--- Log opened Tue Jun 18 00:00:15 2013 |
00:15 | | You're now known as TheWatcher[T-2] |
00:21 | | You're now known as TheWatcher[zZzZ] |
00:21 | <@Azash> | Doo de doo |
00:21 | <@Azash> | After collecting some books I took the most interesting ones and put them into a "to-read" folder and organized it |
00:22 | <@Azash> | Only 332 books in it |
00:22 | <@Azash> | I should be done around 2019 |
00:24 | | ktemkin[awhale] is now known as ktemkin |
00:24 | <@froztbyte> | haha |
00:29 | <@Azash> | froztbyte: http://pastebin.com/KfMVnuzR |
00:29 | <@Azash> | :3 |
00:32 | <@froztbyte> | eeeek, windows paths! |
00:32 | <@froztbyte> | Azash: so, shitcan all the RHCE stuff |
00:32 | <@Azash> | Oh? |
00:32 | <@froztbyte> | RHCE is pointless |
00:32 | <@Azash> | I'm not planning to take actual certs |
00:32 | <@froztbyte> | there's a couple of books in there which are definitely worthwhile though |
00:32 | <@Azash> | I use them more as curricula |
00:33 | <@froztbyte> | yeah, no |
00:33 | <@froztbyte> | they're worthless |
00:33 | <@Azash> | Alright |
00:33 | <@froztbyte> | basically, run a centos vm for a bit if you want to know how to drive a thing |
00:33 | <@froztbyte> | there's a couple of minor things to note, and that's really about it |
00:33 | <@froztbyte> | * they love their lvm |
00:34 | <@froztbyte> | * if you're thinking "why can't I find this package?", it's in dag or the other rpm tree, both of which are unofficial |
00:34 | <@froztbyte> | * yum is still a gigantic piece of shit |
00:34 | <@Azash> | :P |
00:35 | <@froztbyte> | * some daemons follow the actual upstream project name, contrary to nearly every other distro out there (httpd vs apache, for instance) |
00:35 | <@froztbyte> | * `/etc/sysconfig` |
00:35 | <@froztbyte> | * kudzu can fuck up your day if you break its state file |
00:35 | <@froztbyte> | and that's about it |
00:36 | <@froztbyte> | but yeah, run a centos or a fedora for a bit, you'll have to unfuck them enough to gain the knowledge necessary ;p |
00:36 | <@McMartin> | Oh hey, speaking of Yum, it's been awhile since I've updated Iodine |
00:36 | <@Azash> | My VPS is running CentOS and I've managed to avoid problems |
00:36 | <@froztbyte> | McMartin: probably /because/ of yum |
00:36 | <@Azash> | The thing is I want a deeper understanding of administration but I never actually get into real problems |
00:36 | <@froztbyte> | McMartin: I'm remarkably opinionated about distros, I'll openly admit that |
00:36 | <@McMartin> | froztbyte: Not... really? |
00:36 | <@McMartin> | (re "because") |
00:37 | <@froztbyte> | but CentOS seriously deserves the name 1 Cent OS |
00:37 | <@froztbyte> | McMartin: ah |
00:37 | <@froztbyte> | McMartin: it would be for me :D |
00:37 | <@McMartin> | "sudo yum upgrade" is not exactly some huge imposition |
00:37 | <@froztbyte> | McMartin: there was a time where I had shit internet and using yum was a horrible disaster |
00:37 | <@McMartin> | "To upgrade: 253 packages" |
00:37 | <@McMartin> | "To download: 48M" |
00:37 | <@froztbyte> | because it would keep picking different upstream servers at each update, they'd have sync issues, and then it'd invalidate its own cache |
00:38 | <@froztbyte> | burning up more cap and flatlining my link with pointless updates |
00:38 | <@froztbyte> | and when you have 2GB/month, and each yum run costs you 2~6MB... |
00:38 | <@Azash> | The only problem I've had with yum on my veeps is that it is incredibly slow to find the fastest server |
00:38 | <@froztbyte> | (run, not install; it did this even with searches) |
00:38 | <@Azash> | Which seems to defeat the point |
00:38 | <@froztbyte> | Azash: yeah, you can drop that plugin and just point it at a couple of predefined ones |
00:39 | <@froztbyte> | iirc there was also another plugin that did it better |
00:39 | <@Azash> | I'll have to look that up~ |
00:39 | <@Azash> | Thanks for the tip |
00:39 | <@froztbyte> | so, re the administration thing you mention |
00:39 | <@froztbyte> | it's really just another linux |
00:39 | <@froztbyte> | same packages, slightly different install tools |
00:40 | <@froztbyte> | I suppose you could look at things like selinux, those kind of default policies? |
00:40 | <@Azash> | Or just do LFS once or thrice |
00:40 | <@froztbyte> | I don't think there's more than a handful of things which I've only ever found on one distro and not the others (or couldn't be taken to the others) |
00:41 | <@froztbyte> | Azash: an lfs is a good kind of terrible idea, yeah |
00:41 | <@froztbyte> | build a clfs, too |
00:41 | <@froztbyte> | and maybe pull apart some embedded firmwares |
00:41 | <@Azash> | Compiled linux from scratch or what? |
00:41 | <@froztbyte> | cross |
00:42 | <@Azash> | Right |
00:42 | <@froztbyte> | (cross-compiled for another system) |
00:42 | <@Azash> | Yeah |
00:42 | <@froztbyte> | download various home router software packages, pull those apart |
00:42 | <@froztbyte> | see how things get symlinked around, become familiar with various daemons |
00:42 | <@froztbyte> | do random `kill -9`'s on systems and unfuck the result |
00:42 | <@froztbyte> | move daemons to their non-standard paths |
00:43 | <@froztbyte> | moving daemons around is especially useful for quickly getting to know how they piece themselves together ;p |
00:45 | <@Azash> | Haha |
00:45 | <@froztbyte> | (but seriously, everyone should just be using debian ;p) |
00:45 | <@Azash> | I like my xubuntu |
00:45 | <@Azash> | :< |
00:45 | <@froztbyte> | that's technically kinda a debian.. |
00:46 | <@froztbyte> | just a somewhat slathered-in-cream one |
00:46 | <@Azash> | I should specify I mean as go-to distro |
00:46 | <@Azash> | Nice interface and decent support~ |
00:48 | <@Azash> | One thing I suppose worth experimenting with is doing LFS or gentoo and learning how to eg. switch to Wayland, set up a new DE or a new packaging system |
00:48 | <@Azash> | Probably simple, probably interesting |
