| --- Log opened Wed Sep 24 00:00:02 2014 |
| 00:11 | | * McMartin has his first run-ins with the Apache C libraries and specifically APR. |
| 00:11 | <&McMartin> | I am actually pleasantly surprised |
| 00:39 | | Derakon[AFK] is now known as Derakon |
| 00:45 | <&ToxicFrog> | Argh |
| 00:45 | <&ToxicFrog> | Fuck HTML in the eye |
| 00:46 | <@macdjord> | <fuckthisshit oriface="eye" /> |
| 00:47 | <&ToxicFrog> | All I want is <button><input type="text"><button>, with the input field taking up whatever space isn't occupied by the two buttons |
| 00:47 | <&ToxicFrog> | I cannot find any way of doing this |
| 00:49 | <&Derakon> | You may need to use CSS. |
| 00:49 | <&Derakon> | http://stackoverflow.com/questions/1260122/expand-div-to-take-remaining-width might help. |
| 00:54 | <&ToxicFrog> | This looks closer: http://stackoverflow.com/questions/10244927/input-text-auto-width-filling-100-wi th-other-elements-floating |
| 01:19 | | * Derakon arghs at his laptops. |
| 01:20 | <&Derakon> | Got the fucking $30 Lightning cable to connect them, hooked them up, ran Migration Assistant, it detected the old laptop from the new one, tried to copy my account across, it failed because of inadequate disk space on the new computer. |
| 01:20 | <&Derakon> | Try to repeat (because I know there's a way to limit what Migration Assistant copies to keep it under the limit), the new laptop now no longer detects the old one. |
| 01:20 | <&Derakon> | Er, Thunderbolt, not Lightning. Stupid nonstandard connectors. |
| 01:38 | <&ToxicFrog> | So apparently you have to: |
| 01:38 | <&ToxicFrog> | - place the left button with float: left |
| 01:39 | <&ToxicFrog> | - place the right button with float: right |
| 01:39 | <&ToxicFrog> | - only after doing both of those, place the textbox wrapped in a <span> |
| 01:41 | <@RchrdB> | McMartin, I found a lack of documentation other than the header files themselves,but⦠|
| 01:41 | <@RchrdB> | â¦the header files themselves were really nice. |
| 01:43 | | * Derakon makes a new admin account on the new lappy, deletes the old one, and now can find the old laptop in Migration Assistant...but the damned thing says "this user's files don't need to be transferred" when he tries to transfer his old lappy's account over. |
| 01:43 | <&Derakon> | Goddammit, Apple, just let me do the thing. |
| 01:43 | <&Derakon> | Stop trying to be smarter than me! |
| 01:43 | <@RchrdB> | Also the functions all have the same uniform calling convention: everything returns an apr_status_t, and other values may be returned through out pointers. |
| 01:43 | <&ToxicFrog> | Seriously, even the really shit UI toolkits can do better than that. |
| 01:45 | <&ToxicFrog> | Derakon: er, isn't "We're Smarter Than Youâ˘" the entire design philosophy/selling point of Apple software? |
| 01:45 | <@RchrdB> | McMartin, the one thing that I really didn't understand about APR is that while a) it's meant to be the foundation for Apache2, b) it has really good support for event-triggered network programming (like libuv or libev)... |
| 01:45 | <&Derakon> | TF: well, it used to be "we're simple and we work" but apparently that's changed. |
| 01:46 | <@RchrdB> | ...but Apache 2 doesn't (much) do event-triggered networking. |
| 01:49 | <&McMartin> | Heh |
| 01:49 | <&McMartin> | Yeah, actually, most of the stuff I'm auditing here is using their encoder/decoder/string/pool allocation stuff, which is less apr_status_t and more "a libc that doesn't suck" |
| 01:49 | <@RchrdB> | to be fair, I think Apache did get an event-triggered MPM eventually, though I'm not sure if it's still marked experimental or not. |
| 01:50 | <&McMartin> | And I got man-page level goodness just by googling the function names. |
| 01:50 | <&McMartin> | So maybe the docs have improved a little? |
