| --- Log opened Sun Jul 20 00:00:06 2014 |
| 00:04 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 00:04 | | mode/#code [+o Checkmate] by ChanServ |
| 00:09 | <@froztbyte> | McMartin: yeah |
| 00:09 | <@froztbyte> | windows debugging has always seemed nice-by-accident to me (probably just because of market demand) |
| 00:09 | <@froztbyte> | typically on a debian-based system you should be able to get by with installing the debug symbols, but that requires enough beardage on the part of whoever packaged your software |
| 00:10 | <@froztbyte> | (which, having some knowledge of most iterations of that process, can actually mean a pretty significant amount of beardage) |
| 00:13 | <&McMartin> | Heh. |
| 00:14 | <&McMartin> | I don't think it's by accident. WinDbg meets every single UNIX neckbeard stereotype of terrible by-and-for-hardcore-developers design |
| 00:14 | <&McMartin> | I think it's just a different use case |
| 00:14 | <&McMartin> | viz. the basic expectation is that you're actually, like, distributing binaries and want that to be the easy case. |
| 00:15 | <&McMartin> | As opposed to "everything was compiled from source by the same group, so they don't need to split debug symbols, they have the source right there and know what to do with it" |
| 00:16 | <@froztbyte> | McMartin: no, I mean |
| 00:16 | <@froztbyte> | McMartin: rather than it being some arcana which only the hallowed few are familiar with |
| 00:16 | <@froztbyte> | if you have a debugger installed and your CLR app takes a shit, you get a helpful "sup yo, want me to take a look?" in your face |
| 00:17 | <&McMartin> | Oh, right. |
| 00:17 | <&McMartin> | Yeah |
| 00:17 | <&McMartin> | Hell, even if your C++ one does >_> |
| 00:17 | | * McMartin *might* have started saving physical memory dumps on bluesceens once he got kernel debuggers installed for work <_< |
| 00:17 | <@froztbyte> | haha |
| 00:18 | <@froztbyte> | there was a time when I would've known where to start with those (in windows country) |
| 00:18 | <@froztbyte> | but I believe win7 antiquated that knowledge |
| 00:18 | <&McMartin> | "set up autodownload of kernel symbols" |
| 00:18 | <&McMartin> | "!analyze -v" |
| 00:18 | <&McMartin> | (That gives you the stack trace of whatever actually triggered the crash, at least) |
| 00:19 | <@froztbyte> | sure |
| 00:19 | <&McMartin> | Work has a script for it, it's pretty nice, and some scripts to dump *all* threads if that's what you want |
| 00:19 | <@froztbyte> | except I still have very little clue what to do with that :) |
| 00:19 | <&McMartin> | Sure~ |
| 00:19 | <&McMartin> | Step 3 is "stare like stunned fish at stuff that has nothing to do with whatever you knew about"~ |
| 00:19 | <@froztbyte> | while I can traverse a linux system pretty damn well, and would even hazard a BSD or whatever |
| 00:19 | <@froztbyte> | windows is other magic |
| 00:19 | <&McMartin> | Yeah, I can only do anything useful if it was, say, one of my own programs that crashed. |
| 00:20 | <&McMartin> | (We have kernel components in our product) |
| 00:20 | <@froztbyte> | in....last december (2012) I ended up doing some serial code stuff on windows. I'd started with twisted. when I ended up reading the serial implementation and the windows reactor, I organized having a linux VM on the windows machine instead. |
| 00:20 | <&McMartin> | Heh. |
| 00:20 | <&McMartin> | Serial as in COM1:? |
| 00:21 | <@froztbyte> | and all other UART flavouring |
| 00:21 | <&McMartin> | Yikes. Yeah. |
| 00:21 | <&McMartin> | Yeah. |
| 00:21 | <@froztbyte> | :) |
| 00:21 | <&McMartin> | I've been burned by that in VMs -_- |
| 00:21 | <@froztbyte> | I figured out the trick |
| 00:21 | <@froztbyte> | USB-serial device, hardpatch it to the VM |