00:48 | <@McMartin> | I found gentoo to be a gigantic timesink for no gain whatsoever |
00:48 | <@froztbyte> | gentoo is great for one thing only |
00:48 | <@froztbyte> | portage is a goddamn excellent build system, given everything in it |
00:49 | <@froztbyte> | if I could use portage for build automation without having to fuck around with the rest of gentoo, I probably would |
00:49 | <@froztbyte> | Azash: ahahaha "probably simple" |
00:49 | <@froztbyte> | Azash: no. |
00:49 | <@froztbyte> | in fact |
00:49 | <@Azash> | McMartin: One of my favourite lecture moments was our Linux administration course here at the department |
00:49 | <@McMartin> | froztbyte: Does this mean they've fixed the unending cavern of recursive spiders that is use flags |
00:49 | <@Azash> | The lecturer is a grizzled Linux admin who claims to have been doing that since Linux 1.0 came out |
00:50 | <@Azash> | And when he was explaining the practical labs someone asked why he hadn't listed Gentoo as a possible distro |
00:50 | <@froztbyte> | just try to take a system that comes with gnome by default, then purge it and replace with something else |
00:50 | <@froztbyte> | that shit infects * |
00:50 | <@froztbyte> | McMartin: no, but that's the gentoo part of it |
00:50 | <@Azash> | And he shot them down with a short rant on people who prefer wanking with knowledge over actual usability |
00:50 | <@Azash> | It was amazing |
00:51 | <@froztbyte> | McMartin: if I could have use flags as an "enable capability" thing (like ./configure --add-foo), that'd be great |
00:51 | <@froztbyte> | portage makes a nice abstraction to makefiles and shitty scripts |
00:51 | <@froztbyte> | but yeah |
00:51 | <@froztbyte> | not gonna happen |
00:51 | <@froztbyte> | Azash: ahahaha |
00:51 | <@froztbyte> | Azash: awesome |
00:51 | <@Azash> | froztbyte: Then it's an interesting challenge re. earlier |
00:51 | <@McMartin> | Yeah, the part I remember most distinctly was --add-unicode being "pick which half of of the distro you want to be able to install, becuase they will be made mutex without telling you which is what" |
00:52 | <@froztbyte> | hmm |
00:52 | <@froztbyte> | when I had a minor bit of gentoo work (2007), the querying tools on that were better |
00:52 | <@froztbyte> | but knowledge of them were still a bit arcane |
00:52 | <@McMartin> | Certain libraries (wxWindows was the most flagrant offender, but it did it basically everywhere) seemed to exacerbate it |
00:52 | <@Azash> | McMartin: How did you know they became mutex? |
00:52 | <@Azash> | Did you.. monitor the situation? |
00:52 | <@McMartin> | I did |
00:52 | <@froztbyte> | well, that's the one option |
00:52 | <@McMartin> | I also reinstalled the entire OS about six times over the timespan, with varying settings |
00:53 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
00:53 | <@McMartin> | Then I installed Fedora Core 3 on a different machine and noticed everything was working out of the box |
00:53 | <@froztbyte> | option 2 is: later try to install something else, wonder why the hell you can't, then figure out you spent 3 days compiling your way into a corner you can't leave |
00:53 | <@McMartin> | And Fedora Core 3 was *really bad* |
00:53 | <@froztbyte> | FC3 was pretty bad, yes |
00:53 | <@froztbyte> | I had FC2, then FC4 (that hurt), FC6, F7, F9 |
00:53 | <@froztbyte> | and then gave up on fedora |
00:53 | <@McMartin> | And it was so much better than the gentoo experience at the time that abandoning in favor of it was a no-brainer |
00:53 | <@froztbyte> | having to unfuck it over slow internet after every 2 major upgrades |
00:53 | <@Azash> | Poor joke |
00:54 | <@froztbyte> | I just didn't want to |
00:54 | <@Azash> | I produced it but there were no consumers |
00:54 | <@froztbyte> | haha |
00:54 | <@McMartin> | Yeah, if F9 didn't work for you you're at least looking at something fair |
00:54 | <@McMartin> | FC4 I skipped, because I was Warned~ |
00:54 | <@McMartin> | Early FCs are reverse star-trek, only the odd-numbered ones were usable |
00:54 | <@froztbyte> | well, cf that whole internet thing |
00:54 | <@McMartin> | F10 was the first workable even-numbered Fedora IME |
00:54 | <@froztbyte> | at this point in time my acquisition of distros was still from the cover disc of Linux Format |
00:55 | <@McMartin> | The problem now is that Fedora basically is GNOME's playground |
00:55 | <@froztbyte> | which I got here a-month-after-UK-release |
00:55 | <@McMartin> | So my Fedora machine is strictly headless |
00:55 | <@froztbyte> | McMartin: that's always been a problem |
00:55 | <@froztbyte> | rawhide is goddamn awful |
00:55 | <@McMartin> | Yeah, I don't touch rawhide |
00:55 | <@McMartin> | Headless though Iodine's run every Fedora since FC6 |
00:56 | <@froztbyte> | I have a bit (not all that little) of a vendetta against most things GTK too, because I believe the GTK is what's wrong with software quality ;p |
00:56 | <@McMartin> | Heh |
00:56 | <@froztbyte> | McMartin: the fact that ubuntu and fedora still rocked Evolution as the default mail client for so long.... |
00:56 | <@McMartin> | How much of that is GTK and how much is GObject |
00:56 | <@froztbyte> | thunderturd isn't *much* better, but at least it fucking stored things in maildir |
00:56 | <@Reiv> | I'm now reminded of Eudoria. ... thingy. The mail client of yesteryear. |
00:56 | <@froztbyte> | McMartin: hmmmmmmm |
00:57 | <@Reiv> | Back when 'mail client' had actual meaning~ |
00:57 | <@froztbyte> | McMartin: hand-in-hand, I'd say |
00:57 | <@froztbyte> | GObject might be the technical reason |
00:57 | <@froztbyte> | but GTK is the political reason |
00:57 | <@froztbyte> | Reiv: haha |
00:57 | <@Reiv> | I don't actually remember the name! Embarrasing, given I used it for eons. |
00:58 | <@froztbyte> | didn't Eudora have some relation to Evolution? or at least, whatever Evolution was called before |
00:58 | <@McMartin> | Oh cripes |
