This report was released 9 days ago, this hack was widely discussed on HN when it happened (https://news.ycombinator.com/item?id=27152402) and I thought the formal postmortem would be of interest !
> On the same day, the Attacker posted a link to a key that would decrypt files encrypted by the Conti ransomware. [..] Without the decryption key, it is unknown whether systems could have been recovered fully [..] but it is highly likely that the recovery timeframe would have been considerably longer.
Is the implication that they paid the ransom?
The report seems to go out of its way to avoid stating why the attacker posted the decryption key.
The health minister at the time explicitly stated that they did not pay the random, directly or indirectly (e.g. via a third party) although realistically not easily verifiable.
The discussion at the time was the perpetrators didn't expect to have the effect they did, effectively halting the entire health service for several weeks to months. I think the ethics element as the other commenter stated is a valid one, as one is playing with another's life when you interfere with medical operations, routine or otherwise
I imagine the hacker was somewhat upset by the fact that the victim seems unlikely to be able to pay up and people are about to start dying soon. Having blood on your hands is not only a different matter ethically, but changes the likelihood of law enforcement actually doing something against you.
Maybe, but unlikely. I think it's more of an "ethics" issue (read: attackers don't want to get more heat than needed and also the HSE would have trouble paying for it)
Yeah looks like they did gave them the decryption tool.
I just know quite a lot of cases where non-health related systems were hit with ransomware over here, and that was the route they took to recover the data.
As usual people ignore messages that basically told them what was happening. Reminds me of the Target hack where they installed some anti hacking system which immediately tossed out warnings which seemed excessive so they turned it off for a few months.
But security is an expense and people don't like paying money.
A financial company I worked for in mid 2000's decided the only thing they needed to do was buy some encryption for the disks their databases ran on, which of course would do nothing to keep someone from just using SQL to extract all our customers credit card data.
What is an acceptable signal to noise ratio for a security tool to be useful? clearly some amount of false positives to any real threat ratio causes people to just ignore it completely. Cue me looking at my npm vulnerabilities with I install packages lol.
We’re not talking about thermal noise here. Each and every signal has a determinate source. You need to go through each and every one, but doing this effectively often involves paying lots of money to “some nerds” (rather than your own in house supplicants) and that’s where this kind of thing usually falls down.
