| Time |
Nick |
Message |
| 00:11 |
|
shadowzone joined #minetest-dev |
| 00:23 |
|
crazyR joined #minetest-dev |
| 00:30 |
|
iqualfragile_ joined #minetest-dev |
| 00:50 |
|
Kalabasa joined #minetest-dev |
| 00:59 |
|
iqualfragile joined #minetest-dev |
| 01:01 |
|
blaise joined #minetest-dev |
| 01:32 |
|
NakedFury joined #minetest-dev |
| 01:49 |
|
kaeza joined #minetest-dev |
| 02:29 |
|
shadowzone joined #minetest-dev |
| 02:51 |
|
est31 joined #minetest-dev |
| 03:02 |
T4im |
hm… can someone please correct lua_api.txt claiming minetest.get_craft_recipe(output) would return a input.method, when it actually returns an input.type? :) |
| 03:03 |
|
est31 joined #minetest-dev |
| 03:07 |
* T4im |
opens it as issue then |
| 03:09 |
VanessaE |
hmmmm: it isn't a backtrace, but is this helpful at all in diagnosing MichaelEh's crashing problem? http://pastebin.com/rPCyueF1 |
| 03:09 |
VanessaE |
just got it today |
| 03:11 |
hmmmm |
hello |
| 03:11 |
hmmmm |
no that's not helpful |
| 03:11 |
VanessaE |
damn. |
| 03:11 |
hmmmm |
see it's never helpful unless we have the symbols |
| 03:12 |
hmmmm |
if we had the exact build setup and the exact version we'd be able to match up the symbols with what's getting generated so we'd have a snowball's chance in figuring out where what went wrong |
| 03:13 |
hmmmm |
but even then... this is -O3 so most symbols are screwed up beyond relief |
| 03:13 |
VanessaE |
well I know he's tried 0.4.11-release, and he's running Win 7.1 Enterprise |
| 03:13 |
hmmmm |
VanessaE, is michaeleh's problem consistent? are you able to get in touch with him? |
| 03:14 |
VanessaE |
absolutely consistent since I enabled strict version checking on my servers. |
| 03:14 |
hmmmm |
okay so it happens on connecting to certain mp servers |
| 03:14 |
hmmmm |
this all helps |
| 03:14 |
VanessaE |
and yeah, the above is from a forum PM today, so it's easy to get ahold of him |
| 03:14 |
VanessaE |
I asked him in my reply if there's any chance of him being able to use a debugger, but I don't expect he'll be able to |
| 03:14 |
hmmmm |
see if I were sapier at this point I'd start making educated guesses as to what might be going wrong and looking at the code responsible for handling that |
| 03:14 |
VanessaE |
(when I read up on it, I couldn't make much sense of it myself) |
| 03:15 |
VanessaE |
however! |
| 03:15 |
hmmmm |
just to be sure, this guy is running windows right? |
| 03:15 |
VanessaE |
he also said in his forum messages that it also affects his SP games too |
| 03:15 |
VanessaE |
yeah, Win 7.1 Enterprise |
| 03:15 |
hmmmm |
oh yeah. |
| 03:15 |
VanessaE |
https://forum.minetest.net/viewtopic.php?f=10&t=4057&p=168553#p168552 |
| 03:15 |
hmmmm |
can you compile a debug version for him?? |
| 03:15 |
VanessaE |
here's the whole exchange, save for the message that included that last crash report. |
| 03:16 |
hmmmm |
or rather that's not helpful |
| 03:16 |
hmmmm |
that's not going to print the symbol hmmm |
| 03:16 |
VanessaE |
I haven't the foggiest idea how to cross-compile for windows, however he's also tried sfan5's recent build as well |
| 03:16 |
VanessaE |
(...-1702c34-win64.7z that is, if I am correct) |
| 03:17 |
Brains |
win 15 |
| 03:17 |
hmmmm |
oh |
| 03:17 |
hmmmm |
yeah i don't cross compile for windows, if anything I compile in a VM |
| 03:17 |
hmmmm |
but i've been having vm problems with windows lately |
| 03:17 |
T4im |
hmmmm: btw, did you forget fixing the nil concatenation on deleteblocks? :D just noticing that's still there… |
| 03:17 |
hmmmm |
yeah i forgot that |
| 03:17 |
hmmmm |
i'll get it |
| 03:18 |
