20:08 -!- oerjan has quit (Quit: Nite).
00:20:27 <fizzie> Error while checkign syntax automatically: (irony-server-error "irony-server is broken! Cannot open load file: No such file or directory, lisp-mnt")
00:20:40 <fizzie> Nothing ever works right if you don't keep using it. :/
00:53:10 <fizzie> Welp. Now there are no logs. :/
00:54:03 <fizzie> My deploy.sh is based on building at home and rsync'ing the binaries to the server. Turns out there's too much version shift between the systems now for that to work:
00:54:10 <fizzie> /home/esowiki/esologs: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.28' not found (required by /home/esowiki/esologs)
00:55:01 <fizzie> Well, at least I found this out by trying to restart the server component, so *collecting* logs should keep working.
00:55:20 <fizzie> I guess I can just build it on the server it needs to go to then.
00:56:15 -!- b_jonas has quit (Quit: leaving).
00:56:49 <fizzie> (Of course it doesn't have any of the necessary dependencies for building things.)
01:21:54 -!- xkapastel has quit (Quit: Connection closed for inactivity).
01:38:37 <fizzie> ...and of course the GCC 6.x available on the server's Debian version is too old to support <optional>...
01:50:12 <j4cbo> it might be <experimental/optional>
02:53:21 -!- salpynx has joined.
02:53:47 <fizzie> Oh, so it is.
02:54:05 <fizzie> Also managed to work around it by producing a statically linked binary.
02:54:27 <fizzie> It has the usual "Using 'getaddrinfo' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking" warnings, but seems like it was good enough.
02:56:41 <fizzie> Maybe by the next time this becomes an issue, buster has become the new stable.
02:58:44 <fizzie> (Also, there's no <experimental/variant>, so...)
03:21:00 -!- dauie_ has quit (Remote host closed the connection).
03:38:19 -!- salpynx has quit (Quit: Page closed).
05:14:57 -!- atslash has joined.
05:33:46 <esowiki> [[Talk:Eodermdrome]] https://esolangs.org/w/index.php?diff=59662&oldid=59497 * Salpynx * (+2398) answering my own question
05:49:29 -!- FreeFull has quit.
05:51:55 <esowiki> [[Talk:EsoScript]] N https://esolangs.org/w/index.php?oldid=59663 * Cortex * (+130) Created page with "== ... == This is literally just MediaWiki syntax. ~~~~"
06:26:20 -!- nfd9001 has joined.
06:31:23 -!- nfd9001 has quit (Ping timeout: 246 seconds).
06:31:35 <esowiki> [[Talk:Eodermdrome]] https://esolangs.org/w/index.php?diff=59664&oldid=59662 * Salpynx * (+534) /* Computational class / -completeness */
08:14:47 -!- arseniiv has joined.
09:00:40 -!- sebbu has quit (Read error: Connection reset by peer).
09:01:11 -!- sebbu has joined.
09:04:56 -!- Lord_of_Life has quit (Ping timeout: 268 seconds).
09:09:57 -!- Lord_of_Life has joined.
09:20:26 -!- heroux has quit (Ping timeout: 250 seconds).
09:21:28 -!- heroux has joined.
09:38:56 -!- AnotherTest has joined.
10:22:00 -!- wob_jonas has joined.
10:22:10 <wob_jonas> fizzie: are you trying to install something to the server that runs HackEso?
11:09:11 <fizzie> I was trying to update the bit that runs esolangs.org/logs, though it's the same server, sure.
11:12:35 <wob_jonas> I see
11:12:45 <wob_jonas> so it at least doesn't have to run inside the hackeso sandbox
11:13:28 <fizzie> Yes. Although that does share userland with... well, the intermediate layer of sandboxing.
11:16:18 <wob_jonas> sure, a large part of the file system is shared read only
11:17:28 <fizzie> HackEso really goes at least three levels deep -- there's the (I think KVM-based) Bytemark VPS, which runs a systemd-nspawn namespace-container-style "chroot on steroids" thing, which then has HackEso's UML kernel.