| 00:22 | <@froztbyte> | most of the VM software now takes care of unbinding the port reference in windows |
| 00:22 | <@froztbyte> | or at least allocating it to busymode appropriately |
| 00:22 | <&McMartin> | VirtualBox apparently accurately simulates unreliable UARTs and does so by default -_- |
| 00:22 | <@froztbyte> | oh? |
| 00:22 | <@froztbyte> | this was virtualbox at the time |
| 00:22 | <&McMartin> | VMware basically just hooks a FIFO to the simulated system and calls it a day |
| 00:22 | <@froztbyte> | I did find all kinds of interesting shit though |
| 00:22 | <@froztbyte> | McMartin: lol |
| 00:22 | <@froztbyte> | McMartin: does that actually end up working right for timings/buffers? |
| 00:23 | <&McMartin> | I dunno. Would not surprise me if it did not |
| 00:23 | <@froztbyte> | haha |
| 00:23 | <&McMartin> | To be clear, I'm talking about opening \\Devices\Serial0 here |
| 00:23 | <&McMartin> | Which is a layer of abstraction up |
| 00:24 | <@froztbyte> | oh god, don't even speak to me about UNC shit |
| 00:24 | <&McMartin> | But if you just do that and jam data through it, VMware gives you an orderly stream of data and VBox will randomly lose bytes. |
| 00:24 | <&McMartin> | OK, fine, /dev/serial0. |
| 00:24 | <@froztbyte> | hahahah |
| 00:24 | <@froztbyte> | NOT THE SAME THING!~ |
| 00:24 | <@froztbyte> | McMartin: UNC is kinda interesting |
| 00:24 | <@froztbyte> | especially on the filesystem side |
| 00:24 | <@froztbyte> | and the non-uniform things you can do depending on which API you hit |
| 00:25 | <&McMartin> | Yeah, it also breaks a lot of presuppositions if you walk in from POSIX |
| 00:25 | <@froztbyte> | afaik it's possible to write a file like "A File.txt" as well as "A file.txt", through one API |
| 00:25 | <@froztbyte> | and then of course literally every other presentation layer takes a shit |
| 00:25 | <&McMartin> | I bet. |
| 00:26 | <@froztbyte> | you can also blow past the path length limit with at least 2 different API calls |
| 00:26 | <@froztbyte> | which becomes pretty entertaining because you can't actually reverse that easily ;p |
| 00:26 | <@froztbyte> | hilarious side note: the linux drivers *don't* have these issues. so guess what you do when you have this issue... |
| 00:27 | <&McMartin> | Let's see, this is not actually Samba's domain, is it, so it'd be ntfstools? |
| 00:28 | <@froztbyte> | ntfs3g, actually, these days |
| 00:28 | <@froztbyte> | also the vfat driver |
| 00:28 | <@froztbyte> | I *think* the vfat driver can also make a bigger file on a vfat volume than the windows stuff will allow you to |
| 00:28 | <@froztbyte> | although I say that under correction |
| 00:28 | <&McMartin> | It doesn't match limits I don't think, no. |
| 00:29 | <@froztbyte> | it respects spec, but not 1:1 on windows' implementation |
| 00:29 | <@froztbyte> | which can again cause some hilarity |
| 00:29 | <&McMartin> | (There are windows builds of the ntfstools, at least, and I know they see real use) |
| 00:29 | <@froztbyte> | .....hah |
| 00:29 | <@froztbyte> | actually, I wonder. |
| 00:29 | <@froztbyte> | ntfstools was kindashitty at one point, with only readonly support being default |
| 00:30 | <@froztbyte> | rw was experimental |
| 00:30 | <@froztbyte> | but I recall something landing in a kernel branch or some shit at some stage |
| 00:30 | <@froztbyte> | (ntfs3g is aldslkjasdlkjasd because fuse) |
| 00:30 | <@froztbyte> | aha: https://www.kernel.org/doc/Documentation/filesystems/ntfs.txt |
| 00:31 | <&McMartin> | Yeah, ntfsresize in particular sees some use in the Windows world |
| 00:31 | <@froztbyte> | huh, weird |
| 00:31 | <@froztbyte> | I would've thought a pirated symantec thing would see more |
| 00:31 | <@froztbyte> | what's that software called? |