hmmmm |
also fwiw while I was aimlessly flying around in single player today I got a rather abrupt crash |
| 03:18 |
hmmmm |
:( was not able to replicate |
| 03:18 |
VanessaE |
hmmmm: did you notice his "CP sp3 compatibility mode" comment? |
| 03:18 |
VanessaE |
XP* |
| 03:18 |
hmmmm |
yeah |
| 03:18 |
hmmmm |
i haven't the slightest clue how that could make a difference. it'll be obvious in hindsight |
| 03:19 |
hmmmm |
so |
| 03:19 |
VanessaE |
I know exactly jack shit about windows :-/ |
| 03:19 |
T4im |
that's a rather good attribute to have, if you ask me |
| 03:19 |
hmmmm |
how do commercial game developers track down these kinds of bugs with minimal information? |
| 03:19 |
hmmmm |
1). they have a very limited number of official builds that they have all the symbols for |
| 03:20 |
T4im |
don't they employ professional testers for that? |
| 03:20 |
VanessaE |
pro testers and a metric fuckton of configuration variants. |
| 03:20 |
hmmmm |
2). they have lots of time on their hands and low paid grunts to do this sort of gruntwork |
| 03:20 |
VanessaE |
either that or they all link some kind of minimal debugger into their programs and none of us ever notice it :P |
| 03:21 |
VanessaE |
or they have special builds they can send to crashing users for exactly tha purpose |
| 03:21 |
VanessaE |
that* |
| 03:21 |
VanessaE |
"here, run this and send me the output when it dies", etc. |
| 03:21 |
hmmmm |
I notice an error.log all the time for stuff like Far Cry that has nothing except the faulting address and the exception type |
| 03:21 |
hmmmm |
right |
| 03:21 |
VanessaE |
maybe we need a standard procedure for that in MT? |
| 03:22 |
hmmmm |
alrighty every time I feel like my load is getitng lighter I get more work todo |
| 03:22 |
VanessaE |
heh |
| 03:22 |
VanessaE |
sorry :) |
| 03:22 |
crazyR |
lol |
| 03:22 |
hmmmm |
i'll have to add a minidump mechanism for windows if a certain build parameter is specified |
| 03:22 |
hmmmm |
but in general I need to do more work |
| 03:22 |
hmmmm |
so freaking lazy |
| 03:24 |
hmmmm |
the mapgen params system is so FUBAR. i need to desperately do something about this. |
| 03:27 |
hmmmm |
btw was I imagining it or lately have jungles in mapgen v6 not been working? |
| 03:27 |
VanessaE |
they still work for me? |
| 03:27 |
hmmmm |
on a new map |
| 03:27 |
VanessaE |
lemme re-check in my test world. |
| 03:28 |
hmmmm |
regardless, if it is broken, it's fixed now |
| 03:28 |
VanessaE |
(a world I have set aside for delete-regen events) |
| 03:28 |
VanessaE |
nope, jungles work fine, just where I expected it to be. |
| 03:30 |
hmmmm |
VanessaE, could this be MichaelEh's problem? https://github.com/minetest/minetest/issues/2219 |
| 03:32 |
VanessaE |
hmmmm: it is possible - I DO have the forbidden_names mod installed, which uses that function to reject gibberish, "Guestxxxx" and the like, but I don't see how that would apply to 0.4.10 connecting in a restricted-client-version situation? |
| 03:32 |
hmmmm |
that's a problem caused by Zeno's game.cpp refactor. iirc he's aware of it |
| 03:33 |
hmmmm |
just hang tight and we'll see if the fix for that fixes MichaelEh's problem as well |
| 03:34 |
hmmmm |
..regardless, we'll need to add the minidump functionality |
| 03:34 |
hmmmm |
I also have a possible fix for the multi-second lag problem |
| 03:34 |
VanessaE |
ok. |
| 03:36 |
|
Zeno` joined #minetest-dev |
| 03:36 |
VanessaE |
ah look who shows up ;) |
| 03:37 |
Zeno` |
:3 |
| 03:38 |
VanessaE |
you know, I haven't the foggiest idea how to see the max lag.. |
| 03:38 |
Zeno` |
type /status ? |