00:58 | <@McMartin> | My parents still use Eudora |
00:58 | <@froztbyte> | Ximian |
00:58 | <@froztbyte> | hmm, nothing on wikipedia |
00:58 | <@Reiv> | McMartin: Wait, it's still /alive/? |
00:59 | <@froztbyte> | oh god |
00:59 | <@froztbyte> | oh god oh god oh god |
00:59 | <@froztbyte> | on the wikipedia page for Evolution (which is distilled from a feature page elsewhere, I'm sure) |
00:59 | <@froztbyte> | * Synchronization via SyncML with SyncEvolution and with Palm OS devices via gnome-pilot. |
00:59 | <@froztbyte> | ahahahaha |
00:59 | | * froztbyte could *never* get any of that shit to work |
00:59 | <@froztbyte> | ever. |
00:59 | <@Reiv> | Oh man, Palm OS |
01:00 | <@Reiv> | That's a sad tale or three and no mistake |
01:00 | <@Reiv> | And also mildly baffling; they pulled a Nokia and I'm not even sure *how* |
01:00 | <@froztbyte> | even though I had 4 devices on hand which were listed as *perfect* support |
01:00 | <@Reiv> | I mean, Palm was around for ages as the Proto Smartphone (just sans the phone bit, and even then) |
01:00 | <@froztbyte> | Reiv: (bad) luck and momentum |
01:00 | <@froztbyte> | they had momentum to keep trying to build on what they had |
01:00 | <@Reiv> | And then actual smartphones showed up and they... fell over? |
01:01 | <@froztbyte> | then crowds like motorola came out and pushed colour screen things on cell phone hardware |
01:01 | <@froztbyte> | HP took a bit of a shit on them with WinCE gear |
01:01 | <@froztbyte> | then people started doing the WinCE thing more, Palm managed to save a little bit of face again during this time (I don't know the exact details, but know they had a minor resurgence) |
01:02 | | himi [fow035@D741F1.243F35.CADC30.81D435] has joined #code |
01:02 | | mode/#code [+o himi] by ChanServ |
01:02 | <@froztbyte> | then the iPhone happened (which was really a "skunkworks" sort of thing, since Apple hired people to focus on being properly far ahead of anyone else), and it all became a shitfest |
01:03 | <@McMartin> | And, 253 updates later, time to reboot to finish the upgrades |
01:03 | | McMartin [mcmartin@Nightstar-c2f37ed0.pltn13.sbcglobal.net] has quit [[NS] Quit: Kernel and X upgrades] |
01:03 | <@froztbyte> | haha |
01:03 | <@froztbyte> | Reiv: so the other fun part of this |
01:04 | <@froztbyte> | Reiv: Nokia is still actually doing "okay" as a business (well, last I checked), it's just their mobile division that's getting drunk and puking all over the corner |
01:04 | <@Azash> | You know nokia wellies are a huge cultural icon here |
01:05 | <@Azash> | Can't imagine they rival the phone business though :P |
01:05 | <@Reiv> | Funny enough, their windows smartphones are actually pretty good |
01:05 | | * froztbyte can't comment, never used |
01:06 | <@froztbyte> | Azash: yeah, the whole rubber business is still pretty ingrained, as I understand it |
01:06 | <@froztbyte> | and their two-way radio gear also still finds its way to many places |
01:07 | <@Azash> | Shame about their military radio :P |
01:07 | <@froztbyte> | hehe |
01:08 | <@Reiv> | Something wrong with it? |
01:08 | <@Azash> | Reiv: The military announced they are replacing them with toughbooks |
01:09 | <@Azash> | Was a brief convo about it earlier |
01:13 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has quit [[NS] Quit: Gone.] |
01:17 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has joined #code |
01:36 | | McMartin [mcmartin@Nightstar-c2f37ed0.pltn13.sbcglobal.net] has joined #code |
01:36 | | mode/#code [+ao McMartin McMartin] by ChanServ |
01:36 | <&McMartin> | That was odd, Nightstar wouldn't let me connect |
01:38 | <@celticminstrel> | That happens to me occasionally... |
01:39 | <@celticminstrel> | Usually connecting directly to a server (instead of to irc.nightstar.net) solves it, as I recall. |
01:39 | <@celticminstrel> | ...for some reason I'm currently doing that anyway. |
01:39 | <&McMartin> | Yeah, I just rmreconns'd it and then re-did the connect |
01:39 | <@celticminstrel> | rmreconns? |
01:39 | <&McMartin> | My other servers were fine |
01:39 | <&McMartin> | That's an irssi command that says "don't try to re-log in yourself" |
01:45 | | Typherix is now known as Typh|offline |
01:46 | <@Azash> | Or rather, cancels an existing reconnect series |
01:47 | <&McMartin> | Sure |
01:50 | | Derakon is now known as Derakon[AFK] |
01:55 | | * McMartin flips through ctypes |
01:56 | <&McMartin> | Hmm, one will want to do a bit of massaging on that before Monocle will talk cleanly to it, I think. I return a lot of things that should be treated as void *s but are not, in fact void * |
02:01 | <&McMartin> | It looks like OCaml systems like to handle this by representing void *s as strings of length sizeof(intptr_t) |
02:02 | | ktemkin is now known as ktemkin[work] |
02:10 | | Derakon[AFK] is now known as Derakon |
02:19 | | * ToxicFrog tries out btsync, declares it tasty |
02:21 | <&McMartin> | Is that some kind of cross between rsync and bittorrent? |
02:26 | <&ToxicFrog> | It's dropbox over bittorrent, basically |
02:26 | <&ToxicFrog> | You connect your machines to a tracker or DHT and it syncs arbitrary directories between all machines with the same key |
02:27 | <&ToxicFrog> | Basically it is the continuous n-way sync tool with local storage I've been looking for for ages |
02:31 | <&ToxicFrog> | Gonna version all my dotfiles into it~ |
02:32 | <&McMartin> | Neat |
02:33 | <&McMartin> | So when a client has a change, it tells the tracker "change block X to be Y" and everyone else mysteriously realizes their DL is corrupt and "fixes" it? |
02:35 | <&ToxicFrog> | I think it actually has a higher-level sync protocol running on top of that and just uses bittorrent for client discovery and data transmission |
02:36 | <&McMartin> | It does seem like you'll need something higher-level to distinguish between "I need to download" and "I've got an upload" |