11:18:24 <wob_jonas> ah
11:18:32 <wob_jonas> right, that's why it said
11:18:51 <wob_jonas> ``` cat /proc/cpuinfo
11:18:52 <HackEso> processor: 0 \ vendor_id: User Mode Linux \ model name: UML \ mode: skas \ host: Linux dysnomia.zem.fi 4.9.0-5-amd64 #1 SMP Debian 4.9.65-3+deb9u2 (2018-01-04) x86_64 \ bogomips: 5134.74
11:19:01 <wob_jonas> it doesn't even tell what instruction set it supports anywhere
11:19:21 <wob_jonas> mind you, you can still find that out the intel way
11:19:27 <wob_jonas> it's just funny how sparse that cpuinfo is
11:19:37 <wob_jonas> I mean instruction set extensions
11:19:45 <wob_jonas> it does say x86_64 as the base cpu arch
11:20:45 <fizzie> Yeah. The middle level shares a kernel with the top level, while the bottom level shares the userland. Although those are both just Debian stable installations, so they're more or less identical.
11:22:21 <fizzie> s/shares the userland/& with the middle level/
11:22:42 <fizzie> That's also why it says "dysnomia.zem.fi" -- that's the hostname in the container, though it has no presence in the public DNS since it's not directly reachable. The VPS itself is called something else.
11:22:52 <wob_jonas> isn't the top level also running on a virtual machine by a hosting provider?
11:22:56 <fizzie> Yes.
11:23:04 <wob_jonas> so it's like four level deep at leat
11:23:04 <fizzie> Top level from my perspective, I guess.
11:23:24 <wob_jonas> at what level does the chat log script and chat http server run?
11:23:36 <wob_jonas> and the irc client
11:23:57 <wob_jonas> and the fetch HTTP client
11:25:02 <wob_jonas> is the network of the middle layer already restricted by namespaces, with holes only to tunnel the irc connection, the http interface for the version control contents, and fetch?
11:25:22 <fizzie> The "esobot" (the IRC client and log-writer) and "esologs" (the esolangs.org/logs web server) binaries both run on the top level together with MediaWiki, though not for any particular reason.
11:25:49 <wob_jonas> I see
11:25:53 <fizzie> HackEso's own machinery (so multibot, hackbot, the `fetch command and so) run on the middle one, and the `commands on the bottom one.
11:26:29 <wob_jonas> I see
11:27:50 <fizzie> And yes, the container networking is namespaced. Outgoing, it allows TCP to ports 80, 443 and 6665 to any host (for `fetch and the IRC connection) and UDP 53 to the VPS provider's DNS servers.
11:28:03 <wob_jonas> oh!
11:28:13 <wob_jonas> so they're not tunneled, those are just passed through
11:28:27 <wob_jonas> I thought they were connected to an external fetch server
11:28:50 <wob_jonas> and an external http server for the version control interface
11:29:17 <wob_jonas> and then the hg repository lives in the middle level?
11:31:24 <fizzie> There's a virtual ethernet interface that's a point-to-point link between interfaces 'host0' in the container, and 've-dysnomia' in the top level, and there's iptables NAT rules. So not really tunneled, no.
11:31:58 <fizzie> There's no networking out of the UML, but historically on HackEgo (when it was still working) that was based on a tunnel to a proxy server.
11:33:09 <fizzie> And yeah, the hg repository is on the middle level (and mounted read-only to the bottom), although the filesystem hierarchy of the middle level is accessible from the top.
11:33:24 <wob_jonas> yeah, so you can back it up from the top
11:33:33 <esowiki> [[2/9 of an esolang]] https://esolangs.org/w/index.php?diff=59665&oldid=59526 * Cortex * (+1)
11:33:40 <wob_jonas> how much of this was inherited from hackego?
11:34:20 <fizzie> In fact, I think I might have the hgweb server running on the top level directly, though that's also for no special reason.
11:35:41 <fizzie> HackEgo was broadly similar, but in place of the container setup it just had a conventional chroot.