| 03:38 |
VanessaE |
oh derp |
| 03:39 |
VanessaE |
it's < 1.0 for me |
| 03:39 |
VanessaE |
(in a run of the mill mt_game world) |
| 03:40 |
VanessaE |
in a dreambuilder world on the other hand, it's 9-10 seconds |
| 03:40 |
VanessaE |
even though it takes only about 1 second to respond to the command |
| 03:41 |
VanessaE |
(that's when the mapgen code is actively running) |
| 03:42 |
Zeno` |
max_lag is a weighted average so it will slowly go down |
| 03:42 |
VanessaE |
yep, it is. |
| 03:42 |
Zeno` |
i.e. it doesn't really show the *current* "max_lag" |
| 03:42 |
VanessaE |
*shrug* |
| 03:42 |
hmmmm |
probably a dumb question, but how do you use libbacktrace? |
| 03:43 |
|
electrodude512 joined #minetest-dev |
| 03:44 |
Zeno` |
hmmmm, no idea |
| 03:44 |
hmmmm |
i believe it exposes one specific standard-ish looking posix function |
| 03:47 |
|
diemartin joined #minetest-dev |
| 03:47 |
hmmmm |
hrmm I wonder if it's bad to use this https://gcc.gnu.org/ml/gcc-patches/2013-11/msg02055.html |
| 03:48 |
hmmmm |
my goal is to get sufficient debugging information without telling people to run a debugger |
| 03:56 |
kahrl |
I think I actually just figured #2219 out |
| 03:56 |
ShadowBot |
https://github.com/minetest/minetest/issues/2219 -- Singleplayer segfaults if server rejects client |
| 03:56 |
kahrl |
look at the destructor of Database_SQLite3, then look at the constructor |
| 03:57 |
kahrl |
it seems the server is deleted so early that it never gets around to calling verifyDatabase |
| 03:57 |
hmmmm |
:( I'm sorry kahrl |
| 03:57 |
hmmmm |
that was a total fuckup of my own |
| 03:58 |
kahrl |
it happens :) |
| 03:58 |
hmmmm |
are you going to write the fix for it? |
| 03:58 |
kahrl |
sure |
| 03:58 |
hmmmm |
ok |
| 03:58 |
hmmmm |
I introduce so many bugs |
| 04:02 |
kahrl |
pushing in a few minutes: https://gist.github.com/kahrl/5d0da5b59e9d46851645 |
| 04:43 |
|
Hunterz joined #minetest-dev |
| 04:58 |
|
Megaf joined #minetest-dev |
| 05:00 |
|
chchjesus joined #minetest-dev |
| 05:00 |
|
selat joined #minetest-dev |
| 05:51 |
|
Miner_48er joined #minetest-dev |
| 06:19 |
hmmmm |
what do you guys think about removing se_trans_func and handling Win32 exceptions via a handler that does a minidump and spits out some debug info |
| 06:23 |
hmmmm |
from what I'm seeing of the current win32 debug handlers it seems to be mostly useless. it just prints mostly useless crap the default win32 handler would print anyway |
| 06:32 |
hmmmm |
the thing is, you're not supposed to print to logfiles in se_trans_func as per msdn |
| 06:39 |
|
Hunterz joined #minetest-dev |
| 06:57 |
|
hmmmmm joined #minetest-dev |
| 07:10 |
|
ImQ009 joined #minetest-dev |
| 07:46 |
|
jin_xi joined #minetest-dev |
| 07:56 |
|
nrzkt joined #minetest-dev |
| 07:56 |
nrzkt |
Hi all ! |
| 07:57 |
nrzkt |
may the C++ force be with you |
| 08:42 |
|
kilbith joined #minetest-dev |
| 08:52 |
|
chchjesus__ joined #minetest-dev |
| 08:57 |
|
selat joined #minetest-dev |
| 09:06 |
|
FR^2 joined #minetest-dev |
| 09:54 |
|
ImQ009 joined #minetest-dev |
| 10:25 |
|
casimir joined #minetest-dev |
| 10:31 |
|
kaeza joined #minetest-dev |
| 10:50 |
|
DFeniks joined #minetest-dev |
| 10:51 |
|
DFeniks joined #minetest-dev |
| 11:29 |
|
gregorycu joined #minetest-dev |
| 12:06 |
|
Kalabasa joined #minetest-dev |
| 12:20 |
|
diemartin joined #minetest-dev |
| 12:47 |
|
twoelk joined #minetest-dev |
| 13:21 |
|
crazyR joined #minetest-dev |
| 13:30 |
|
twoelk joined #minetest-dev |
| 13:42 |
|
casimir joined #minetest-dev |
| 13:59 |
|
T4im joined #minetest-dev |
| 14:29 |
|