02:54 | | Vorntastic [Vorn@Nightstar-c769d083.sub-70-211-9.myvzw.com] has joined #code |
02:55 | <@Reiv> | Can others, um, leap in on your transmissions? |
02:58 | | Typh|offline is now known as Typherix |
03:00 | <&ToxicFrog> | Reiv: clarify? |
03:00 | <&ToxicFrog> | Other people can't download your files without knowing the key, if that's what you're asking |
03:01 | <&ToxicFrog> | (which is a 32-digit alphanumeric) |
03:15 | <@Reiv> | hm, safeish then, I suppose |
03:21 | < ktemkin[work]> | Inter-node traffic is secured with 256-bit AES. |
03:28 | < ktemkin[work]> | Looks like the AES key is generated from your key-string, which is apparently at minimum 32 alphanumeric characters (160 bits of entropy). |
03:31 | < Vorntastic> | I thought an alnum was 6 bits. |
03:33 | | * Reiv eyes that. |
03:33 | < ktemkin[work]> | Vorntastic: You need six bits to fit all 36 values, yes. |
03:33 | < ktemkin[work]> | But it's closer to 2^5 (32) than it is to 2^6 (64). |
03:33 | <@Reiv> | So your 256bit security is in fact 32bit. |
03:34 | < Vorntastic> | No, I thought you had 62 values.. |
03:34 | < ktemkin[work]> | Vorntastic: Depends if it's case sensitive. |
03:34 | < ktemkin[work]> | The documentation says "20 bytes of entropy", so I'm assuming it's not. |
03:34 | < Vorntastic> | Hng. |
03:35 | < ktemkin[work]> | 20 * 8 = 160, which is 32*5 |
03:35 | < Vorntastic> | Where do you vget 32, reiv? |
03:36 | < ktemkin[work]> | Your 256-bit security is actually 160-bit. |
03:36 | < ktemkin[work]> | If you choose a keystring that's of minimum length; you can go beyond that to get the full 256. |
03:37 | <@Reiv> | If the key is built off a previous, smaller key, then the bigger key is only as secure as the smaller. |
03:37 | < Vorntastic> | Right but the small key is 160 minimum. |
03:41 | < ktemkin[work]> | Unrelated: Has anyone tried out Opalrb? |
03:43 | < ktemkin[work]> | (It's yet another compiles-to-JS language. It has a few things that make it look a little nicer than CoffeeScript, but also a few things that make me feel as though it'd be more trouble/overhead than it's worth.) |
03:47 | < Vorntastic> | I don't really get compiles-to-js. Js is already pretty sensible as long as you avoid the thorny bits. |
03:49 | | Orthia [orthianz@Nightstar-7c657700.callplus.net.nz] has joined #code |
03:49 | | mode/#code [+o Orthia] by ChanServ |
03:51 | < ktemkin[work]> | CoffeeScript feels cleaner to me than raw JS; it's nice to be able to do quick lambdas, string interpolations, and comprehensions. |
03:52 | < ktemkin[work]> | I don't have a problem with JS itself; it just seems like it requires me to repeat myself when creating certain control structures. |
04:00 | | Typherix is now known as Typh|offline |
04:04 | <&McMartin> | Yeah, CS is pretty obviously "file off the thorny bits" |
04:05 | < ktemkin[work]> | They may go too far, replacing the prototypal inheritance with a classical one. |
04:06 | | Vorntastic [Vorn@Nightstar-c769d083.sub-70-211-9.myvzw.com] has quit [Ping timeout: 121 seconds] |
04:06 | | Vorntastic [Vorn@Nightstar-c769d083.sub-70-211-9.myvzw.com] has joined #code |
04:11 | | Typh|offline is now known as Typherix |
04:11 | | Typherix is now known as Typh|offline |
04:24 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
04:24 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
04:25 | | Vorntastic [Vorn@Nightstar-c769d083.sub-70-211-9.myvzw.com] has quit [[NS] Quit: Bye] |
04:33 | | Reivles [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has joined #code |
04:34 | | Orthia [orthianz@Nightstar-7c657700.callplus.net.nz] has quit [Ping timeout: 121 seconds] |
04:48 | <@Azash> | Is there a legitimate reason why switches can't mediate between separate layer 1 techs? |
04:54 | | Kindamoody[zZz] is now known as Kindamoody |
05:13 | | Turaiel is now known as Turaiel[Offline] |
05:16 | | Reivles is now known as Orthia |
05:28 | | Turaiel[Offline] is now known as Turaiel |
05:47 | | ktemkin[work] is now known as ktemkin |
06:09 | | Orthia [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has quit [Ping timeout: 121 seconds] |
06:12 | | Orthia [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has joined #code |
06:12 | | mode/#code [+o Orthia] by ChanServ |
06:17 | | Turaiel is now known as Turaiel[Offline] |
06:18 | | Derakon is now known as Derakon[AFK] |
06:29 | | Kindamoody is now known as Kindamoody|out |
06:37 | <~Vornicus> | Azash: because switches are Really Dumb |
06:38 | | ErikMesoy|sleep is now known as ErikMesoy |
06:38 | <~Vornicus> | Very often they do their job with physical wires. |
06:57 | | Xires is now known as ^Xires |
07:05 | | himi [fow035@D741F1.243F35.CADC30.81D435] has quit [Ping timeout: 121 seconds] |
07:08 | | You're now known as TheWatcher |
07:37 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [Ping timeout: 121 seconds] |
07:53 | | AverageJoe [evil1@Nightstar-4b668a07.ph.cox.net] has joined #code |
08:00 | < jeroud> | A hub is the "connect all the ports" thing. |
08:02 | < jeroud> | A switch has a little bit of logic and watches MAC addresses fly past so it can filter out traffic that doesn't belong on various ports and reduce collisions. |
08:03 | < JBeshir> | The closest thing to a legitimate reason is, "if it does that we call it a bridge". |
08:13 | | You're now known as TheWatcher[afk] |
08:17 | <@Azash> | Ah, thanks JBeshir |
08:30 | < AverageJoe> | at work we use an 80k appliance which monitors our network and catches malware |
08:30 | < AverageJoe> | i found a way to bypass the checks |
08:31 | < AverageJoe> | 80 thousand dollars per license. highway robbery. |
08:31 | < AverageJoe> | anywho, intrinsic functions fool the engine. |
08:31 | < AverageJoe> | inline assembly code for shit. |
08:32 | < AverageJoe> | from what little info they've shared, the product uses a custom VM image and an API monitor to pick up common malware traits |