11:36:00 <wob_jonas> I see
11:36:56 <wob_jonas> `? pcb
11:36:57 <HackEso> pcb? ¯\(°​_o)/¯
11:40:45 -!- Lord_of_Life_ has joined.
11:42:27 -!- Lord_of_Life has quit (Ping timeout: 240 seconds).
11:43:05 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
11:48:57 <wob_jonas> `datei
11:48:58 <HackEso> 2019-02-01 11:48:58.062828658 +0000 UTC February 1 Friday 2019-W05-5
12:34:03 -!- PaddyF has joined.
13:10:13 -!- PaddyF has left ("bbl").
13:34:09 -!- john_metcalf has joined.
13:39:41 -!- xkapastel has joined.
14:03:27 -!- atslash has quit (Ping timeout: 240 seconds).
14:07:30 -!- atslash has joined.
14:13:53 -!- atslash has quit (Ping timeout: 268 seconds).
14:14:22 -!- atslash has joined.
14:49:59 -!- Essadon has joined.
15:02:16 <wob_jonas> `? viking
15:02:17 <HackEso> viking? ¯\(°​_o)/¯
15:20:07 -!- Sgeo_ has joined.
15:23:02 -!- Sgeo has quit (Ping timeout: 246 seconds).
15:40:59 -!- sleepnap has joined.
16:15:42 -!- mniip has quit (Ping timeout: 604 seconds).
16:35:24 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
17:45:22 -!- dre has joined.
17:46:06 -!- dre has quit (Read error: Connection reset by peer).
18:42:25 -!- b_jonas has joined.
18:43:44 -!- b_jonas has set topic: Welcome to the international cloning factory for esoteric programming language discussion, design, development and deployment! | https://esolangs.org | logs: https://esolangs.org/logs/ http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf.
18:50:29 <fizzie> It's a shame other languages don't really have a Hoogle/Hayoo-like thing, at least as far as I know. Maybe it wouldn't work so well, but in Go I'm not so familiar with the standard(/quasistandard) libs and could easily look up some functions by type signature.
18:51:30 <b_jonas> how much "deployment" do we really do?
18:54:34 <orin> b_jonas: well fungot is arguably a "deployed" befunge progam
18:54:35 <fungot> orin: we are in the process of what i suppose the minister can, preferably, the termination, of his now even more counterproductive fiscal rule, but
18:55:06 -!- oerjan has joined.
19:01:51 <fizzie> b_jonas: Not much. Back when the cloud at cost system was still running, I was considering using it as a staging server for MediaWiki upgrades.
19:05:58 <fizzie> Other than that, I have a homegrown configuration pusher I use for all my systems (a git repo for configs, plus a bunch of Python to SSH in, verify that the live file matches the last-pushed revision and replace that with the latest) that also takes care of updating all the config files on both the top and middle levels. It's got a special case to use machinectl/nsenter to inject itself into the
19:06:04 <fizzie> container, so the container doesn't need to be SSH-reachable.
19:06:21 <fizzie> (The mounts are visible outside, but the UID/GID lookup is easier inside the container.)
19:07:56 <fizzie> It basically runs "ssh ... sudo [nsenter ...] python3 -" and feeds in a program that contains both the config file payload and the logic to insert it in place.
19:10:11 <fizzie> For code changes to esologs/esobot, there's just a shell script to do "bazel build -c opt //esologs //esobot" + rsync to put the binaries and the static files (.css, .js) in the right place. Didn't feel right putting those in the config pusher.
19:11:24 -!- mniip has joined.
19:11:28 <fizzie> I think that's about it. I'm always a little bit nervous when it comes time to update MediaWiki (it probably actually is way past time) because even though it's backed up, I haven't really ran any disaster recovery tests, and don't have a test environment to try on.
19:12:50 <fizzie> At least for the esologs/esobot binaries I can run them locally. There's actually a dozen lines of code in the esologs web server that if it's built without NDEBUG, it can serve the CSS/JS files for local testing. Normally that's done by nginx and it answers only for the dynamic parts.