younishd joined #minetest-dev |
| 14:46 |
|
SopaXorzTaker joined #minetest-dev |
| 15:00 |
|
DFeniks joined #minetest-dev |
| 15:12 |
|
kilbith joined #minetest-dev |
| 15:14 |
|
disablec1 joined #minetest-dev |
| 15:18 |
|
hmmmm joined #minetest-dev |
| 15:23 |
|
shadowzone joined #minetest-dev |
| 15:28 |
|
MinetestForFun joined #minetest-dev |
| 15:39 |
|
ImQ009 joined #minetest-dev |
| 15:41 |
|
alexxs joined #minetest-dev |
| 15:49 |
hmmmm |
http://www.openwall.com/lists/oss-security/2015/01/27/9 |
| 15:52 |
nrzkt |
minetest is using this function ? |
| 15:52 |
hmmmm |
wouldn't it be a nice time to find out? |
| 15:52 |
nrzkt |
no it doesn't use |
| 15:52 |
sfan5 |
(stefan stefan-pc) ~ % sudo pacman -Q glibc |
| 15:52 |
sfan5 |
glibc 2.20-6 |
| 15:53 |
sfan5 |
yay arch linux |
| 15:53 |
nrzkt |
i didn't found gethost in src |
| 15:53 |
* sfan5 |
hides |
| 15:53 |
nrzkt |
yaourt -Q glibc |
| 15:53 |
nrzkt |
:D |
| 15:53 |
nrzkt |
no need to sudo for -Q sfan5 |
| 15:54 |
sfan5 |
i always type sudo in front of pacman because most of the time I'm doing either -S <pkg> or -Syu |
| 15:54 |
nrzkt |
i know |
| 15:58 |
|
niebieski joined #minetest-dev |
| 16:06 |
|
shadowzone joined #minetest-dev |
| 16:09 |
|
MinetestForFun joined #minetest-dev |
| 16:10 |
celeron55 |
that's a very minor vulnerability with barely ever any way of exploiting it |
| 16:11 |
Kray |
it requires local access to the machine |
| 16:11 |
Kray |
or specifically configured exim |
| 16:11 |
celeron55 |
it's basically a concern for the whatever fs-level way of making a program always run as root |
| 16:11 |
celeron55 |
what's it even called, i forgot |
| 16:11 |
Kray |
suid? |
| 16:12 |
celeron55 |
yes, in a multiuser environment that's going to be a problem for some thing somewhere |
| 16:12 |
celeron55 |
maybe rarely |
| 16:13 |
Kray |
i have never run a multiuser setup but i'd think a properly set up one doesn't allow normal possibly untrusted users to access any suid binaries to begin with |
| 16:13 |
Kray |
at least any that would have any reason to call that function |
| 16:14 |
nrzkt |
it can be exploited by DNS poising too |
| 16:14 |
nrzkt |
poisoning* |
| 16:15 |
|
Taoki_1 joined #minetest-dev |
| 16:15 |
celeron55 |
nrzkt: hmm, that might be a real issue too |
| 16:19 |
T4im |
local access? I heared it was a RCE |
| 16:19 |
T4im |
heard* |
| 16:21 |
T4im |
and provided that it is most likely used by tor, there's a real good chance its commonly exploited by certain parties for a while :P |
| 16:21 |
nrzkt |
tor is a good way for this because every outgoing node can do dns poisoning |
| 16:26 |
|
kilbith joined #minetest-dev |
| 16:27 |
T4im |
mail systems probably use that for antispam measures on a regular basis too,… any chance minetest makes use of that indirectly in banning mods or server publishing? |
| 16:28 |
* T4im |
found "# host: ip of server (use gethostbyname(hostname) to get from a dns name)" in the protocol doc |
| 16:28 |
nrzkt |
function is not called in code |
| 16:28 |
nrzkt |
doc said it but grep doesn't found it in code |
| 16:35 |
T4im |
hm… I guess lua sockets make use if it… (irc mods)… |
| 16:35 |
celeron55 |
minetest resolves stuff by using getaddrinfo |
| 16:35 |
celeron55 |
i wonder what curl does; it's used for the server list and some other things |
| 16:39 |
T4im |
they seem to use gethostbyname_r() for ipv4 |
| 16:42 |
T4im |
and gethostbyname if _r is not available |
| 16:42 |
|
SopaXorzTaker joined #minetest-dev |
| 17:01 |
|
CraigyDavi joined #minetest-dev |
| 17:04 |
|
Hunterz joined #minetest-dev |
| 17:11 |
|
PilzAdam joined #minetest-dev |
