This shit reminds me of back a few years ago now when TF2 would not work on Arch Linux unless you downloaded and compiled some lib32 package from the AUR, I think this only got fixed when they released a 64bit update for the game(?) it was ridiculous.
Skills issue, proton is fine. VM is gonna perform worse. Setting up GPU passthru is a pita in some cases.
pacman I stumbled upon a somewhat amusing relic from the pre-CoC era of Linux:@@ -248,14 +250,21 @@ static int key_search(alpm_handle_t *handle, const char *fpr,
_alpm_log(handle, ALPM_LOG_DEBUG, "looking up key %s remotely\n", fpr);
- err = gpgme_get_key(ctx, fpr, &key, 0);
+ /* gpg2 goes full retard here. For key searches ONLY, we need to prefix the
+ * key fingerprint with 0x, or the lookup will fail. */
+ fpr_len = strlen(fpr);
+ MALLOC(full_fpr, fpr_len + 3, RET_ERR(handle, ALPM_ERR_MEMORY, -1));
+ sprintf(full_fpr, "0x%s", fpr);
+
+ err = gpgme_get_key(ctx, full_fpr, &key, 0);
+
if(gpg_err_code(err) == GPG_ERR_EOF) {
_alpm_log(handle, ALPM_LOG_DEBUG, "key lookup failed, unknown key\n");
/* Try an alternate lookup using the 8 character fingerprint value, since
* busted-ass keyservers can't support lookups using subkeys with the full
* value as of now. This is why 2012 is not the year of PGP encryption. */
- if(strlen(fpr) > 8) {
- const char *short_fpr = fpr + strlen(fpr) - 8;
+ if(fpr_len - 2 > 8) {
+ const char *short_fpr = memcpy(&full_fpr[fpr_len - 8], "0x", 2);
_alpm_log(handle, ALPM_LOG_DEBUG,
"looking up key %s remotely\n", short_fpr);
err = gpgme_get_key(ctx, short_fpr, &key, 0);
@@ -268,6 +277,8 @@ static int key_search(alpm_handle_t *handle, const char *fpr,
}
}
+ free(full_fpr);
+
if(gpg_err_code(err) != GPG_ERR_NO_ERROR) {
goto error;
}
--
1.7.9.5pacman binary today. 
ah great. FOSS, and linux are irrelevant because everything will be in the cloud, and use ai? The most jeeted video he has ever made. And he has made videos where he's just talking about stuff in India.
I miss SLI too man, but it's time to let it go.
I'm not interested in output of some dalit. Begone, jeet!
You negroes have the worst taste in youtubers I swear. If I have to see one more Brodie/Lunduke/whateverfaggot video in here I'm airdropping Slav Power on this thread again.
It's been suspected for a long time that many of the big DDOS mitigation companies are the reason behind the majority of the issues plaguing the hosting of independent websites. So it's not entirely out of the realm of possibility. What are you hosting exactly?
git and cgit (git read-only frontend). The bots are crawling all of the files on cgit and even with different commits. Its causing cpu usage to hit 100% on all cores. Im hosting it on a potato (raspberry 3b) which works fine as long as its used by real users.
Are you running your server on your own hardware or something? I do get AI traffic on mine but after checking my traffic on [VPS provider of choice] just now, it's like 18 Kb/s average traffic towards my site. Then again, it's an insanely lightweight website and I have my own fail2ban etc on top of whatever [VPS provider of choice] does for anti-DDoS.
Nevermind.
I didn't know this until I told the IT guy at work about my F13 and he's like 'oh cool, Linus invested in that' and I just said "Well now I regret it."
Come up with a fictitious user agent for your real users to spoof, then serve goatse to any non-matching user agent.
At one point I toyed with the idea of modifying one of those git frontends to randomly generate the code views on the web. Basically, when an AI goes to scrape that data, they just get a bunch of gibberish that looks kinda like code. A human (or well behaved AI) could just hit the “download repo” button and get the code at a much lower cost to the server, but badly behaved AIs would get their data set slowly poisoned. This would make these frontends less useful to humans, but it would be kinda funny regardless.
I blocked windows and macintoch user agents now and after a while the AI bots stopped spamming. All of the AI bots had windows/mac user agents. Since my website only contains linux/bsd software I dont care if windows/mac users can't access it.
>Tries to show a example of something that uses low cpu power
Good to hear. I'm one of the few defenders of that series, in large part because it echos my (admittedly limited) experience with Linux.
Which is exactly the point. I keep seeing Linux fanboys pretend to not understand this extremely simple aspect of the challenge, or act as if it's a stupid rule. Makes me suspect they know the Linux community has a massive elitism issue.
I'm going to go on a related, but seemingly off topic rant.
"You should not do your own research and just follow something some retard told you 5 years ago" is a stupid rule. Call me an elitist for simply telling someone that they should go to Linux like anything else, take your time to learn.
The peak concurrent player count of cyberpunk is 1 million. The combined total of the peak player count of the 4 indie games you mentioned is only 800k. Sure, they are averaging more players now, but thats how shit works, they are not story games, they are simulators/world builders, meaning infinite replayability.
No one is saying "Why are you trying to play games with kernel level anticheats?" as though its a sin to play them. You want to use Linux, you have to deal with the fact those games are inherently broken either intentionally or not on Linux.