Time |
Nick |
Message |
00:36 |
|
pounce joined #minetest |
01:04 |
|
proller joined #minetest |
01:10 |
|
smk joined #minetest |
01:30 |
|
SFENCE joined #minetest |
02:17 |
|
SFENCE joined #minetest |
02:25 |
|
MTDiscord1 joined #minetest |
02:48 |
|
acarrico joined #minetest |
04:19 |
|
v-rob joined #minetest |
04:59 |
|
Markow joined #minetest |
05:00 |
|
MTDiscord joined #minetest |
05:00 |
|
TheCoffeMaker_ joined #minetest |
05:49 |
|
SFENCE joined #minetest |
06:01 |
|
gregon joined #minetest |
06:05 |
|
tarsovbak joined #minetest |
06:51 |
|
SFENCE joined #minetest |
07:00 |
|
lemonzest joined #minetest |
07:01 |
|
SFENCE joined #minetest |
07:05 |
|
SFENCE joined #minetest |
07:07 |
|
YuGiOhJCJ joined #minetest |
07:08 |
|
Verticen joined #minetest |
07:11 |
|
SFENCE joined #minetest |
07:12 |
|
SFENCE joined #minetest |
07:25 |
|
definitelya joined #minetest |
07:27 |
|
SFENCE joined #minetest |
07:34 |
|
calcul0n joined #minetest |
08:18 |
|
Glaedr joined #minetest |
08:59 |
|
Talkless joined #minetest |
09:00 |
|
gregon joined #minetest |
09:03 |
|
gregon joined #minetest |
09:15 |
|
mrkubax10 joined #minetest |
09:17 |
|
___nick___ joined #minetest |
09:17 |
|
TomTom_ joined #minetest |
10:10 |
|
___nick___ joined #minetest |
10:13 |
|
___nick___ joined #minetest |
10:23 |
|
jaca122 joined #minetest |
10:45 |
|
SFENCE joined #minetest |
13:33 |
|
proller joined #minetest |
13:51 |
|
tarsovbak1 joined #minetest |
13:54 |
|
mazes_83 joined #minetest |
14:04 |
|
tarsovbak joined #minetest |
14:11 |
Norkle |
:] |
14:15 |
|
Talkless joined #minetest |
14:17 |
|
SFENCE joined #minetest |
14:18 |
|
tarsovbak1 joined #minetest |
14:49 |
|
Talkless joined #minetest |
14:59 |
|
definitelya joined #minetest |
15:01 |
|
Talkless joined #minetest |
15:02 |
|
tarsovbak joined #minetest |
15:10 |
|
mrkubax10 joined #minetest |
15:11 |
|
appguru joined #minetest |
15:13 |
|
tarsovbak joined #minetest |
15:27 |
|
TheCoffeMaker joined #minetest |
16:20 |
|
SFENCE joined #minetest |
16:31 |
|
SFENCE joined #minetest |
16:42 |
|
SFENCE joined #minetest |
16:45 |
|
SFENCE joined #minetest |
16:55 |
|
s20 joined #minetest |
17:09 |
|
Desour joined #minetest |
17:11 |
|
SFENCE joined #minetest |
17:25 |
|
tarsovbak joined #minetest |
17:40 |
|
tarsovbak1 joined #minetest |
17:47 |
|
tarsovbak joined #minetest |
18:24 |
|
v-rob joined #minetest |
18:26 |
|
gxt joined #minetest |
18:26 |
|
fling joined #minetest |
18:35 |
|
SFENCE joined #minetest |
18:46 |
|
alguien joined #minetest |
18:48 |
alguien |
cheapie, hi - i have a question/bug report |
18:49 |
alguien |
If I pass the touchscreen a quazi-texture with only the botched modifier "[combine!" my client crashes (and I am hardlocked, because it crashes every time I try to rejoin) |
18:49 |
alguien |
However, when I try to render the same quazi-texture using formspec directly, it doesn't crash my client |
18:51 |
alguien |
Would you happen to know why that is? Would you want a bug report with the two test cases? If so, where's your bug tracker? Or if you think it's a lua api bug or something, let me know I should report to the engine. |
19:23 |
|
v-rob joined #minetest |
19:36 |
|
TheCoffeMaker joined #minetest |
19:43 |
|
TheCoffeMaker joined #minetest |
19:44 |
|
acarrico joined #minetest |
19:59 |
|
mrkubax10 joined #minetest |
20:13 |
cheapie |
In case someone else sees alguien again before I do - client crashing sounds like an engine bug (but it's hard to tell when the actual crash message wasn't given...). If there is a bug in the mod I would recommend having mt-mods fix it in their fork, as the original is no longer maintained. |