08:33 | < AverageJoe> | checks for shit like IsDebuggerPresent, sleep(), internetopenconnection(), etc. |
08:34 | < AverageJoe> | good times |
08:51 | | AverageJoe [evil1@Nightstar-4b668a07.ph.cox.net] has quit [[NS] Quit: Leaving] |
09:45 | | You're now known as TheWatcher |
10:23 | | Reivles [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has joined #code |
10:24 | | Netsplit *.net <-> *.split quits: @PinkFreud, @Reiv, jerith |
10:25 | | Orthia [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has quit [Ping timeout: 121 seconds] |
10:25 | | Netsplit over, joins: @PinkFreud, &jerith, @Reiv |
10:28 | | ktemkin is now known as ktemkin[avol] |
10:40 | | * TheWatcher sets about overhauling code, tries to get his head back into this old codebase |
10:42 | | * Syka fetches TheWatcher the ceremonial "what the hell was I *thinking*" hard liquor |
10:45 | | * Azash makes a note to celebrate major refactors in the future by holding up print-outs of the worst bugs and screaming 'Kali-Ma' |
10:49 | <@TheWatcher> | ... |
10:49 | <@TheWatcher> | pffft |
10:51 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [Operation timed out] |
10:58 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
10:58 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
11:39 | | Reivles is now known as Orthia |
11:43 | | Reivles [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has joined #code |
11:43 | | Orthia [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has quit [Ping timeout: 121 seconds] |
12:08 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
12:08 | | mode/#code [+o himi] by ChanServ |
12:16 | <@Azash> | Who was Swedish here again? |
12:17 | <@Azash> | gnolam, it was |
12:34 | | Reivles is now known as Orth |
12:47 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
13:04 | < [R]> | <Azash> Is there a legitimate reason why switches can't mediate between separate layer 1 techs? <-- use-case? |
13:05 | <@Azash> | More of a curiosity thing |
13:12 | <@gnolam> | Azash: hmm? |
13:17 | <@Azash> | gnolam: So I just had a demo for my database PHP project thing |
13:17 | <@Azash> | And I'm at the stage where I'm so tired I'm laughing at stupid things |
13:17 | <@Azash> | My login is just verified with a lone "yes $username" on the screen |
13:17 | <@Azash> | My test account is called b?rje |
13:18 | <@Azash> | So in the middle of the demo I log in and I stare for a moment at the little "yes b?rje" in the corner and almost have a breakdown |
13:23 | <@gnolam> | O...k. |
13:23 | <@gnolam> | But here, have this: http://i.imgur.com/cZq8oQn.png |
13:24 | <@Azash> | I suppose it's not as funny for Sweden-swedes |
13:25 | < Syka> | hee |
13:25 | <@Azash> | I don't intpol much, what's the eight-ball represent? |
13:38 | <@froztbyte> | yeah, finnish swedes are a different breed |
13:38 | <@froztbyte> | <Azash> Is there a legitimate reason why switches can't mediate between separate layer 1 techs? |
13:38 | <@froztbyte> | you get some that can |
13:38 | <@froztbyte> | (mixed fibre/copper stuff, for instance) |
14:15 | | Orth [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has quit [Ping timeout: 121 seconds] |
14:42 | | You're now known as TheWatcher[afk] |
15:29 | | Derakon[AFK] is now known as Derakon |
15:39 | | jerith [jerith@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
15:57 | <@iospace> | HAHAAHA |
15:57 | <@iospace> | that moment when you just figured out a bug that was puzzling a senior dev for a while |
15:58 | | * Azash throws confetti |
15:58 | <&Derakon> | Off-by-one error? Flipped sign? Incorrect indentation? |
15:59 | <@iospace> | Derakon no, he read a register so he could clear it, then didn't read the register he needed for the next if statement |
15:59 | <&Derakon> | Reading clears registers in whatever system you're using? |
16:00 | <@iospace> | not exactly, he was reading then writing the values back (clearing an ISR reg) |
16:00 | <@iospace> | i was just being simple |
16:01 | <&Derakon> | Righto. |
16:05 | <@iospace> | welp, not the entire problem, but definitely part of it |
16:09 | | * Azash slumps in his chair |
16:09 | <@Azash> | Networking exam over, now I just need to finish my PHP project by Sunday and I'm free to work on my own stuff until September |
16:23 | | * iospace attacks Azash with SMBus/I2C |
16:24 | <@Azash> | The whips on the bus go round and round |
16:29 | | jerith [jerith@Nightstar-bf52129d.slipgate.za.net] has joined #code |
16:29 | | mode/#code [+ao jerith jerith] by ChanServ |
16:33 | | Typh|offline is now known as Typherix |
16:41 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code |
16:41 | | mode/#code [+o celticminstrel] by ChanServ |
16:46 | <@iospace> | BINGO |
17:02 | | ErikMesoy is now known as Harrower |
17:47 | | jeff [NSwebIRC@2D9871.A95144.C9ADC1.F9C673] has joined #code |
18:35 | <@froztbyte> | iospace: I need to do some stuff soon with ODB2 |
18:36 | <@froztbyte> | still not sure how I feel about that |
18:36 | <@iospace> | ODB2? |
18:36 | <@iospace> | (name is ringing a bell) |
18:36 | <@froztbyte> | the onboard sensor stuff on cars and such |
18:36 | <&Derakon> | Builtin diagnostic system for automobiles. |
18:36 | <@iospace> | ah |
18:36 | <@iospace> | i haven't used that |
18:37 | <&Derakon> | You can get little plugin thingies that will tell you why your Check Engine light is on, that kind of thing. |
18:37 | <@Tamber> | OBD* |
18:38 | <@froztbyte> | Tamber: yeah, I keep seeing that |
18:38 | <@froztbyte> | I have no goddamn idea if it's supposed to be OBD or ODB |
18:38 | <@Tamber> | It's meant to be OBD, for On Board Diagnostics. |
18:38 | <@froztbyte> | because some spec docs I found refer to it as "ODB2 (Onboard Diagnostics protocol)", and then I'm just like ".....????????" |
18:39 | <@Tamber> | o.0 |
18:39 | <@froztbyte> | seriously. |
18:40 | <&Derakon> | Wikipedia will redirect ODB2 to OBD, and doesn't otherwise mention "ODB" anywhere in the article. |