| 01:50 | <@RchrdB> | I might have just been looking in the wrong places. |
| 01:50 | <&McMartin> | See, for instance: http://apr.apache.org/docs/apr-util/0.9/group__APR__Util__Base64.html |
| 01:50 | <@RchrdB> | I *was* able to write some things that successfully put bytes onto TCP sockets just with my "read the .h files" approach |
| 01:51 | <&McMartin> | Yeah |
| 01:51 | <@RchrdB> | ugh no that's javadoc shit |
| 01:51 | <@RchrdB> | I'd rather read the .h files. |
| 01:51 | <&McMartin> | I'm doing a code audit so that is the part I needed~ |
| 01:51 | <&McMartin> | (Not sure why it thinks decode_binary involved EBCDIC though) |
| 01:51 | | * Derakon eyes http://www.hardturm.ch/luz/2014/02/fixing-os-x-10-9-mavericks-migration-from-ext ernal-volume/ |
| 01:51 | <&Derakon> | My fucking god, how can Apple be this incompetent. |
| 01:52 | <&Derakon> | Step 1: mount old computer in "target disk mode", which basically makes its hard drive into an "external hard drive" for another computer. |
| 01:52 | <@RchrdB> | hey, EBCDIC probably counts as a secret code by now because nobody will remember it |
| 01:52 | <&Derakon> | Step 2: run Migration Assistant. |
| 01:52 | <@RchrdB> | I think that the docs you point out there are generated from the .h files and have the same text in them. |
| 01:52 | <&Derakon> | Step 3: Migration Assistant logs out all user accounts (because they might be replaced by migration) and unmounts all attached drives. |
| 01:52 | <&McMartin> | RchrdB: Right, but ANAICT that's actually "base64_decode something that might have nulls in it and isn't an ascii, null-terminated string) |
| 01:52 | <&McMartin> | s/)/"/ |
| 01:52 | <&Derakon> | Step 4: where's the drive you want to migrate from? Who fucking knows! |
| 01:53 | <&McMartin> | (It's what you call to get a usable version of, say, a public key) |
| 01:56 | | * Derakon sighs, apparently has to drag his third computer into the mix to re-mount the old laptop via SSH after starting Migration Assistant on the new laptop. |
| 01:56 | <&Derakon> | This is deplorable. |
| 01:56 | <&Derakon> | Also really not what I wanted to be spending my evening on. |
| 01:57 | <&McMartin> | Jesus H. |
| 02:02 | <&Derakon> | Right, now it finally sees the drive and account on it...let's see if it'll figure out there's not enough space, and let me customize what to copy. |
| 02:08 | <&ToxicFrog> | Hmm. Object search feature. |
| 02:08 | <&ToxicFrog> | Sort by level first, or by object type? |
| 02:08 | <&Derakon> | Is this System Shock? |
| 02:09 | <&ToxicFrog> | Yes. |
| 02:09 | <&Derakon> | I'd say level first -- "I want to find the <X> on this level" makes more sense to me than "I want to find all <X> in the game". |
| 02:09 | <&Derakon> | But I've never played SS so I might have the wrong perspective. |
| 02:09 | <&ToxicFrog> | Right, but there's a "search this level only" button thatdoes that |
| 02:09 | <&ToxicFrog> | This is for when you click "search all maps" |
| 02:09 | <&Derakon> | Oh? Okay, then *shrug* |
| 02:14 | <@macdjord> | ToxicFrog: Item type. |
| 02:20 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Connection closed] |
| 02:53 | | Turaiel[Offline] is now known as Turaiel |
| 03:02 | <&ToxicFrog> | Hmm. Problem: searching for items in containers gets you the blueroom location rather than the container location. |
| 03:02 | <&ToxicFrog> | I'll have to fix that before this can really be considered production-ready. |
| 03:11 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 03:11 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 04:03 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 04:12 | | Alek [omegaboot@Nightstar-c8t.a00.36.73.IP] has quit [Connection reset by peer] |
| 04:12 | | Alek [omegaboot@Nightstar-c8t.a00.36.73.IP] has joined #code |
