9 ms·
Ancient Linux servers: The blighted slum houses of the Internet
- NoodleIncident 13y agoWould it be feasible to write a virus that only exploits vulnerabilities more than, say, 18 months old, to either encourage or force sysadmins to upgrade? If you get access, then you can look for ssh keys and whatnot to figure out who to send form letters to. "Your server is vulnerable to x, y, and z: please upgrade a, b, and c." It would be more fun if the virus tried to upgrade a, b, and c itself, though. Even if it fails, a broken server that used to process credit cards is better than a vulnerable one, right?
- kjs3 13y agoYou don't fix bad behavior with bad behavior.
- greenyoda 13y agoAs we know from recent history, many companies have pursued criminal charges against people who have broken into their systems, even if they had no intention to do harm. And DAs are very happy to prosecute these cases. So this would be a great virus to write if you wanted to add a prison sentence to your resume. (And you could be held responsible for the cost of restoring the servers after your failed upgrades, which could run into the millions of dollars if you damaged enough servers.)
- bediger4000 13y agoBack in 1996, I heard two different people complain that since Microsoft convinced anti-virus companies to support Windows 95, and drop support for Windows 3.11, new viruses for Windows 3.11 were forcing them to upgrade to Windows 95. At least one of these folks was perfectly happy with Windows 3.11, had a huge stash of Windows 3.11 games on CDs and just plain didn't want to upgrade. He saw the whole thing as a charade to force people to shell out for upgrades.
- mikeash 13y agoThis is playing out again pretty much word-for-word with the termination of support for Windows XP.
- justin66 13y agoIt would have been a more valid complaint for someone compelled to upgrade from Win3.1 to Windows 95. At this stage, Windows XP users are like Japanese soldiers stranded on an island, growing old not knowing they lost the war. Except without the excuse of being on an island.
- rprospero 13y agoAs someone stuck supporting a Windows XP server, I know we lost the war. We raised a big white flag and rolled out a welcome mat for our new overlords. Still waiting for someone to show up, though. (To break the metaphor, our server is hooked up to a PCI-X card which runs a radiation sensor. The drivers for the card will not run under anything after XP. The vendor has discussed possibly considering looking into designing a new card that would might run under Vista, but that's probably five years down the road.)
- derekp7 13y agoSounds like you need a new vendor? As an alternative, could you run XP in a minimum VM, which has that particular PCI card's address passed directly to the VM? Then make sure that there is no other way into the XP VM (i.e., don't run any other apps on it, and keep it only on the host-only network).
- keithpeter 13y agoXP Embedded? Still has support.
- HeyLaughingBoy 13y agoNot for long. XPe is already at EOL. Support ends in 2016 and license availability ends in 2017. Our products using XP Embedded are expected to still be in the field well past 2020. We are already making plans to transition to Windows 8 simply because it will soon be impossible to buy XPe licenses.
- backdraw_ 13y ago> Even if it fails, a broken server that used to process credit cards is better than a vulnerable one, right? "I saw that you dropped your wallet on the floor, so I threw it into the river so nobody could steal your credit cards."
- andrewflnr 13y agoIf returning the wallet (securing the server) is out of the question, throwing my wallet in the river could easily be preferable to leaving it loose. Of course if someone told me that in meatspace I wouldn't believe them... hm.
- backdraw_ 13y agoYou should only intentionally kill their server if you're willing to compensate them for the revenue they lost while they try to figure out what the hell happened. An alternate analogy would be something like: "I saw you forgot to install a lock on door, so I superglued it shut."
- ChuckMcM 13y agoThis is an interesting problem. Too many people just ignore their machine once it "works" and leave it to the wild, or their attention span drops and ignore it. The painful part is when previously "good" internet sites are now injecting sophisticated malware unbeknownst to the owner.
- incision 13y agoWhen I read this I can't help but think about how often I run across people in development or design who don't seem to see difficulty in or value of system administration. It's pretty trivial to turn up systems that work in the most basic sense, but given enough time or load things start to break down. The skill isn't in making things work, it's making them work efficiently, securely and automating what will keep them in that state while planning ahead to make it as easy as possible to recover, replace, extend and transition from them in the future.
- dijit 13y agoSysadmins create an intangible benefit to companies, I've had a lot of jobs come up that are 'DevOps', by which they mean (incorrectly), that I'm a developer who 'knows how to configure apache'. Someone remarked today that 'Sysadmin' may return to being a term of endearment as they specialise in unfucking the systems created quickly and scaled organically (much like taking on a DBA after 15 years of having a massively growing dataset). it'll probably mean people are far more specialised.
- jvagner 13y agoSee: mainframe operators.
- derefr 13y agoAre you sure that's what sysadmins themselves do? In practice, the "sysadmin" seems to be "that guy who does all the stuff that we don't know how to automate." The people who do the automation--if it's repeatable, instead of per-site one-off stuff--are just called developers. E.g., the Docker developers, or the Erlang developers.