19:22:28 <b_jonas> orin: yeah
19:23:43 <b_jonas> fizzie: I see
19:27:23 <oerjan> b_jonas: has IOCCC ended, since you removed it from the topic?
19:35:28 <b_jonas> oh darn
19:35:31 <b_jonas> no, it hasn't ended
19:35:32 <oerjan> someone should ask doctor rakethorn why he keeps not wearing shirts - although my obvious guess is that he's got orders to try to seduce agatha, but he could be a bit more subtle about it especially if he wants others not to interfere
19:35:41 -!- b_jonas has set topic: 2019 IOCCC ends March 15th -- http://www.ioccc.org/2019/rules.txt | Welcome to the international cloning factory for esoteric programming language discussion, design, development and deployment! | https://esolangs.org | logs: https://esolangs.org/logs/ http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf.
19:35:48 <b_jonas> thanks
19:35:59 -!- Phantom_Hoover has joined.
19:35:59 -!- Phantom_Hoover has quit (Changing host).
19:36:00 -!- Phantom_Hoover has joined.
19:48:32 <oerjan> well, "guess", i think his conversation with wooster pretty much implied it.
19:55:47 <b_jonas> `? liver
19:55:48 <HackEso> liver? ¯\(°​_o)/¯
19:55:49 <b_jonas> `? livery
19:55:50 <HackEso> livery? ¯\(°​_o)/¯
19:57:23 -!- LKoen has joined.
20:27:25 -!- leonheart has quit (Ping timeout: 256 seconds).
20:34:23 <esowiki> [[Talk:Eodermdrome]] https://esolangs.org/w/index.php?diff=59666&oldid=59664 * Oerjan * (+416) Not that simple
20:57:36 -!- Phantom_Hoover has quit (Remote host closed the connection).
20:59:33 -!- leonheart has joined.
21:02:17 -!- Phantom_Hoover has joined.
21:02:17 -!- Phantom_Hoover has quit (Changing host).
21:02:17 -!- Phantom_Hoover has joined.
21:03:23 -!- FreeFull has joined.
21:03:59 -!- LKoen has quit (Remote host closed the connection).
21:08:35 -!- FreeFull has quit (Quit: kernel update).
21:09:26 -!- FreeFull has joined.
21:10:11 -!- LKoen has joined.
21:17:57 <oerjan> @metar ENVA
21:17:59 <lambdabot> ENVA 012050Z 09009KT CAVOK M04/M10 Q1013 RMK WIND 670FT 12009KT
21:32:01 -!- leonheart has quit (Ping timeout: 256 seconds).
21:37:29 -!- leonheart has joined.
21:47:19 -!- leonheart has quit (Ping timeout: 256 seconds).
21:51:27 -!- arseniiv has quit (Read error: Connection reset by peer).
21:51:27 -!- arseniiv has joined.
21:59:28 <fizzie> @metar EGLL
21:59:28 <lambdabot> EGLL 012150Z AUTO 36007KT 320V030 5000 -RASN BKN010 OVC029 02/01 Q0995
21:59:31 <fizzie> Wet.
22:01:43 <shachaf> @metar KOAK
22:01:43 <lambdabot> KOAK 012153Z 14016KT 10SM SCT046 BKN055 OVC100 15/09 A2978 RMK AO2 RAB2058E19 SLP084 P0000 T01500089
22:03:58 <oerjan> ...I can't claim much better on that :P
22:04:05 <oerjan> wong wrindow
22:26:59 -!- b_jonas has quit (Quit: leaving).
22:39:51 <kmc> wish i was cool enough to be able to read metar
22:42:31 <Phantom_Hoover> which of these numbers means 'fucking cold'
22:43:30 <fizzie> The temperature (in celsius) is the first half of the M04/M10, 02/01, 15/09 pair.
22:44:21 <kmc> so it's 2°C at EGLL and 15°C at KOAK?
22:44:42 <fizzie> Yes. And -4°C at ENVA.
22:44:50 <kmc> ah, M = minus
22:44:55 <kmc> got it
22:45:05 <kmc> what's the second number?