18:40 | <&Derakon> | I think it's just that "DB" has a fairly strong basis as a unit of speech, so people naturally want to say ODB instead of OBD. |
18:41 | <@froztbyte> | yeah, that's probable |
18:47 | <@Tamber> | 's certainly easier to say "ODB" than "OBD", for me. And then I end up feeling somewhat annoyed at it. |
18:49 | <@Azash> | https://en.wikipedia.org/wiki/Ol%27_Dirty_Bastard |
19:02 | | Turaiel[Offline] is now known as Turaiel |
19:07 | <@froztbyte> | haha |
19:07 | <@froztbyte> | Azash: so |
19:07 | <@froztbyte> | Azash: tell me things about your network exam |
19:09 | <@Azash> | froztbyte: Basic networking really |
19:09 | <@Azash> | Hang on I have it here |
19:10 | <@Azash> | 1. a) what causes/how to detect packet errors/loss b) Explain NAT c) How does router get MAC, what layer d) How can switches identify correct receiver |
19:11 | <@Azash> | 2. a) What is flow and congestion control, why needed, what purpose, what difference b) How does TCP congestion work in <practical example> c) What if the example fails like bla bla Reno and Tahoe differences |
19:11 | <@Azash> | 3. abc) student gets www page, explain entire process in detail |
19:13 | <@froztbyte> | 1a makes me unhappy, and 3abc would probably not have had enough space for the real answer |
19:14 | <@froztbyte> | but I see |
19:19 | <@Azash> | 1a I just answered "caused by unreliable channels, congestion or running out of TTL and detected with ECCs and duplicate ACKs/timeouts" |
19:20 | <@froztbyte> | see, that's exactly the kind of thing I'm talking about |
19:20 | <@froztbyte> | they're "simple" questions, but have extensively deep answers |
19:21 | <@Azash> | 3abc I covered, let's see.. DHCP, ARP, DNS, SYN-SYNACK-ACK, HTTP GET, FIN-ACK-FIN-ACK |
19:21 | <@Azash> | Yeah it's a fairly intro course so you're not meant to show real mastery |
19:21 | <@froztbyte> | 3abc's answer was probably something like "do a DNS lookup, do local link shit to get to gateway, NAT stage if present, tcp connection neg, get content" |
19:22 | <@froztbyte> | but that's not the entire process in detail :( |
19:22 | <@froztbyte> | (all the intermediary routing, link translation stuff, sdh handover, etc) |
19:22 | <@Azash> | Yeah none of that is covered in the intro course |
19:22 | <@Azash> | There's a follow-up at MSc level that is deliciously painful |
19:22 | <@froztbyte> | wait, which level are you now? |
19:23 | <@Azash> | Things like "homework: the packet is formed by bla bla, write it out in bit level" |
19:23 | <@froztbyte> | otherwise stated as "why the hell only at MSc?" |
19:23 | <@froztbyte> | Azash: "meh" |
19:23 | <@froztbyte> | the number of times I debug things with tcpdump at L2.... |
19:23 | <@Azash> | This course is an intro level thing done at the start of the second BSc year |
19:23 | <@froztbyte> | I've had to find mismatched/missing SIP packets in ~90Mbps |
19:23 | <@Azash> | The lecturer for the networking course actually did a wireshark/tcpdump lab that is apparently going to be part of the BSc from now on |
19:23 | <@froztbyte> | for a single call session |
19:24 | <@Azash> | I might have mentioned my tcpdump woes earlier :b |
19:24 | <@froztbyte> | haha |
19:24 | <@froztbyte> | yes |
19:24 | <@froztbyte> | ah well, doesn't sound too terrible |
19:24 | <@froztbyte> | still a couple of things in it which offend me, but it wouldn't be academia if it actually reflected the real world correctly ;3 |
19:25 | <@Azash> | haha |
19:25 | <@Azash> | The problem is our BSc is.. how should I put it, "width-first" ? |
19:25 | <@froztbyte> | same that I've seen this side |
19:25 | <@Azash> | Then when you have a basic taste for everything you can pick your more advanced things for MSc courses and BSc filler |
19:25 | <@froztbyte> | is your BSc a 4 year affair? |
19:25 | <@Azash> | 3 |
19:26 | <@froztbyte> | okay, so do you not have the Hons stage? |
19:26 | <@Azash> | Nah |
19:26 | <@froztbyte> | (here it's B -> Hons - M -> PhD, afaik) |
19:31 | <@Azash> | froztbyte: http://pastebin.com/dCNG73SN |
19:32 | <@froztbyte> | I see |
19:33 | <@Azash> | Over here you do a BSc in 3, and an MSc in 2 (you're presumed to do both and automatically have the right if you get accepted) |
19:33 | <@Azash> | Then a licenciate or PhD which should take 4-6 |
19:34 | <@froztbyte> | does education at university level still benefit from state funding? |
19:34 | <@froztbyte> | (I recall there's something at the school level which does) |
19:35 | <@Azash> | All our education, universities included, is publically funded |
19:35 | | You're now known as TheWatcher |
19:48 | | Kindamoody|out [Kindamoody@Nightstar-05577424.tbcn.telia.com] has quit [Client exited] |
19:49 | | Kindamoody|autojoin [Kindamoody@Nightstar-05577424.tbcn.telia.com] has joined #code |
19:49 | | mode/#code [+o Kindamoody|autojoin] by ChanServ |
19:49 | | Kindamoody|autojoin is now known as Kindamoody |
20:25 | <@Alek> | publicly. |
20:28 | <@froztbyte> | it appears to be an acceptable form |
20:28 | <@froztbyte> | but it depends who accepts it, I guess |
20:41 | <@iospace> | http://theprofoundprogrammer.com/post/27593573683/text-when-i-find-out-who-wrote -this-im-going me while porting code |
20:44 | < RichyB> | I really think that a RoR photo should not be showing a fast-moving object. http://theprofoundprogrammer.com/post/46872400756/text-ruby-on-rails-is-differen t-but-im-open |
20:44 | | * TheWatcher twitches |
20:45 | <@iospace> | RichyB: that's also the april fools one :V |
20:45 | < RichyB> | Oh |
20:45 | < RichyB> | OH |
20:45 | < RichyB> | huh |
20:46 | <@TheWatcher> | Set up gitlab in work, we need a github-alike and it's the best one I could find (hell, it's practically a carbon copy of parts of github). Needs RoR. |
20:46 | < RichyB> | I am not good at observing things |
20:46 | <@TheWatcher> | Good gods, that was painful |
20:46 | <&ToxicFrog> | Snrk: http://theprofoundprogrammer.com/post/42413774647/leonsbuddydave-eclipse-4-3-cha ngelog-added |