- incision 13y ago>"Are you sure that's what sysadmins themselves do?" By my standard yes, that's what they do. Of course, I expect there are a whole host of common definitions and countless pet ones. >"In practice, the "sysadmin" seems to be "that guy who does all the stuff that we don't know how to automate." I would call that person an operator or technician with the amount of manual work and a sort of single-threaded skillset being defining characteristics. >"The people who do the automation--if it's repeatable, instead of per-site one-off stuff--are just called developers. E.g., the Docker developers, or the Erlang developers." I can't claim to have met many people who are called or would call themselves developers who have the breadth of knowledge in the layers below the interpreter or compiler to manage or automate across many dissimilar systems. I imagine this is depends on the environment though, the younger, more homogeneous and closer the company product is to the technology they employ the more likely I'd expect what I would consider developers able to handle those matters. When a company gets older, more segmented and the business functions further from the technologies I could see the role being more that of a sysadmin.
- keithpeter 13y agoEnterprise Linux and version numbering[1]: I'm a bit worried by quotes like "The mass infection works against servers running version 2.6 of the Linux operating system kernel, some using releases from 2007 or earlier, Lee said." Some servers running 2.6.x kernels are as secure as anything out there! [1] https://access.redhat.com/site/security/updates/backporting/ https://access.redhat.com/site/security/updates/backporting/
- duaneb 13y agoWhy is that worrying? I don't read into that an implication that linux 2.6 is insecure, just the terribly vulnerable versions they were using.
- mpyne 13y agoThe problem is that if you simply say that, something like "2.6.9 is vulnerable", it might not actually be true. Enterprise Linux vendors might actually be running 2.6.9 but with relevant security patches backported.
- keithpeter 13y agoWhen it get read out of context by phbs.
- bananas 13y agoThis is quite funny because the other day I was asked by a friend to look at the box serving his company site just on a whim. I've never looked at or discussed the tech he was using. Turns out he paid a company for "managed service" in 2004. They were crap so he stopped paying for the managed bit in 2006 but paid the £90/month for the 1U rack space. He was running IMAP and a static site. IMAP stopped working one day so he moved it to hosted Exchange and shrugged it off. The static site was fine and has been updated via FTP since by his design agency who don't know arse from elbow. So the first thing I did was nmap it. SSH, HTTP, SMTP, IMAP open. Fair enough. SSH'ed in. Some interesting surprises in there. 10 year old DL360 G2 with one dead disk in the array with CentOS 3 on it. CPU spammed 100% with just about every bit of crap that could work its way through the ancient OpenSSH build (and presumably the cgi-bin and IMAP server (cyrus). Sooooo... FTP'ed all the files of it, fired up a digital ocean box and locked it down, uploaded the static site to it, changed the A record on his DNS and shut the old box down. I know this will only postpone the problem for another couple of years but what can you do with people like this? All I got was a "meh" and a crate of lager for my troubles ( granted it was 45 mins work but...)
- sentenza 13y agoWhat... what hosting provider will not notice this? Shouldn't their abuse mail address be glowing red?
- dredmorbius 13y agoDo you have any idea of how many abuse addresses are /dev/nulled? Sadly: black-hat takedown pages are sometimes your best sources of information. A few weeks back, a site I visit periodically was pwned via what appears to have been a Wordpress bug, based on similarities between it and other sites listed by the black-hat hacker. I fired off mail to the standard contact addresses for the site (abuse, webmaster, postmaster), all of which bounced. The listed WHOIS contact actually worked (an argument against domains by proxy), but I also reached out to the upstream hosting provider, who had several other hosts similarly taken down, though other providers were also affected. Sadly, and from my own experience, monitoring abuse for smaller shops (1-2 man admin team, far too many fires going on) is often a very, very low priority.
- foohbarbaz 13y agoI am running a box built with Fedora Core 4 (2007 vintage). Never patch any systems. Why would I? If I am running a service facing the internet, it's custom built and patches would do it no good. Why would I wait for vendor to release a patch? If a service is external, I will watch out for vulnerabilities and rebuild ASAP before any patches are out. Besides, 90% of the time my custom build is not even vulnerable to a particular problem. If I am NOT running a service, why would I care about patches for it? Why would I wholesale patch a server anyway? If somebody breaks in and gets a local shell, all is lost anyway. If they are not in, they are dealing with externally facing services only, see above. There are specific and counted number of daemons on every machine. This whole patch-update thing is misguided and for people that want assurances and no responsibility.
- ars 13y agoPlease tell me you are joking. The is no way you could possibly keep up with every vulnerability of everything installed on your server. > If somebody breaks in and gets a local shell, all is lost anyway. That is not true at all. You should run your server such that someone could get a shell running as the apache user - and still be able to do very little. They could read files and the database (which is bad), but not modify any files (which would be worse).
- andrewflnr 13y agoWRT local shells, it might well be a good idea to assume someone who got a shell as apache could use a privilege escalation 0-day and do some more damage. Hopefully your deployment process is such that starting from scratch isn't a huge hardship. I'd appreciate of someone with actual security experience (not me) weighed in...
- foohbarbaz 13y agoIt's very naive to think that you can protect a box from somebody with a local shell. Never worked.
- foohbarbaz 13y ago
- Zaephyr 13y agoA browser add-in that would check and warn/block loading when the OS and language versions aren't sufficiently patched might be useful security mechanism.