22:45:16 <fizzie> That's the dew point.
22:45:25 <fizzie> When it's M00/M00, that's called "cow temperatures".
22:45:31 <fizzie> Because that's how a cow goes.
22:45:48 <fizzie> (Not official METAR jargon.)
22:45:53 <oerjan> or swims, in that case
22:46:29 -!- LKoen has quit (Quit: “It’s only logical. First you learn to talk, then you learn to think. Too bad it’s not the other way round.”).
22:46:36 <oerjan> actually, more like gets stuck in ice, i guess
22:46:58 <fizzie> The 012153Z part is a partial timestamp in DDHHMMZ format, so the report was 21:53 UTC on the 1st day of some unspecified month.
22:48:34 <fizzie> And the 14016KT is wind direction 140°, 16 knots. Sometimes it has extra stuff, like "...10G20KT" is 10 knots, gusts of 20, or "VRB" in place of a direction.
22:52:08 <fizzie> 10SM and 5000 are visibilities, in statute miles and metres respectively. SCT046, BKN055 OVC100 are cloud covers (scattered, broken, overcast) with heights in 100s of feet. A2987 and Q0995 are the air pressure in some units (mmHg and hPa?), and that's pretty much all I remember.
22:52:10 -!- leonheart has joined.
22:52:40 <fizzie> Oh, and precipitation is various codes, so -RASN is weak (-) rain (RA) and snow (SN).
22:53:53 <fizzie> I think I've only seen DZ, RA, SN, GR, BR, FG and HZ (drizzle, rain, snow, hail, mist, fog, haze), but there's a lot more.
22:56:27 -!- AnotherTest has quit (Ping timeout: 264 seconds).
23:01:10 <shachaf> what about FU
23:01:23 <shachaf> definitely had that one here tdnh
23:05:43 <fizzie> Smoke, apparently. Haven't seen that here.
23:05:47 <fizzie> Also no code for drones.
23:07:58 <FireFly> They might have to add one soon
23:08:24 <FireFly> how else would UK airports report the condition at the runway accurately...
23:08:43 <Phantom_Hoover> @metar EGPH
23:08:43 <lambdabot> EGPH 012250Z VRB02KT 9999 FEW022 BKN038 01/00 Q1002
23:08:58 <Phantom_Hoover> not even freezing
23:09:57 <fizzie> Does the "PH" there mean Phantom_Hoover? Is it your own airport?
23:10:03 -!- leonheart has quit (Ping timeout: 256 seconds).
23:10:15 <Phantom_Hoover> literally yes
23:10:18 <Phantom_Hoover> it's edinburgh airport
23:15:32 <oerjan> `icao EGPH
23:15:33 <HackEso> Edinburgh (EDI, EGPH)
23:16:46 <Phantom_Hoover> i don't actually live in edinburgh any more but i don't know the icao for glasgow airport
23:16:56 <Phantom_Hoover> or... anywhere except edinburgh really
23:16:57 <fizzie> `airport glasgow
23:16:58 <HackEso> Glasgow (GLA, EGPF) \ Wokal Field Glasgow International Airport (GGW, KGGW) \ Glasgow Railway Station (?, ?) \ Glasgow Buchanan Bus Station (?, ?) \ Glasgow City Heliport (?, EGEG) \ Glasgow Industrial (?, 07MT)
23:17:04 <Phantom_Hoover> khaf and ksfo i suppose
23:17:08 <fizzie> Very close, lexicographically.
23:17:15 <Phantom_Hoover> @metar egeg
23:17:15 <lambdabot> No result.
23:17:20 <Phantom_Hoover> damn
23:17:25 <Phantom_Hoover> @metar egpf
23:17:26 <lambdabot> EGPF 012250Z AUTO 28007KT 9999 NCD 01/M03 Q1004
23:18:03 <fizzie> @metar EGLC
23:18:03 <lambdabot> EGLC 012250Z AUTO 34005KT 280V010 8000 -RA SCT010 BKN018 OVC024 02/01 Q0995
23:18:16 <fizzie> I flew from that place the other day for the first time.
