0

I am looking for opinions and point of view of experienced people on this question, as I am not able to find a lot of resources. Let me define the terms first, so that we all talk about the same things.

What I call a fault is an unintentional problem that may occur in software or hardware, because of a computation resources problem, a bad implementation, a natural event ...

A cyber-attack is an intentional, malicious action to compromise a system (steal data, tamper with data, modify system behaviour ...).

My question is simple: is there a chance that, when an attacker attacks a system, the occurrence of faults of this system increase? If yes, is there any sources to justify this claim?

In my opinion, it is the case. If an attacker uploads a script on a resource-constrained execution platform, this script may cause the platform to not have enough available resources to execute its expected tasks. Hence, the fault probability is higher than when the script is not there.

| improve this question | |
2

Yes, there's definitely a good chance that occurrence of faults in a system increase in case of attacks and compromised systems.

On a compromised system you may have any or even all of the following:

  • tasks running in background (that may interfere with legit processes)

  • terminated processes (many things like backup-related processes will be stopped on an infiltrated system in the case of ransomware-like scenario)

  • altered general system settings (may affect many things from specific programs to the overall system functionality)

  • altered user accounts (changing account passwords can prevent various things from running under these accounts)

  • unusually high resource usage (in the case of miners, which can make other legit processes crash or have low performance)

  • scripts altering results of normal actions (like outlook scripts stealing passwords and then spamming with your account)

Any of them can cause a legit process/task/program to become faulty.

| improve this answer | |
  • Thanks a lot. That's what I was looking for. I am still trying to find some research articles/technical papers about this but it is not so easy to find interesting sources. – Ecterion Mar 11 '19 at 7:46
  • Yes, that's true, since it's more a matter of logic - that's why online articles are not focused on such matters. – Overmind Mar 11 '19 at 7:56
  • Most of the works are matter of logic to be honest. What i am interested in is how an attack impact the occurrence of such errors? Is there a point where some attacks impact this occurrence so much that the probability for an error to occur goes beyond an accepted threshold? If yes, is it possible to modify the system such that an attack will not provoke the above? – Ecterion Mar 11 '19 at 8:05
  • Well it really depends on the type of attack you deal with and type of service you want to protect. A ransmoware 100% guaranteed will terminate all your backup-related jobs/programs/tasks, so it's a certain occurrence of errors in this case. To reduce the risk you will have to anticipate what attacks you would expect and act accordingly (like making the backup independent of the local machine in case of this example). – Overmind Mar 11 '19 at 9:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.