20:46 | < RichyB> | TheWatcher: why do that instead of using Github itself? |
20:47 | <@gnolam> | TheWatcher: The "on rails" bit refers to the part where Ruby ties you to them and then laughs diabolically while twirling its moustache. |
20:47 | < RichyB> | You can just give Github money and put private repos up. |
20:47 | <@TheWatcher> | Yeah, I can /just see/ the student's reactions to that suggestion |
20:47 | <@iospace> | yeah |
20:47 | <@iospace> | ruby on rails |
20:47 | <@iospace> | ugh |
20:47 | < RichyB> | We use Bitbucket for git repo hosting in work because their pricing structure's better for us, but yeah. |
20:48 | <@iospace> | that's why i dropped from SE for CS |
20:48 | <@TheWatcher> | Or we can pay github a shitton of cash we don't have (university!) for a license... and still have to install RoR. |
20:48 | < RichyB> | er, no? |
20:48 | < RichyB> | Github runs on Github's servers, not yours. |
20:49 | < Syka> | RichyB: um |
20:49 | < RichyB> | Anyway. Do you actually need privacy of these repos? It's only studenty code. |
20:49 | < Syka> | RichyB: github enterprise |
20:49 | <@iospace> | RichyB: yes |
20:49 | <@iospace> | it's called "copying code" |
20:49 | < RichyB> | Syka: yeah, I wasn't suggesting that. |
20:49 | <&ToxicFrog> | TheWatcher: do you actually need all of the online diffs and stuff, or just git repo hosting? Because the latter is easy. |
20:49 | <@TheWatcher> | RichyB: students doing assessed exercises |
20:49 | < RichyB> | Okay. |
20:50 | < RichyB> | Why do you need any of the Github features that Gitweb and Gitorious don't have? |
20:51 | < Syka> | gitweb has private repos? |
20:51 | < RichyB> | Let's back up another step |
20:52 | <@Tamber> | And around, and around, the argument goes~ |
20:52 | < RichyB> | why do you need a web-anything? |
20:52 | <@TheWatcher> | RichyB: you've never actually taught undergrads, have you? |
20:52 | < RichyB> | Not intentionally. |
20:52 | <@TheWatcher> | Especially ones doing things like CS and BM |
20:52 | <@TheWatcher> | The flailing over subversion repos was phenominal because they simply couldn't work it out |
20:53 | <@TheWatcher> | and we did most of the setup work for them |
20:53 | < RichyB> | ...you need a github workalike. |
20:53 | < RichyB> | With a big shiny "make e |
20:53 | <@TheWatcher> | they're moving all the students to git, and getting them to self-organise |
20:53 | < RichyB> | With a big shiny "make repo" button and so on. |
20:53 | <@TheWatcher> | Yeah |
20:54 | <@TheWatcher> | Bit shiny buttons, nice shiny web interface, extensive help pages, handholding |
20:55 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
20:55 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
20:57 | <@TheWatcher> | gitweb would be hilarious, gitorious would probably work, but gitorius is another ruby app anyway and I preferred the look and operation of gitlab |
20:57 | < RichyB> | Yeah, sorry for being a dick. |
20:57 | < RichyB> | I did not understand what you were using it for at all. |
20:58 | < RichyB> | Actually what you're doing here is something along the lines of "have every student able to just hump a button on a web form to get a clone URL for a fresh git repo that they can push their homework into" |
20:59 | <@TheWatcher> | Almost, yeah. Although we're trying to encourage more flexibility - basically want them using it to vc everything |
21:00 | <@TheWatcher> | (which makes private repos even more vital, given some of the stuff you find public on github!) |
21:00 | <@TheWatcher> | (>.<) |
21:00 | < RichyB> | If you could count on actual competence then the entire thing would just be a machine set up with isolated shell accounts. :) |
21:01 | | Typherix is now known as Typh|offline |
21:02 | | Kindamoody is now known as Kindamoody[zZz] |
21:04 | <&ToxicFrog> | "vc everything"? |
21:05 | | * Vornicus vc's TF |
21:05 | <&ToxicFrog> | (here, you just get an SVN repo hosted on the department servers and instructions on how to access it via the command line) |
21:05 | < RichyB> | Sure. |
21:06 | < RichyB> | I version-control my .emacs file, my Christmas shopping list, my .plan file, my lecture notes... |
21:06 | <&ToxicFrog> | (I would expect it to be even easier using git; 'type "git clone <url>"') |
21:07 | <&ToxicFrog> | Oh. vc == VC == version control. |
21:07 | <@iospace> | we have our SVN linked up with our ticket system |
21:07 | <@iospace> | it's rather nice actually |
21:09 | <@gnolam> | It'd be funnier if he wanted them to Viet Cong everything. |
21:09 | <@TheWatcher> | ... |
21:09 | <@gnolam> | "Hey, why are there bamboo spikes all over the floor of the computer lab?" |
21:09 | <@iospace> | o_O |
21:10 | <@gnolam> | "And where the hell did all these tunnels come from?" |
21:10 | <&ToxicFrog> | TheWatcher: anyways, good luck, here the idea of a "web interface" for student version control is laughable |
21:11 | <&ToxicFrog> | But then, there isn't really organized teaching of version control until second year anyways, and even then it only happens because the prof who teaches CIS*2750 every year is a complete hardass about it |
21:12 | <@TheWatcher> | Oh, they'll need to use the command line client too, but they need th web interface for management and organisation, sharing, etc |
21:12 | <@TheWatcher> | Also, hey, I managed to not kill dromed with my scripts again, shock and amazement |
21:13 | <&ToxicFrog> | What sort of management and organization? |
21:14 | <@TheWatcher> | creating repos, organising teams of students working on projects, checking contributions, that sor tof thing |
21:15 | <&ToxicFrog> | Aah. |
21:15 | <&ToxicFrog> | (here, repo creation happens automatically at the start of the course, team repos are created ad hoc by the prof, and contribution checking is done with svn log) |
21:16 | | Zemyla [zemyla@Nightstar-8fbb7981.compute-1.amazonaws.com] has quit [Ping timeout: 121 seconds] |