| 04:12 | | mode/#code [+o Alek] by ChanServ |
| 04:12 | | HotShot [fake@Nightstar-95455m.sfldmi.sbcglobal.net] has quit [[NS] Quit: Trespassers will be shot, Survivers will be shot again! [Time wasted on Mirc 6hrs 36mins 4secs]] |
| 04:36 | <&ToxicFrog> | It's alive: http://funkyhorror.ancilla.ca/toxicfrog/ss1/maps/map.html |
| 04:36 | <&ToxicFrog> | Try searching for something :D |
| 04:36 | <@macdjord> | > love |
| 04:39 | <@Reiv> | ToxicFrog: So, why is it that complex level design took such a hit? |
| 04:41 | | RchrdB [0x2ba22e11@Nightstar-qe9.aug.187.81.IP] has quit [[NS] Quit: Leaving] |
| 04:41 | <&ToxicFrog> | Reiv: ? |
| 04:42 | <&Derakon> | Reiv: because those complex levels are composed of very simple building blocks. |
| 04:42 | <&Derakon> | As you start upping the detail of what's in a level, you have to drastically cut down on something to compensate, and level complexity is what went. |
| 04:42 | <&Derakon> | (Not really surprising, since lots of people got lost in these old maps) |
| 04:43 | <&ToxicFrog> | Derakon: I note that SS1 also had a metric shitload of detail objects, in many places rivaling modern games for object density (though not, obviously, 3d model complexity etc); click "show all" on one of the floors and try to make any sense of the result. |
| 04:43 | <@Alek> | TF: it'd be nice to have a legend by the selection list of what color each item corresponds to. |
| 04:43 | <@Alek> | just my $.02 |
| 04:44 | <&ToxicFrog> | Thief did the same, and in fact they spent a fair amount of effort writing the renderer so that it could handle levels with significantly more incidental objects than other games of the time. |
| 04:44 | <&ToxicFrog> | Alek: I should probably add that, yeah. |
| 04:46 | <&ToxicFrog> | Derakon: also, in terms of overall stuff in a level, I don't think that's really gone down; it's just been unrolled. |
| 04:47 | <&ToxicFrog> | Rather than getting, say, 64x64 square meters of interconnected space to explore, you get 8x512 square meters of alternating arenas and cutscenes. |
| 04:48 | | * Derakon nods. |
| 04:48 | <&ToxicFrog> | I actually blame Half-Life for that; it introduced (or at least popularized) linear levels with storytelling via in-engine cutscenes (rather than during intermissions, via found records, or not at all) |
| 04:49 | <&Derakon> | And yeah, I wasn't speaking so much to "unique object density" as to "detail density", which incorporates the complexity of the models. |
| 04:49 | <&ToxicFrog> | The trick there being that if the level is a big explorable space, you have much less control over how the player explores it, which -- from a game designer perspective -- makes your job harder. |
| 04:50 | <&ToxicFrog> | Valve put a lot of effort into making sure that the player encountered enemies, weapons, game mechanics, and storyline events in a specific and carefully tuned order. |
| 04:50 | <&ToxicFrog> | Which is not, in and of itself, a bad thing, but then absolutely fucking everyone copied them, and generally not as well, either. |
| 04:51 | <&Derakon> | And again, this makes sense when you're trying to sell a disposable "experience" -- you have a lot more control over the events when everything's linear and scripted. |
| 04:51 | <&ToxicFrog> | Derakon: I don't think individual model complexity is really relevant here, though -- the level designers aren't the same people making the models. |
| 04:51 | <&Derakon> | But such games don't generally get as much replay as the more open-ended ones. |
| 04:52 | <&Derakon> | TF: mm, but I can't help feeling that more detailed props start standing out more as they get re-used. |
| 04:53 | <&Derakon> | If you have a lovingly-crafted vase with a lovingly-crafted crack in it, and you start seeing that vase everywhere, it looks kind of odd. |