20:33 |
|
SFENCE joined #minetest |
20:42 |
Krock |
rubenwardy: the rubenwardy.com certificate expired about 2 hours ago |
20:44 |
rubenwardy |
It would be nice if certbot could support namecheap so I don't have to do this manually every 3 months |
20:46 |
rubenwardy |
I suppose I could run my own ns/dns servers |
20:47 |
Krock |
cronjob? |
20:47 |
rubenwardy |
fixed |
20:47 |
MTDiscord |
<paradust> isn't there something that can run on the webserver itself? |
20:47 |
rubenwardy |
Krock: I need to manually edit my DNS settings to verify the wildcard cert |
20:47 |
MTDiscord |
<paradust> ah |
20:47 |
Krock |
fixed. except the image that you linked in the FOSDEM post is N/A: https://rubenwardy.com/static/portfolio/freenode_live_2018_full.jpg |
20:48 |
Krock |
@ https://forum.minetest.net/viewtopic.php?p=430812#p430812 |
20:48 |
rubenwardy |
content.minetest.net is fine as it's a single certificate, not a wildcard, only does well-known verification |
20:48 |
rubenwardy |
Removed the image |
20:49 |
rubenwardy |
looks like people have written manual scripts for this https://simon.aldrich.cc/blog/2021/04/automating-lets-encrypt-wildcard-certificates-with-the-namecheap-api/ |
20:49 |
Krock |
so cronjob it is. |
20:50 |
rubenwardy |
I already use a cronjob, you need more than that |
20:51 |
Krock |
petition to remove the internet with all its complicated issues |
20:52 |
Mantar |
yeah, rewrite the internet in rust plz :) |
20:53 |
rubenwardy |
namecheap is the second most used registrar in the world, I feel like the options should be less sketchy |
20:53 |
rubenwardy |
I suppose most people just don't have ~30 subdomains |
20:55 |
Krock |
almost all domains listed on freedns do have >30 subdomains with the difference that they're not using a certificate for them |
20:55 |
|
alguien joined #minetest |
20:55 |
Krock |
so yes your case is special. |
20:56 |
alguien |
cheapie, got my message? |
20:57 |
cheapie |
alguien: Client crashing sounds like an engine bug (but it's hard to tell when the actual crash message wasn't given...). If there is a bug in the mod I would recommend having mt-mods fix it in their fork, as the original is no longer maintained. |
20:57 |
Krock |
alguien: https://irc.minetest.net/minetest/2024-02-10 |
20:58 |
alguien |
if it were the engine, wouldn't it crash when showing the formspec sans the touchscreen? |
20:58 |
alguien |
thanks Krock |
20:58 |
cheapie |
Are you using the exact same formspec the touchscreen came up with? (and you're still not giving the error message) |
20:58 |
alguien |
and thanks for the reply cheapie |
20:59 |
alguien |
cheapie, something so terrible happens that the log doesn't record anything before crashing |
20:59 |
alguien |
I'll give you the formspec code which works, and touchscreen code which crashes |
21:00 |
alguien |
well works, fails gracefully with the message in the log. That one I can give you as well, sec |
21:00 |
cheapie |
What, it just exits without saying *anything*, not even a segfault? |
21:01 |
|
Waffelo joined #minetest |
21:03 |
|
Verticen joined #minetest |
21:08 |
alguien |
cheapie, https://paste.debian.net/1306904/ |
21:09 |
alguien |
could be segfualt, can't find it in the log now, trying |
21:09 |
alguien |
idk, it's like i saw "core dumped" running strace minetest |
21:10 |
alguien |
but in the log I don't see any additional info |
21:10 |
alguien |
Wait, i'll recreate it once again, will see what it says exactly |