21:18 | | * TheWatcher eyes some of the staff members who need to check contributions |
21:18 | <@TheWatcher> | Let's say that doesn't work too well for them ;.; |
21:18 | <@TheWatcher> | (Yes, staff) |
21:19 | <@TheWatcher> | (I despair at the world sometimes) |
21:30 | | Typh|offline is now known as Typherix |
21:33 | | Zemyla [zemyla@Nightstar-8fbb7981.compute-1.amazonaws.com] has joined #code |
21:33 | | mode/#code [+o Zemyla] by ChanServ |
21:56 | < RichyB> | I saw the best example of the XY problem ever on freenode earlier today. |
21:56 | <@froztbyte> | brand new: http://cloudfixesit.tumblr.com/ |
21:56 | < RichyB> | Best in that the person suffering it actually went with the correct solution. |
21:58 | < RichyB> | Their problem, "How do I debug this problem I'm having with this deprecated feature that nobody likes in this six year old version of Zope?" |
21:59 | <@froztbyte> | that.....doesn't only qualify as the XY Problem |
21:59 | < RichyB> | The actual solution, "Jesus, why do you think you want to use that in the first place? Here's the Pyramid tutorial instead." |
21:59 | < RichyB> | \o/ |
22:00 | <@froztbyte> | hahaha |
22:00 | <@froztbyte> | RichyB: fwiw, that'll go down swimmingly in the other channel, I think |
22:01 | <@froztbyte> | and there was even a quote in the twisted quotefile once, somewhat on topic for that |
22:01 | <~Vornicus> | that /is/ brand new |
22:01 | <@froztbyte> | yeah, I know |
22:01 | <@froztbyte> | like, I saw it being made on IRC ;p |
22:03 | <&ToxicFrog> | froztbyte: seems XYy to me. X: I want to debug this ancient feature; Y: I want something functionally equivalent to the ancient feature I'm using that actually works |
22:03 | <@froztbyte> | ToxicFrog: what I mean is |
22:03 | <@gnolam> | To me, it looks like "these are the tools I have - how do I solve it?" rather than a proper XY problem. |
22:04 | <@froztbyte> | ToxicFrog: you also have the problem of "jesus christ, why does this even exist?" |
22:04 | <@froztbyte> | but I think that falls under hysterical raisins category? |
22:04 | < RichyB> | Most of Zope does, yeah. |
22:05 | <&ToxicFrog> | Hysterical raisins? |
22:05 | <&ToxicFrog> | gnolam: except that "replace the tool with one that doesn't suck" was a valid and, in the end, correct solution |
22:05 | <&ToxicFrog> | IMO, "I am trying to get X to do something it is ill-suited for when the actual answer is to throw out X and use Y instead" is the archetypal XY problem |
22:05 | < RichyB> | Hysterical raisins is a vernacular term meaning "historical reasons" and connotes that one regrets what was done. |
22:06 | <@froztbyte> | yesyes, I didn't say it /wasn't/ the XY problem |
22:06 | <@froztbyte> | just said that it's not *only* the XY problem |
22:06 | < RichyB> | Ah, so you two are in agreement. |
22:06 | < RichyB> | wtf http://seclists.org/fulldisclosure/2013/Jun/149 |
22:07 | <@froztbyte> | without clicking, the puppet one? |
22:07 | <@gnolam> | Eh. In a proper XY problem, you can choose any tool you want. If someone is asking about how to do something on an outdated platform, I assume they don't actually have a choice in the matter. |
22:07 | < RichyB> | Joomla. |
22:07 | <@gnolam> | Because then you can go from XY to Apollo 13. |
22:08 | < RichyB> | gnolam: to clarify: fortunately, in this case the problem was "I want to learn to write web-apps" and their choice of X was motivated by "I found this ten-year-old (!) book on Zope lying around." |
22:08 | <@gnolam> | See, that was the vital missing information. |
22:09 | < RichyB> | froztbyte: I hadn't heard of the Puppet one. WTF? |
22:09 | <@froztbyte> | oh, this is joomla |
22:09 | <@froztbyte> | not even surprised |
22:09 | <@froztbyte> | RichyB: ruby/rails issue |
22:09 | <@froztbyte> | RichyB: puppet used the same mechanism and only realized now |
22:10 | < RichyB> | ahahahah |
22:10 | < RichyB> | Oh well. From what I've been told, you really don't want to run the puppet server in the first place. |
22:10 | <@froztbyte> | again, see the other channel for appropriate info ;p |
22:10 | <@froztbyte> | RichyB: yeah |
22:10 | <@froztbyte> | it doesn't scale very well |
22:11 | <@froztbyte> | and you get thundering turds |
22:11 | <@froztbyte> | (not herds, herds can be useful) |
22:11 | < RichyB> | Supposedly you're much better off running puppet standalone and using something else (I've heard of people using git) to pull down latest configs. |
22:11 | <@froztbyte> | yar |
22:11 | <@froztbyte> | and either controlled deployments (per node group, perhaps), or a script that distributes your pull time |
22:12 | <@froztbyte> | so that you don't have everything hitting your server every 5 minutes |
22:12 | <@froztbyte> | well, that's one approach, but yeah |
22:16 | | * Vornicus ADDs at his code |
22:16 | | Harrower is now known as ErikMesoy|sleep |
22:19 | | jeff [NSwebIRC@2D9871.A95144.C9ADC1.F9C673] has quit [Ping timeout: 121 seconds] |
22:26 | <~Vornicus> | It does not help that I haven't organized a C++ project in years, so it's a very tangly mess |
22:37 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
23:07 | < RichyB> | My colleagues are probably going to kill me when they see the single 842-line commit I just pushed. |
23:07 | < RichyB> | Ah well. Part of that was a ~50 line commit message explaining the rationale. |
23:31 | <@Azash> | Commit early, commit rarely |
23:34 | < RichyB> | I couldn't find many smaller atomic parts. |
23:34 | < RichyB> | Okay, except for that one where I corrected another six identifier mis-capitalisations. |
23:36 | <&McMartin> | Vornicus: I seriously ahd enough trouble with that that it was part of the "shift to a C API" discipline. >_< |
23:43 | <~Vornicus> | well, it wouldn't be so bad if I had it organized at all |
23:59 | <~Vornicus> | But it's not, mostly because I'm adding things ad-hoc, from the top and bottom at the same time. |
--- Log closed Wed Jun 19 00:00:29 2013 |