23:18:24 -!- leonheart has joined.
23:18:32 -!- leonheart has quit (Client Quit).
23:19:01 <Phantom_Hoover> `icao eglc
23:19:01 <HackEso> City (LCY, EGLC)
23:19:12 <Phantom_Hoover> just city lol
23:22:08 <fizzie> `fetch share/airports.dat https://raw.githubusercontent.com/jpatokal/openflights/master/data/airports-extended.dat
23:22:09 <HackEso> 2019-02-01 23:22:09 URL:https://raw.githubusercontent.com/jpatokal/openflights/master/data/airports-extended.dat [1632638/1632638] -> "share/airports.dat" [1]
23:22:14 <fizzie> `icao eglc
23:22:15 <HackEso> London City Airport (LCY, EGLC)
23:22:18 <fizzie> That's bette.
23:22:24 <fizzie> ...r.
23:22:54 <fizzie> (airports-extended.dat is the one with also train stations and ferry terminals, so maybe the plain airports.dat would've been better.)
23:38:25 -!- Phantom_Hoover has quit (Remote host closed the connection).
23:41:36 -!- Lord_of_Life_ has joined.
23:42:58 -!- Lord_of_Life has quit (Ping timeout: 245 seconds).
23:43:02 <kmc> `icao ksfo
23:43:03 <HackEso> San Francisco International Airport (SFO, KSFO)
23:43:28 <kmc> usa gets the best icao codes cause we're the best
23:43:41 <kmc> that's also why our phone country code is 1
23:43:57 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
23:44:41 <fizzie> Canada has the same deal when it comes to ICAO codes, it's just XXX <-> CXXX.
23:45:20 <fizzie> And aren't they in the +1 area too?
23:45:23 <kmc> yeah
23:45:28 <kmc> but their 3 letter codes are gibberish
23:45:30 <kmc> YYZ, give me a break
23:48:26 -!- arseniiv has quit (Ping timeout: 246 seconds).
23:49:02 <oerjan> `icao CYYZ
23:49:03 <HackEso> Lester B. Pearson International Airport (YYZ, CYYZ)
23:49:33 <fizzie> Maybe that should print the city as well.
23:49:44 <fizzie> When it's not a substring of the airport name, say.
23:50:05 <fizzie> `` grep CYYZ share/airports.dat
23:50:07 <HackEso> 193,"Lester B. Pearson International Airport","Toronto","Canada","YYZ","CYYZ",43.6772003174,-79.63059997559999,569,-5,"A","America/Toronto","airport","OurAirports"
23:51:23 <kmc> it has train stations too? how does that work?
23:51:53 <fizzie> That's just what they said on the download page.
23:52:16 <fizzie> `` grep Euston.Railway share/airports.dat
23:52:17 <HackEso> 7393,"London Euston Railway Station","London","United Kingdom",\N,\N,51.5284,-0.1331,0,0,"U","Europe/London","station","User"
23:52:36 <fizzie> Nulls (with the \N CSV encoding) on the ICAO and IATA columns, I guess.
23:53:16 <fizzie> (I don't think it has more than a small fraction of railway stations.)
23:53:23 <kmc> some US train stations have codes
23:53:27 <kmc> `icao kzfv
23:53:28 <HackEso> No output.
23:53:36 <kmc> `iata zfv
23:53:36 <HackEso> Philadelphia 30th St Station (ZFV, ?)
23:54:07 <kmc> `iata zep
23:54:08 <HackEso> No output.
23:54:40 <kmc> `iata zyp
23:54:41 <HackEso> Penn Station (ZYP, ?)
23:54:58 <fizzie> `iata LON
23:54:59 <HackEso> All Airports (LON, ?)
23:55:04 <fizzie> Best name.
23:55:28 <fizzie> The internet alleges that there's actually an ICAO code ("EGGO") for the London Metropolitan Area.
23:56:35 <kmc> some airlines in the US will sell you a ticket which includes a rail connection
←2019-01-31 2019-02-01 2019-02-02→ ↑2019 ↑all