21:10 |
|
Markow joined #minetest |
21:11 |
alguien |
oh wait, i found the tab in which I have the trace output |
21:11 |
cheapie |
If it's segfaulting then there's an engine bug here, formspecs shouldn't be able to do that. |
21:11 |
alguien |
Last three lines: Invalid size of image for texture. / +++ killed by SIGFPE (core dumped) +++ / Floating point exception (core dumped) |
21:12 |
cheapie |
Yeah, IMO the engine should just give up trying to process that "texture" and not crash. |
21:12 |
cheapie |
(or even abort processing the formspec entirely, that would be preferable to crashing) |
21:14 |
alguien |
I think it does that in the formspec-only case. As you might've noticed by the background, the error message is a screenshot of the message from within the game |
21:14 |
alguien |
*the background of the image I linked in the paste |
21:15 |
alguien |
And yeah, it makes no sense to me either that one thing works and the other does not. That's why I'm contacting you |
21:17 |
cheapie |
You're not escaping the [ in the texture name in the first one you showed, so it's probably getting confused there and not even getting to the texmod stuff. |
21:17 |
alguien |
will try |
21:18 |
alguien |
nope still works. |
21:18 |
alguien |
no message in-game either, maybe that was from a different try |
21:18 |
alguien |
this is the line now FYR: image[1,1;10,10;\[combine!] |
21:20 |
alguien |
maybe minetest.escape_formspec_string (or whatever the function's called) does someting different? |
21:20 |
alguien |
let me try |
21:20 |
cheapie |
It's been ages since I've poked at this stuff, but you might need two backslashes. |
21:24 |
alguien |
barring any invisible chars, minetest.formspec_escape gives the same: "\[quote!" |
21:24 |
alguien |
ah, trying |
21:26 |
alguien |
Still works. Forgive long delays, I currently have a weird setup |
21:27 |
alguien |
this time the error message appeared saying cannot load image "\[combine!" - perhaps you don't need to escape the open bracket after all? |
21:28 |
cheapie |
I think at this point this is enough of me offering support for an unsupported mod - about all I can really say at this point is to perhaps try using a formspec closer to what the touchscreen does, and if you still get nothing then perhaps talk to an engine dev who might be more familiar with those bits. |
21:29 |
alguien |
ok, thanks for the attmpet |
21:29 |
alguien |
attempt |
21:29 |
cheapie |
If you do find a mod bug then let the mt-mods people know as they'll probably want to fix it in their version if they haven't already. |
21:30 |
alguien |
ok, thanks |
21:34 |
alguien |
I don't get it, I can no longer reproduce the issue |
21:36 |
alguien |
eh, that's enough for today, maybe i get it next time (will report) |
22:18 |
|
SFENCE joined #minetest |
23:26 |
|
YuGiOhJCJ joined #minetest |
23:32 |
|
whosita joined #minetest |
23:32 |
whosita |
Hello. Are node timers "reliable"? |
23:32 |
|
panwolfram joined #minetest |
23:33 |
whosita |
if I start a node timer, and server crashes before it times out... |
23:33 |
whosita |
will it trigger after server is restarted or no? |
23:34 |
whosita |
I guess I can just try... |
23:37 |
sfan5 |
that's not a simple yes no question |
23:38 |
sfan5 |
they are saved more often than just at shutdown at least |
23:41 |
|
Verticen joined #minetest |
23:50 |
Mantar |
re: reliability, node timers can sometimes die silently if there are large numbers of them |
23:51 |
Mantar |
I had to add an ABM to Exile to restart certain nodes when their timers die |