That critical VMware vuln allowed anyone on your network to create new admin users, no creds needed

A critical vulnerability in VMware's vCenter management product allowed any old bod on the same network to remotely create an admin-level user, research by Guardicore Labs has revealed. The astonishing vuln (CVE-2020-3952), details of which were quite spare when VMWare issued a patch last week, was rated by VMware itself as …

[Macs1000]

It's very unique...

No it isn't. It's either unique, or it's not.

[TeeCee]

Re: It's very unique...

That's something he said with one hand behind his back with its fingers crossed and the other one on a piece of wood.

[Mage]

re: very unique

It's not like Hilbert's Hotel. Either it's unique or not. Or very <something> as well as unique.

[RegGuy1]

Re: re: very unique

Hilbert's Hotel is shit. I stayed there once and had to keep moving rooms.

[Reg Reader 1]

Re: re: very unique

Very interesting and, perhaps, unique.

[julian_n]

Re: re: very unique

So did I - my wife got a bit pregnant there.

[Anonymous Coward]

Re: re: very unique

Was that linked to the moving rooms?

Enquiring minds and all that...

[julian_n]

Re: re: very unique

Well the room certainly moved.

[Sam Liddicott]

Are these bugs deliberate?

It's clearly a programmer error - it's not valid to call that function without valid credentials!

At least that's the sort of response I get when I report bugs.

I reported today how bash's printf %q format can leave a dangling unused backslash which voids the whole safety benefit of %q

Apparently it's a programmer error to expect to use %q as advertised.

It's not safe to use a truncating size specifier with %q e.g. %.8q

It could be made safe, but why bother for "a programmer error"?

I don't think these sorts of bugs are deliberate but I know others do.

[Anonymous Coward]

Wow, I am sure it was not a backdoor type thing

Yeah, it is perfect coincidence that each time some bug is discovered, it is always some kind of bug allow remote access control, and I am mean it is just a perfect coincidence because Oracle/Microsoft/Intel/VMware/AMD will never allow XYZ to put back door into their products.

Expect a flood of such backddors to come out in the next few months.

Og my god, NSA is spying us, American heroic companies, please secure my IT against NSA spying.

Giggling!

[Sir Runcible Spoon]

Re: Wow, I am sure it was not a backdoor type thing

"Giggling"

Medication time perhaps?

[Anonymous Coward]

Re: Wow, I am sure it was not a backdoor type thing

That IS his medication.

[chuBb.]

Re: Wow, I am sure it was not a backdoor type thing

Couldn't be that it's more interesting than your common or garden buffer overflow, or lack of input validation/Web server running with too many privileges... that and it's staggering in the fact it works as long as your network connected

[Lorribot]

It is (very?) unique to version 6.7. Something the article studiously or ignorantly fails to mention.

[Nate Amsden]

more so seems to only affect 6.7 systems that were upgraded from earlier versions(which is probably many of them, mine included), doesn't affect new installations.

https://kb.vmware.com/s/article/78543

[amanfromMars 1]

What's IT to Be ... Incoming Hostilities or Outrageous Hospitality ‽

a "malicious actor with network access to port 389 on an affected vmdir deployment may be able to extract highly sensitive information such as administrative account credentials".

Which further translates in vulnerable circles, are extracting highly sensitive information such as administrative account credentials in the here and now for future vmdir deployments. Past and present instances are of no effective consequence whenever there are proactive malicious actors with special stealthy access to networks.

Deny it if you don't believe it, however, simply watch, listen and learn how AI and IT easily proves and shows such things to be perfectly true.

Some think Catastrophic Exploitable Vulnerability, A.N.Others Almighty Useful Facility.