| 04:53 | <&Derakon> | But a chunky, more "symbolic" than "representational" vase can be thrown down anywhere and everywhere without the player blinking. |
| 04:53 | <&ToxicFrog> | ...true, but again, I don't think that has anything to do with how the level is interconnected, just with how you decorate it. |
| 04:53 | <&Derakon> | I'm getting to that. |
| 04:54 | <&ToxicFrog> | It will look just as odd repeated in five successive rooms of a corridor shooter as it will appearing in every dining hall of a Thief level. |
| 04:54 | <&Derakon> | Since each prop becomes harder to re-use, you either have to make more props, accept less densely-decorated scenes, or have fewer scenes / less explorable area. |
| 04:55 | <&ToxicFrog> | (also, this is not universally true. FEAR, for example, got a shitload of mileage out of a set of office furniture props.) |
| 04:55 | <&ToxicFrog> | (and was also completely and inescapably linear) |
| 04:55 | <&Derakon> | The one constant of blanket statemets is that they have exceptions~ |
| 04:55 | <&Derakon> | Er, statements. |
| 04:55 | <&Derakon> | Anyway, I am fast running out of brain, so T-2. |
| 04:56 | <&ToxicFrog> | Likewise. 'night |
| 04:58 | | Derakon is now known as Derakon[AFK] |
| 05:06 | | * Vornicus makes a blanket statement: he's about to be under one. |
| 05:09 | | Kindamoody[zZz] is now known as Kindamoody |
| 05:18 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 05:56 | | Turaiel is now known as Turaiel[Offline] |
| 07:00 | <@Azash> | McMartin: Do you know any good IF engines in python? |
| 07:07 | <&McMartin> | Heh. Does Ren'Py count? |
| 07:10 | <@Azash> | Considering how much Shiz heckles it, probably not :P |
| 07:12 | <@Azash> | Well, if the python requirement is dropped, what are some good ones? |
| 07:32 | <&McMartin> | Inform 7 and TADS 3 are the two I'd consider the best at present for parser-related IF; Twine seems to have the most currency for hypertext IF, but I've liked what I've seen of Undum. |
| 07:34 | <@Azash> | Much appreciated |
| 07:50 | | celticminstrel is now known as celmin|sleep |
| 07:51 | <&McMartin> | I7 is the one I've done all my own work in. |
| 07:52 | <&McMartin> | Well, also its predecessor, I6, which is a more traditional but also much more limited language. |
| 08:07 | <@Alek> | yey, FEAR |
| 08:13 | | Kindamoody is now known as Kindamoody|afk |
| 09:12 | <@Tarinaky> | McMartin: how come all the scripts for I7 are written in I6? |
| 09:13 | <@Tarinaky> | (I experimented with Inform... before working out that while it was easy to use it did nothing to stem the fact that I can't actually write good Fiction) |
| 09:14 | <@Tarinaky> | (So I mostly just implemented shitty game mechanics in I6) |
| 09:41 | | thalass [thalass@Nightstar-nefo3m.bigpond.net.au] has quit [Ping timeout: 121 seconds] |
| 10:10 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 10:13 | | thalass [thalass@Nightstar-nefo3m.bigpond.net.au] has joined #code |
| 10:13 | | mode/#code [+o thalass] by ChanServ |
| 11:01 | <@Tarinaky> | I think someone suggested a way of doing (easy) graphing with Numpy |
| 11:01 | <@Tarinaky> | But I can't find it in my logs. |
| 11:01 | <@Tarinaky> | +yesterday |
| 11:03 | <@Tarinaky> | Alternatively... does anyone here know enough about any of the GNU Numerical Analysis interpretters out there to get me started? |
| 11:03 | <@Tarinaky> | I'm pretty sure I'm in a Matlab shop, and I really can't justify a license. |
| 11:04 | <@Tarinaky> | If it matters: my current problem domain is Audio DSP. |
| 11:04 | <@Tarinaky> | So anything that makes that 'easy' is good. |
| 11:22 | | macdjord [macdjord@Nightstar-7rac1r.mc.videotron.ca] has quit [[NS] Quit: Reboot] |
| 11:35 | | AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has joined #code |
