this post was submitted on 30 Jan 2024
845 points (98.2% liked)

Technology

59598 readers
3147 users here now

This is a most excellent place for technology news and articles.


Our Rules


  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each another!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots


founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] voodooattack 130 points 9 months ago (7 children)

Legacy hardware and operating systems are battle tested, having been extensively probed and patched during their heyday. The same can be said for software written for these platforms – they have been refined to the point that they can execute their intended tasks without incident. If it is ain't broke, don't fix it. One could also argue that dated platforms are less likely to be targeted by modern cybercriminals. Learning the ins and outs of a legacy system does not make sense when there are so few targets still using them. A hacker would be far better off to master something newer that millions of systems still use.

Tell me you know nothing about cybersecurity without telling me you know nothing about cybersecurity. Wtf is this drivel?

[–] ooterness 85 points 9 months ago (1 children)

Simple solution: Don't connect it to the Internet. Hackers hate this one weird trick.

[–] voodooattack 43 points 9 months ago (3 children)

And said trick ends when an attacker manages to socially-engineer their way in. (But maybe they’ll drop floppies instead of flash drives around the block this time)

[–] yesman 33 points 9 months ago (4 children)

You really think that infrastructure IT is dumb unless it can brush off a Stuxnet-like attack by the CIA and Mosad? Most RR traffic signals in the US are run with mechanical logic, physical switches connected to circuits closed by steel wheels on steel tracks. Do you really want a "move fast and break things" tech bro to update all this stuff for us?

All kinds of infrastructure uses ancient software because it's reliable. Updating it just to protect from hackers causing damage is likely to cause that damage unintentionally while doing little to protect from hackers anyhow.

[–] Linkerbaan 9 points 9 months ago (1 children)

It must be updated sometime or risk being archaic and unmanageable. Chances are high they are paying insane amounts for those legacy mechanical switches you mention.

The actual logic is usually very well portable to a more modern ecosystem.

[–] [email protected] 3 points 9 months ago (1 children)

Or these companies could pay to train (no pun intended) technicians to learn the systems they’d like to maintain. No matter how old they are.

Until entropy comes for the actual hardware (assuming they won’t invest in remanufacture or production of replacements). Re-engineering a successfully working system is more costly and might result in worse outcomes, especially in the near term.

[–] Linkerbaan 1 points 9 months ago

Often these system rely on old components which are just not made anymore.

People don't design every switch, computer and chip themselves. They buy whatever mainstream stuff is available at the time and combine it into a system

If you want to resupply those old parts you literally need to search Ebay to buy some weird outdated 2nd hand MSDOS PC to put in your "awesome reliable railway system".

Upgrading at every new whim is of course bad, but once your system reaches legacy age it's often necessary to fully overhaul and modernize it for the next ~15-20 years.

[–] mlg 5 points 9 months ago

Every SCADA related cyber attack and incident has entered the chat.

Even if it's archaic, a lot of these systems aren't secure which can be done relatively easily and cheaply with things like basic firewalls and stunnel.

[–] [email protected] 3 points 9 months ago

Akshually it was recently found that a spy from Holland I think penetrated a chip supply line and installed an infected chip which found it's way into the centrifuge network

[–] RaoulDook 2 points 9 months ago

uses ancient software because it’s reliable

HAHAHA!

I just have to laugh at that idea, since I've been using computers since the days that those OSes were in common use. Reliable is not what I would call a lot of that old stuff for sure.

The bottom line is that ancient software will likely have ancient security vulnerabilities that would be trivial to exploit and take over or destroy those systems. It's not good.

[–] [email protected] 10 points 9 months ago

They could socially engineer their way in regardless of some machine being MSDOS or not. Basically if they can gain physical access to the device, or convince somebody to do something with the device it hardly matters what it was running since it can still be compromised.

[–] [email protected] 2 points 9 months ago (1 children)

Sure, but how likely is this in this specific scenario. We're talking about a system that's not even directly controlling the train but just a display on it. The worst that can happen is that those displays won't work until the system is reinstalled. That's hardly a lucrative target for modern hackers. There's way easier target which are worth something.

[–] voodooattack 4 points 9 months ago* (last edited 9 months ago)

I’m not talking about this specific instance, just that block of misinformation/generalisation. Saying that legacy systems are well-secured because they’re “battle tested” is sheer ignorance.

Take side-channel attacks for example. A timing attack is something programmers from the 60’s and 70’s would not have taken into account when writing their hashing algorithms. And speaking of hashing, what hashing algorithms were available back then? CRC32 or something similar? What about salting? You get the idea.

Not to mention that legacy operating systems don’t get security updates. Let’s assume that DOS is secure (which it definitely isn’t), but if that statement were correct, would it apply to Windows XP as well?

