A point is that one of the more useful basic features of NMap, the SYN partial-handshake scan (default when run as root) can't be replicated by nc. It always leaves marks in connect logs. Hping can replicate that feature though: "hping -8 -S known host.com" will SYN scan all ports listed in /etc/services on host.com

That's really one of those one-in-a-million things. Getting trolled like that, Yahoo vuln, open X server on a home *nix box. The stars really aligned.

NMap is the best there is, period. There's not even specialist scanners that can up it's features, especially since you can set packet flags manually in the more recent versions. It really, really fills it's niche. I use it all the time in my daily life just for benign remote service discovery, and I assume many people do too. I've never had anyone complain about it either.

Okay, i stand corrected. The attack described in that article is obviously a professional targeted heist, especially considering the 0day. Just out of curiosity, how was the attack discovered? It should be quite possible to pull off that kind of attack without discovery even considering the spamming (injecting rookits with steganographic connect-back using dual-stage shellcode and making the website look like harmless viagra spam, assuming that the "unknown vulnerability" is a normal client memory corruption class of vuln). How do you know more subtle attacks aren't passing under your radar?

Whoopsie, I posted anonymous for some reason?

The "masses of probes" are just normal automated botnet attacks, and the "unidentified attacks" are probably just unwashed masses of skiddies. If you want me to believe that a real cyberwar (in this case more aptly named "computer espionage") is up and going you better give me or assure me that you have some sort of evidence (like captured transmissions showing that the attackers know what they are looking for in terms of intercepted/exfiltrated data) showing that you're actually being attacked by foreign governments or skilled people with an actual terrorist agenda. There is nothing in TFA except buzzwords, hyperbole and "x declined to comment".

Yay, the headline quotes one of my favorite poems: http://poetry.poetryx.com/poems/784/

In the TV series, she gets her first body at the hospital as a child after a plane crash. Swapping her cyborg body gender in that situation seems unlikely.

I agree. However, the name of the song is indeed "Aerith's Theme".

Well, it won't change much will it? The DDOS will just follow the DNS update? You probably should invest in load-balancing across multiple connections instead. Or Akamai.

Note however that it's possible to detect contemporary rootkits in situ on a live system, even if any process that runs in ring0 obviously has the upper hand in a pure theoretical sense. One technique for this is to compare data returned from API calls with raw reads of the data (this is the technique used by "RootkitRevealer".)

Yeah, that's all true, but the point is you'd think *someone* should have done it by now. It only takes one person and one worm.

Many (most) bot servers include standalone mail/web server code, usually ripped from sendmail and the like.

Take a look at Schneier's arguments against this: http://www.schneier.com/blog/archives/2008/02/benevolent_worm_1.html. One additional point is that stack/heap overflows and other memory-corrupting vulnerabilities often can't be made to be 100% reliable, and can be difficult to code for different service packs and such. This can be, and is, coded around as a matter of course, but a bug in the exploitation process can have disastrous and unpredictable results (in this case, interruption of a large swath of critical internal office file sharing networks.) This doesn't matter to the criminals, but it presumably matters to any prospective "grey hat" worm authors.

Every time i see one of these high-yield Windows remote execution holes, I'm tempted to couple a timed network-stack-erasing payload to it (24 hours should be enough for it to be able to infect through vpn-connected laptops and such) and send it cracking. Then i always begin to wonder why this hasn't been done already; is the combination of narcissistic recklessness and technical competence really that rare? It could be argued that it's more fun to play pranks and infiltrate corporate and government networks, but we don't even see things like that (I know it was more common up to the early 90s, when the "criminal prankster hacker scene" still existed outside of small tight groups...)? Or do people just cover it up? You sysadmins out there, have you ever had anything like that happen to you, or anyone you know?