| 11:39 | | macdjord [macdjord@Nightstar-7rac1r.mc.videotron.ca] has joined #code |
| 11:39 | | mode/#code [+o macdjord] by ChanServ |
| 12:08 | | AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has quit [[NS] Quit: Leaving] |
| 12:14 | <@Tarinaky> | Octave seems pretty good. |
| 12:15 | <@Tarinaky> | Pretty much just worked out of the box I think. |
| 12:43 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 12:43 | | mode/#code [+o Checkmate] by ChanServ |
| 12:50 | | thalass [thalass@Nightstar-nefo3m.bigpond.net.au] has quit [Ping timeout: 121 seconds] |
| 13:39 | < Xon> | ah todays xkcd is so true; http://xkcd.com/1425/ |
| 14:22 | <@Tarinaky> | Argh at Ruby |
| 14:25 | <@Tarinaky> | Apparently I can't XOR booleans |
| 14:25 | <@Tarinaky> | Because that would be SENSIBLE |
| 14:26 | <@TheWatcher> | wat |
| 14:27 | <@Tarinaky> | Or maybe that's not the issue. |
| 14:27 | <@Tarinaky> | I don't know. |
| 14:27 | <@Tarinaky> | I'm trying to implement a Maximum Length Sequence. |
| 14:28 | <@Tarinaky> | Using an Enumerator (the Ruby equiv to Python's Generators I believe) |
| 14:28 | <@Tarinaky> | I have an Array which holds my registers. |
| 14:28 | <@Tarinaky> | I shift each register left with drop(1) |
| 14:29 | <@Tarinaky> | And then I append (using the << operator) register[0] ^ register[1] |
| 14:29 | <@Tarinaky> | Or I would~ |
| 14:29 | <@Tarinaky> | NoMethodError: undefined method `^' for [1, 1, 1, 1, 1]:Array |
| 14:30 | <@Tarinaky> | Oh hang on. |
| 14:30 | <@Tarinaky> | I think my first guess was right - operator precedence issue. |
| 14:32 | <@Tarinaky> | Yeah. Herp. |
| 14:32 | <@Tarinaky> | It was evaluating '<<' first. |
| 14:34 | <@Tarinaky> | Okay... now it isn't doing XOR properly :( |
| 14:34 | | Alek [omegaboot@Nightstar-c8t.a00.36.73.IP] has quit [Ping timeout: 121 seconds] |
| 14:36 | <@Tarinaky> | Oh... |
| 14:37 | <@Tarinaky> | Ahah! |
| 14:37 | | Alek [omegaboot@Nightstar-c8t.a00.36.73.IP] has joined #code |
| 14:37 | | mode/#code [+o Alek] by ChanServ |
| 15:01 | | thalass [thalass@Nightstar-su58ca.bigpond.net.au] has joined #code |
| 15:01 | | mode/#code [+o thalass] by ChanServ |
| 15:27 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 15:33 | | * Julius unhappily works on rejiggering someone's Windows, after an unknown party made a factory reset of their Win8. |
| 15:33 | < Julius> | Their documents are still there, but settings and installed programs were wiped. |
| 16:11 | | RchrdB [0x2ba22e11@Nightstar-952.jvr.168.194.IP] has joined #code |
| 16:11 | | mode/#code [+o RchrdB] by ChanServ |
| 16:17 | | Checkmate [Z@Nightstar-ro94ms.balk.dk] has joined #code |
| 16:17 | | mode/#code [+o Checkmate] by ChanServ |
| 16:37 | | celmin|sleep is now known as celticminstrel |
| 16:41 | | Orthia [orthianz@Nightstar-jpq.2ff.224.119.IP] has quit [Ping timeout: 121 seconds] |
| 16:42 | | Orthia [orthianz@Nightstar-jpq.2ff.224.119.IP] has joined #code |
| 16:42 | | mode/#code [+o Orthia] by ChanServ |
| 16:46 | <&McMartin> | Tarinaky: I7's compiler generates I6 instead of Z-Code or Glulx code directly. Is that what you're asking? If so, it does so because that means the Z and backend work has already been done like 20 years ago |
| 16:46 | <&McMartin> | I7 extensions are generally *not* written in I6 unless they're exposing VM services |
| 17:19 | | Checkmate [Z@Nightstar-ro94ms.balk.dk] has quit [Ping timeout: 121 seconds] |
| 17:22 | | macdjord is now known as macdjord|slep |
| 17:38 | <@Azash> | https://i.imgur.com/BTKF8Hb.jpg |
| 17:42 | <&McMartin> | For after you tested that new iWave functionality |
| 17:42 | <@Tarinaky> | McMartin: I don't know if the situation has changed, but the code examples I saw for various hing like variables and such were all i6 |