| 00:31 | <&McMartin> | Partition Magic? |
| 00:31 | <@froztbyte> | that one |
| 00:31 | <&McMartin> | Man |
| 00:31 | <&McMartin> | i haven't used that in probably 15 years |
| 00:32 | <@froztbyte> | haha |
| 00:32 | <@froztbyte> | I last used it in '04, I think |
| 00:32 | <&McMartin> | Oh, right |
| 00:32 | <@froztbyte> | to merge two partitions |
| 00:32 | <&McMartin> | Yeah, more like 10 |
| 00:32 | <&McMartin> | Because it wasn't until XP's own disk utilities got good enough that I could toss it |
| 00:32 | <@froztbyte> | I think this is also giving you a fair idea of when last I really windows'd ;p |
| 00:32 | <@froztbyte> | McMartin: yeah, XP was /just/ on the cusp of the logvol stuff |
| 00:32 | <@froztbyte> | but you really needed to know your shit |
| 00:32 | <@froztbyte> | and potentially have a fair handful of disks on hand |
| 00:32 | <&McMartin> | Heh. |
| 00:32 | <&McMartin> | Yeah |
| 00:33 | <&McMartin> | And then there was "linux boot CD with fdisk installed" becase gparted wasn't widely available/didn't exist/didn't work |
| 00:33 | <@froztbyte> | hah |
| 00:33 | <&McMartin> | Which I think is how I rolled in the Win98 world. |
| 00:33 | <@froztbyte> | it's odd, I never actually fdisk'd much |
| 00:33 | <@froztbyte> | I cfdisk'd |
| 00:33 | <@froztbyte> | then sfdisk'd |
| 00:34 | <@froztbyte> | then partman-auto/expert-recipe'd |
| 00:34 | <@froztbyte> | relatedly, some fun from http://linux.die.net/man/8/sfdisk |
| 00:35 | <@froztbyte> | -O file |
| 00:35 | <@froztbyte> | Just before writing the new partition, output the sectors that are going to be overwritten to file (where hopefully file resides on another disk, or on a floppy). |
| 00:35 | <@froztbyte> | -I file |
| 00:35 | <@froztbyte> | After destroying your filesystems with an unfortunate sfdisk command, you would have been able to restore the old situation if only you had preserved it using the -O flag. |
| 00:35 | <&McMartin> | Hee hee |
| 00:35 | <&McMartin> | If only. |
| 00:35 | <&McMartin> | If only you had not pointed the output of -O to the disk you proceeded to destroy |
| 00:36 | <@froztbyte> | well, I suspect most people only find -O and -I when they're up shit creek already |
| 00:36 | <@froztbyte> | I do wonder how many cries of anguish that sentence has produced, though |
| 00:36 | <@froztbyte> | McMartin: fwiw, my suspicion is grounded in the same bedding as that "there are 2 types of people. those who backup, and those who haven't lost a disk yet" |
| 00:36 | <@froztbyte> | saying* |
| 00:37 | <@froztbyte> | sfdisk is really neat though |
| 00:38 | <@froztbyte> | it's helped me save some raid arrays a couple of times |
| 00:38 | <&McMartin> | cool |
| 00:38 | <@froztbyte> | the other really handy part is when you need to do some simulation stuff |
| 00:39 | <@froztbyte> | you can just steal the disk headers from some existing disks, and throw them on blockdevices |
| 00:39 | <@froztbyte> | and force iscsi initiators to read those |
| 00:39 | <@froztbyte> | and then trick some dodgy VMs into believing you've given them a real raid-backed iscsi lun |
| 00:40 | | * froztbyte needed a way to test rhel's gfs replication in a way that'd roughly mirror a symmetrix+fibrechannel setup... |
| 00:40 | <@froztbyte> | (also, open-iscsi is (used to be?) a turd) |
| 00:54 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 02:15 | | Kindamoody[zZz] is now known as Kindamoody |
| 04:04 | | Attilla [uid13723@Nightstar-ed0oqj.irccloud.com] has quit [[NS] Quit: Connection closed for inactivity] |
| 04:08 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [[NS] Quit: Program Shutting down] |
| 04:47 | | celticminstrel [celticminst@Nightstar-6pf18d.dsl.bell.ca] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] |