All I’m saying is that the article is dead wrong. As software developers in this century, we’ve come a long way. We’ve developed security best practices, written libraries and frameworks, and come up with mitigations for a lot of these security vulnerabilities. These solutions are something that closed-source legacy systems (and anything without active maintenance) would never benefit from.

[–] [email protected] 31 points 9 months ago (2 children)

It really depends if these systems (that appear to control arrival boards) are on a network or not. If they're not, then there is minimal risk to leave them the way they are. Somebody would need physical access to the devices to do harm. If they are on a network then that's a pretty big deal, but some attacks could be mitigated against by tunnelling and/or additional packet filtering to ensure the integrity of messages.

Continuing on a railway theme you should be FAR more worried all the devices that run up and down the side of railway lines - PLCs that talk with each other and operations centres to control things like lights, junctions, crossings etc. If they're more than 5 years old then chances are then all that traffic is in the clear, and because these things live in boxes by the railway line, it wouldn't take much to break into a network and potentially kill people by running two trains into each other.

[–] [email protected] 16 points 9 months ago (2 children)

the job was advertised as being remote.....

[–] [email protected] 11 points 9 months ago (1 children)

The job might be remote, doesn't mean the system is remote. For all you or I know they want somebody to reverse engineer the protocol of this thing, which could be some weird board & driver that hooks into an old PC so they can switch it out for something else.

[–] [email protected] 14 points 9 months ago (1 children)

It's in the job description, remote access is available via a repurposed laparoscope robot and webcam placed in front of the original terminal keyboard and CRT

[–] [email protected] 2 points 9 months ago* (last edited 9 months ago) (1 children)

I think you are pulling my leg... But if that's true that's super cool.

[–] [email protected] 1 points 9 months ago

A remote KVM through a portal would be the actual way an air gapped system would be accessed, yeah... Spoofing ps/2 or Din with a teensy would probably be needed to use new hardware for the KVM. Maybe a SFF PC with an analog input capture card...

[–] [email protected] 5 points 9 months ago

Well yes. You can code software remotely. That doesn't mean the end system is reachable through the network. Given it's DB, I bet these systems are still patched by floppy. Until very recently they've used floppy's to distribute train schedules to be displayed in the train.

[–] nexusband 4 points 9 months ago

Exactly. And these things are on an internal bus network, but they are not connected to the internet.

[–] [email protected] 15 points 9 months ago

they can execute their intended tasks without incident

Now if only the Deutsche Bahn could do that too

[–] mlg 12 points 9 months ago

Lmao they don't know all the exploits people learn first are the brutally insane and easy stuff that works on outdated machines like heartbleed and eternal blue.

[–] maness300 10 points 9 months ago (1 children)

What exactly is the issue? Everything mentioned is true.

It even goes further when you consider how newer technology often incorporates more technology, which means a greater attack surface.

Tell me you know nothing about cybersecurity without telling me you know nothing about cybersecurity.

Oh, the ironing. Sad how you have >100 upvotes.

[–] voodooattack -5 points 9 months ago* (last edited 9 months ago)

Not sure how to link a reply on lemmy so I’ll just copy from another comment I wrote here:

I’m not talking about this specific instance, just that block of misinformation/generalisation. Saying that legacy systems are well-secured because they’re “battle tested” is sheer ignorance.

Take side-channel attacks for example. A timing attack is something programmers from the 60’s and 70’s would not have taken into account when writing their hashing algorithms. And speaking of hashing, what hashing algorithms were available back then? CRC32 or something similar? What about salting? You get the idea.

Not to mention that legacy operating systems don’t get security updates. Let’s assume that DOS is secure (which it definitely isn’t), but if that statement were correct, would it apply to Windows XP as well?

All I’m saying is that the article is dead wrong. As software developers in this century, we’ve come a long way. We’ve developed security best practices, written libraries and frameworks, and come up with mitigations for a lot of these security vulnerabilities. These solutions are something that closed-source legacy systems (and anything without active maintenance) would never benefit from.

The “ironing” is lost on you in this case.

[–] [email protected] 10 points 9 months ago (1 children)

Cybersecurity != Safety Critical

[–] [email protected] 5 points 9 months ago (1 children)

It is when safety-critical systems are the target of a cyberattack.

[–] [email protected] 3 points 9 months ago

Doesn't sound like this system is safety critical. You should be more worried if some hacker can change train signs from stop to go. If you ever ride on a train and see steel boxes by the side of the track, those are control systems and they run up and down the line. They might be locked, or possibly alarmed but that's about the extent of their protection. A simple attack would be to just take an axe to one, or set fire to it. A more sophisticated attack could snoop on the profinet traffic and do something evil.