| 17:42 | <@Azash> | McMartin: Haven't heard of bendghazi yet? |
| 17:42 | <@Azash> | McMartin: http://www.dailydot.com/technology/iphone-6-plus-bendgate/ |
| 17:42 | <&McMartin> | Tarinaky: I've been using the system since its initial public beta and I literally have no idea what you are talking about |
| 17:43 | <@Tarinaky> | Okay. |
| 17:43 | <@Tarinaky> | I made it up then. |
| 17:43 | | * Julius scratches his head. |
| 17:43 | <&McMartin> | The closest I can guess is that it used to use the I6 library and people would drop to it because they knew how to do it that way |
| 17:43 | <&McMartin> | As per things like |
| 17:43 | < Julius> | I've solved this problem before. |
| 17:43 | < Julius> | But I don't remember how. |
| 17:43 | <&McMartin> | Include (- has pluralname; -) when defining the whatever. |
| 17:44 | < Julius> | I've got two PCs and a router between, and both can ping the router, but one can't ping the other. |
| 17:44 | < Julius> | But the other can ping the first one. |
| 17:44 | <&McMartin> | Which would now be handled via "The whatever is plural-named", and at release required including a plurality extension and saying things like "the pile of marbles act plural" |
| 18:49 | <&McMartin> | http://seclists.org/oss-sec/2014/q3/649 |
| 18:52 | <&ToxicFrog> | Hooooooly shit. |
| 18:53 | <&McMartin> | http://seclists.org/oss-sec/2014/q3/650 is the vuln disclosed in detail |
| 18:55 | <&ToxicFrog> | I'm not pro at sshd, but AIUI exploiting this through sshd requires you to actually be able to authenticate, yes? |
| 18:55 | <&McMartin> | I'm not totally clear on that, actually. |
| 18:56 | <&McMartin> | s/650/651/ for a test you can run, I guess? |
| 19:00 | <&ToxicFrog> | Not only does it require authentication, it looks like my default configuration doesn't set SSH_ORIGINAL_COMMAND. |
| 19:01 | <&McMartin> | It does seem to require authentication. I get a bad result though not a code execution. |
| 19:03 | <&McMartin> | With the command "ssh 127.0.0.1 '() { ignored; }; /usr/bin/id'" |
| 19:03 | <&McMartin> | I get prompted for my password, and on success get: |
| 19:03 | <&McMartin> | bash: -c: line 0: syntax error near unexpected token `)' |
| 19:03 | <&McMartin> | bash: -c: line 0: `() { ignored; }; /usr/bin/id' |
| 19:03 | <&McMartin> | That seems like it is trying to execute it but doesn't actually know how to to do functions in environment variables. |
| 19:05 | <&ToxicFrog> | Well |
| 19:05 | <&ToxicFrog> | What normally happens is that SSH_ORIGINAL_COMMAND gets set to '() { ignored; }; /usr/bin/id' and exported to the child bash process |
| 19:06 | <&ToxicFrog> | bash, while reading its environment variables, sees 'oh, this one starts with "() {"' when it gets to SSH_ORIGINAL_COMMAND and executes it. |
| 19:06 | <&ToxicFrog> | Things go wrong. |
| 19:06 | <&ToxicFrog> | What is happening on my ssh -- and yours, it looks like -- is that SSH_ORIGINAL_COMMAND is not getting set by sshd. |
| 19:06 | <&ToxicFrog> | So bash never sees the bad variable definition and goes on to try to execute the actual command passed to it by sshd, which is not valid bash, resulting in the error you see. |
| 19:07 | <&McMartin> | Aha |
| 19:08 | <&McMartin> | Hopefully this is because that is the default and not because some white hat ran in and injected some code to tighten up my config~ |
| 19:08 | <&ToxicFrog> | Heh |
| 19:09 | <&ToxicFrog> | Oh I see |
| 19:09 | <&ToxicFrog> | I didn't understand what "ssh forced commands" are |
| 19:10 | <&ToxicFrog> | You can set a thing server-side to restrict a certain key to only running a certain command |
| 19:10 | <&ToxicFrog> | So that do, say, ssh monitor@myserver -i uptimemon.pub, and on server, ~monitor/.ssh/authorized_keys forces it to run 'uptime' rather than the login shell. |
| 19:11 | <&ToxicFrog> | That is the forced command. |
