pytm icon indicating copy to clipboard operation
pytm copied to clipboard

DO02 - huh?

Open colesmj opened this issue 6 years ago • 14 comments

"DO02": { "description": "Potential Process Crash or Stop", "source": (Process, Datastore, Element), "target": Process, "condition": "target.handlesCrashes is False", },

What is the thought process behind this threat? Is it that the Process can crash, or that it could crash in some security-relevant way? It would seem to me that if the concern is that a Process may crash, then the conditions one might check for would be susceptibility to a buffer overflow in unmanaged code, not whether or not it can "handle" crashes, whatever that means...

colesmj avatar Oct 21 '18 15:10 colesmj

yep - i think this might be a "migration" from the MSFT tool repertoire. I think the stress is on the handlesCrashes value - as in "your process should be able to handle crashes gracefully". Perhaps the description should better reflect that.

izar avatar Oct 26 '18 21:10 izar

More than the description - when would someone set handlesCrashes to True?

colesmj avatar Oct 27 '18 00:10 colesmj

In a Process, if it knows how to clean up (no core files, no temporary files, perhaps restarts after crash) , notifies and logs crash?

izar avatar Oct 30 '18 18:10 izar

Or if it handles all exceptions?

On Tue, Oct 30, 2018, 14:46 Izar Tarandach [email protected] wrote:

In a Process, if it knows how to clean up (no core files, no temporary files, perhaps restarts after crash) , notifies and logs crash?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/izar/pytm/issues/31#issuecomment-434420545, or mute the thread https://github.com/notifications/unsubscribe-auth/AlkM-iJq5EgLAnUS9GpvRMC_X4-08EKRks5uqJ6FgaJpZM4XyhAl .

--

-- Matt Coles

colesmj avatar Oct 30 '18 19:10 colesmj

It may be time to introduce some logic into the attributes?

obj.handlesExceptions = True + obj.cleansUp = True | ==> obj.handlesCrashes = True
obj.restartsAfterCrash = True +

izar avatar Oct 30 '18 21:10 izar

This threat is targeted more towards those applications that reveal stack traces when something goes wrong etc. I'm on the keeping it simple side for the threats as I'm afraid adding logic into attributes may raise the complexity and entry barrier for folks adopting pytm. What do you think?

shambho avatar Oct 30 '18 22:10 shambho

I don't know, I see it as a logical progression - as in beginners can keep it simple but more advanced users may use the logic if they need.

izar avatar Oct 30 '18 22:10 izar

Either way it sounds like there is enough confusion that this threats needs an edit, or split into 2 - handles crashes, and exposes sensitive data on error. Adding metadata such as the CWE would be useful here.

On Tue, Oct 30, 2018, 18:40 Izar Tarandach [email protected] wrote:

I don't know, I see it as a logical progression - as in beginners can keep it simple but more advanced users may use the logic if they need.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/izar/pytm/issues/31#issuecomment-434495908, or mute the thread https://github.com/notifications/unsubscribe-auth/AlkM-vAJbW-bkBBdbD9ZZutODufF5vF8ks5uqNVUgaJpZM4XyhAl .

--

-- Matt Coles

colesmj avatar Oct 30 '18 22:10 colesmj

PRs! We need PRs! :smile:

izar avatar Oct 30 '18 22:10 izar

Just a reply to Rohit on complexity and barrier to entry - threats and condition logic can be complex without impacting the users who want to keep things simple in how they describe the objects analyzed for threats. Making the objects "flat" in their properties enables this, with perhaps limits on what can be determined.

On Tue, Oct 30, 2018, 18:51 Izar Tarandach [email protected] wrote:

PRs! We need PRs! 😄

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/izar/pytm/issues/31#issuecomment-434498538, or mute the thread https://github.com/notifications/unsubscribe-auth/AlkM-ujDdZ1KafAUfWfAs8ojtz0E45fUks5uqNfhgaJpZM4XyhAl .

--

-- Matt Coles

colesmj avatar Oct 30 '18 22:10 colesmj

Cool, sounds good!

I'll update this threat since I'm working on creating a schema for the threats.

Just to make sure, I'll update the condition for this threat as: are you handling process crashes gracefully and/or are you revealing sensitive info on crash. Ok?

shambho avatar Oct 30 '18 23:10 shambho

Might be best to split into 2 as I suggested earlier. A process that handles a crash may still expose data, and a process that avoids exposing data on crash may not handle the crash properly; they lead to different weaknesses.

On Tue, Oct 30, 2018, 19:00 Rohit Shambhuni [email protected] wrote:

Cool, sounds good!

I'll update this threat since I'm working on creating a schema for the threats.

Just to make sure, I'll update the condition for this threat as: are you handling process crashes gracefully and/or are you revealing sensitive info on crash. Ok?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/izar/pytm/issues/31#issuecomment-434500857, or mute the thread https://github.com/notifications/unsubscribe-auth/AlkM-qA8Y3N-hZa79KjvMAOBumw8FLY4ks5uqNnxgaJpZM4XyhAl .

--

-- Matt Coles

colesmj avatar Oct 30 '18 23:10 colesmj

Okay

shambho avatar Oct 30 '18 23:10 shambho

Looks like this issue is not valid anymore, this threat has been repurposed and handleCrashes prop is now unused. Can we close this?

nineinchnick avatar Oct 04 '20 11:10 nineinchnick

Stale issue as indicated earlier.

colesmj avatar Jun 25 '23 14:06 colesmj