| 04:59 | | Kindamoody is now known as Kindamoody|afk |
| 05:15 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 05:15 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 05:17 | | Derakon is now known as Derakon[AFK] |
| 05:58 | | Ghozer [fake@Nightstar-3oumjf.sfldmi.sbcglobal.net] has joined #code |
| 06:00 | | HotShot [fake@Nightstar-aln4c5.sfldmi.sbcglobal.net] has quit [NickServ (RECOVER command used by Ghozer)] |
| 06:00 | | Ghozer is now known as HotShot |
| 06:27 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 06:53 | | HotShot [fake@Nightstar-3oumjf.sfldmi.sbcglobal.net] has quit [[NS] Quit: Trespassers will be shot, Survivers will be shot again! [Time wasted on Mirc 54mins 24secs]] |
| 07:22 | | Turaiel is now known as Turaiel[Offline] |
| 07:52 | | Kindamoody|afk is now known as Kindamoody |
| 10:45 | | Kindamoody is now known as Kindamoody|afk |
| 10:48 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 10:50 | | Attilla [uid13723@Nightstar-ed0oqj.irccloud.com] has joined #code |
| 11:01 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 11:01 | | mode/#code [+o himi] by ChanServ |
| 12:07 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 12:07 | | jerith [jerith@Nightstar-ngcnma.slipgate.za.net] has quit [[NS] Quit: leaving] |
| 12:10 | | jerith [jerith@Nightstar-ip7ar2.slipgate.net] has joined #code |
| 12:10 | | mode/#code [+ao jerith jerith] by ChanServ |
| 14:14 | | Checkmate [Z@Nightstar-rshgm5.cust.comxnet.dk] has joined #code |
| 14:14 | | mode/#code [+o Checkmate] by ChanServ |
| 14:16 | | Checkmate [Z@Nightstar-rshgm5.cust.comxnet.dk] has quit [[NS] Quit: If the world didn't suck, we'd all fall off.] |
| 14:21 | | Checkmate [Z@Nightstar-rshgm5.cust.comxnet.dk] has joined #code |
| 14:21 | | mode/#code [+o Checkmate] by ChanServ |
| 15:07 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 15:11 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 15:11 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 15:21 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 15:21 | | mode/#code [+o himi] by ChanServ |
| 15:37 | | celticminstrel [celticminst@Nightstar-6pf18d.dsl.bell.ca] has joined #code |
| 15:37 | | mode/#code [+o celticminstrel] by ChanServ |
| 15:38 | | Turaiel[Offline] [Brandon@Nightstar-5st1eu.mi.comcast.net] has quit [Ping timeout: 121 seconds] |
| 15:39 | | Turaiel[Offline] [Brandon@Nightstar-5st1eu.mi.comcast.net] has joined #code |
| 15:43 | | Turaiel[Offline] is now known as Turaiel |
| 16:42 | | Syka [the@Nightstar-d16r4u.vividwireless.net.au] has quit [Ping timeout: 121 seconds] |
| 17:14 | | Syka [the@Nightstar-d16r4u.vividwireless.net.au] has joined #code |
| 20:02 | | Kindamoody|afk is now known as Kindamoody |
| 21:08 | | Kindamoody is now known as Kindamoody[zZz] |
| 21:16 | | Derakon[AFK] is now known as Derakon |
| 21:46 | | macdjord [macdjord@Nightstar-7rac1r.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 21:57 | | macdjord [macdjord@Nightstar-7rac1r.mc.videotron.ca] has joined #code |
| 21:57 | | mode/#code [+o macdjord] by ChanServ |
| 21:59 | | Orthia [orthianz@Nightstar-mvf.9c1.224.119.IP] has quit [Ping timeout: 121 seconds] |
| 22:03 | | Orthia [orthianz@Nightstar-s5d.ntf.224.119.IP] has joined #code |
| 22:04 | | mode/#code [+o Orthia] by ChanServ |
| 22:08 | | Orthia [orthianz@Nightstar-s5d.ntf.224.119.IP] has quit [Ping timeout: 121 seconds] |
| 22:08 | | Orthia [orthianz@Nightstar-s5d.ntf.224.119.IP] has joined #code |
| 22:08 | | mode/#code [+o Orthia] by ChanServ |
| 23:02 | | Checkmate [Z@Nightstar-rshgm5.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 23:18 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 23:18 | | mode/#code [+o Checkmate] by ChanServ |
| --- Log closed Mon Jul 21 00:00:22 2014 |