| 19:11 | <&ToxicFrog> | Iff a forced command is configured, SSH_ORIGINAL_COMMAND is set, so that the forced command can see what the user was trying to do. |
| 19:16 | <&McMartin> | OK, that is scary, but less "I need to take my machine off the network for awhile" scary. |
| 19:24 | | Kindamoody|afk is now known as Kindamoody |
| 19:26 | <@RchrdB> | oh fuck me that is scary |
| 19:26 | <@RchrdB> | so if you set a bash script as an ssh forced command, you get boned? |
| 19:27 | <&McMartin> | And if the attacker has the key to run the forced command. |
| 19:27 | <&McMartin> | Or if you are running bash CGI scripts (spoiler: don't do that) |
| 19:27 | <&ToxicFrog> | Yeah, the attacker has to actually be able to authenticate. |
| 19:31 | <@RchrdB> | "the attacker has to actually be able to authenticate" is entirely realistic for, say, public Git hosting services that permit the SSH interface. |
| 19:33 | <&ToxicFrog> | This is true. But it's not generally an issue for people running sshd at home for personal use. |
| 19:34 | <@RchrdB> | no |
| 19:34 | <@RchrdB> | major turd-in-trousers moment for github, gitorious, bitbucket, etc. |
| 19:45 | | EvilDarkLord [jjlehto3@Nightstar-evu5hu.org.aalto.fi] has joined #code |
| 19:47 | < EvilDarkLord> | Hey. Does someone know what the configuration option in Firefox is to make the browser try for more than ~20 seconds before giving up on getting to a webpage? I am under poor network conditions but I have pages to load that I'm not in a hurry with. |
| 19:47 | < EvilDarkLord> | (This is also the reason I'm not googling this for myself right now) |
| 20:19 | < [R]> | Http.response.timeout |
| 20:21 | < [R]> | Dom.script_max_run_time |
| 20:21 | <@RchrdB> | dom.script.max_run_time doesn't affect downloads, does it? |
| 20:21 | < [R]> | Both are measured in seconds. |
| 20:22 | < [R]> | http://smallbusiness.chron.com/make-firefox-wait-longer-time-out-41249.html |
| 20:22 | < [R]> | My source on that. |
| 20:22 | <@RchrdB> | ooh |
| 20:23 | <@RchrdB> | *dom.max_script_run_time |
| 20:23 | | * RchrdB manually turns that down to 1, because there is absolutely no public website that I want to visit which runs javascript for more than 1 wall-clock second. :P |
| 20:31 | < EvilDarkLord> | Thanks, turning up dom.max_script_run_time seems to have done the trick. |
| 20:51 | | Kindamoody is now known as Kindamoody[zZz] |
| 21:14 | | RchrdB [0x2ba22e11@Nightstar-952.jvr.168.194.IP] has quit [[NS] Quit: Leaving] |
| 21:15 | | * froztbyte uses some softwares to recover his music collection from 12 years of atrophied neglect and went-to-multiple-disks duplication and fracturing |
| 21:15 | <@froztbyte> | beets + https://gist.github.com/froztbyte/abf18031e94b4ec780d6 \o/ |
| 21:15 | <@froztbyte> | 3664 of 42944 covered ;D |
| 21:15 | <@froztbyte> | this is gonna take weeks |
| 21:36 | | Turaiel[Offline] is now known as Turaiel |
| 21:51 | | Reiv_ [NSwebIRC@Nightstar-q8avec.kinect.net.nz] has joined #code |
| 21:54 | | Reiv [NSwebIRC@Nightstar-q8avec.kinect.net.nz] has quit [Ping timeout: 121 seconds] |
| 21:59 | | Reiv_ is now known as Reiv |
| 22:27 | | RchrdB [0x2ba22e11@Nightstar-qe9.aug.187.81.IP] has joined #code |
| 22:27 | | mode/#code [+o RchrdB] by ChanServ |
| 22:51 | | grindhold [quassel@Nightstar-uufabm.zebra.fastwebserver.de] has quit [Operation timed out] |
| 22:55 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 22:55 | | mode/#code [+o Checkmate] by ChanServ |
| 22:55 | | Turaiel is now known as Turaiel[Offline] |
| 22:56 | | grindhold [quassel@Nightstar-uufabm.zebra.fastwebserver.de] has joined #code |
| 23:37 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 23:37 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| --- Log closed Thu Sep 25 00:00:18 2014 |