| 17:17 |
puhfa |
https://github.com/minetest/minetest/commit/d576235409c1c21c25167ccd611fc847e9d23cb8 <- glad to see that was fixed, but doesnt the modulo make the and there redundant? |
| 17:17 |
puhfa |
or can getParam2() return negative values in that case? |
| 17:19 |
|
CraigyDavi joined #minetest-dev |
| 17:21 |
puhfa |
nevermind, i was just brainfarting around again |
| 17:24 |
|
Wayward_One joined #minetest-dev |
| 17:35 |
|
Calinou joined #minetest-dev |
| 17:43 |
|
shadowzone joined #minetest-dev |
| 17:44 |
|
shadowzone_ joined #minetest-dev |
| 17:48 |
|
selat joined #minetest-dev |
| 17:53 |
|
crazyR joined #minetest-dev |
| 18:06 |
|
SopaXorzTaker joined #minetest-dev |
| 18:19 |
|
ImQ009 joined #minetest-dev |
| 18:25 |
|
blaise joined #minetest-dev |
| 18:33 |
|
Krock joined #minetest-dev |
| 18:36 |
|
est31 joined #minetest-dev |
| 18:37 |
|
OldCoder joined #minetest-dev |
| 18:37 |
|
NakedFury joined #minetest-dev |
| 18:42 |
|
electrodude512 joined #minetest-dev |
| 18:46 |
Krock |
Yeah. Tomorrow's the big day |
| 18:47 |
Calinou |
what happens tomorrow? |
| 18:47 |
Krock |
> RELEASE DATE: 2015-01-29** |
| 18:54 |
hmmmm |
tomorrow is also Turbo Thursday for $4 dominos pizzas |
| 18:55 |
hmmmm |
I know I said thursday we release but I may want to pause that so I could get in some last-minute additions and fixes |
| 18:56 |
hmmmm |
been busy with work lately |
| 18:59 |
|
jin_xi joined #minetest-dev |
| 19:05 |
|
younishd joined #minetest-dev |
| 19:28 |
|
rubenwardy joined #minetest-dev |
| 19:28 |
puhfa |
VanessaE: you made them neat meshes for CNCd objects recently. did you have some shading issues with them too (with smooth_lighting enabled)? |
| 19:29 |
puhfa |
i was wondering if missing mtl files might have something to do with that. at least i couldnt find them. where should i look for those? |
| 19:29 |
puhfa |
because if thats the case, its not an issue with the engine after all |
| 19:36 |
|
ImQ009 joined #minetest-dev |
| 20:07 |
|
NakedFury joined #minetest-dev |
| 20:23 |
|
cHyper joined #minetest-dev |
| 20:39 |
|
Player_2 joined #minetest-dev |
| 20:51 |
|
shadowzone joined #minetest-dev |
| 20:55 |
|
Miner_48er joined #minetest-dev |
| 21:11 |
|
JZTech101 joined #minetest-dev |
| 21:14 |
VanessaE |
puhfa: I did make those models, yes, however the engine does not use mtl files at all |
| 21:15 |
puhfa |
ah, okay. i was trying to find the obj loaded code but gave up |
| 21:15 |
puhfa |
didnt remember if it did |
| 21:35 |
|
est31 joined #minetest-dev |
| 21:42 |
|
shadowzone joined #minetest-dev |
| 21:49 |
|
NakedFury joined #minetest-dev |
| 21:55 |
|
est31 joined #minetest-dev |
| 21:57 |
|
acerspyro joined #minetest-dev |
| 21:59 |
|
cheapie joined #minetest-dev |
| 22:05 |
|
Megal joined #minetest-dev |
| 22:35 |
shadowzone |
How come we have the "Change Keys" button and the "f7" to toggle 3rd person view isn't in there? also, before you tell me to git pull and recompile I just did that about 5 minutes ago. |
| 22:39 |
|
cheapie joined #minetest-dev |
| 22:39 |
acerspyro |
shadowzone: "Well, implement it yourself" |
| 22:40 |
shadowzone |
I am looking at the code right now. |
| 22:41 |
acerspyro |
lol it was a joke |
| 23:07 |
est31 |
is there any lua function to determine the size of some text? |
| 23:07 |
|
Ace joined #minetest-dev |
| 23:07 |
est31 |
wrong channel |
| 23:08 |
acerspyro |
y = ax + b |
| 23:09 |
|
AceXintense joined #minetest-dev |
| 23:29 |
|
electrodude512 joined #minetest-dev |
| 23:36 |
|
crazyR joined #minetest-dev |
| 23:43 |
|
shadowzone joined #minetest-dev |
