Bug 1106954

Summary:	RFE: Randomize hardware addresses when scanning for networks
Product:	[Fedora] Fedora	Reporter:	Miloslav Trmač <mitr>
Component:	kernel	Assignee:	Kernel Maintainer List <kernel-maint>
Status:	CLOSED UPSTREAM	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	rawhide	CC:	dcbw, gansalmon, itamar, jonathan, kernel-maint, linville, madhu.chinakonda, mchehab
Target Milestone:	---	Keywords:	FutureFeature
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2017-05-02 14:01:23 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Miloslav Trmač 2014-06-09 18:44:15 UTC

"In iOS8, Wi-Fi scanning behavior has changed to use random, locally administrated MAC addresses

Probe requests (management frame sub-type 0x4)
Probe responses (management frame sub-type 0x5)"
(e.g. http://arstechnica.com/apple/2014/06/ios8-to-stymie-trackers-and-marketers-with-mac-address-randomization/ )

This sounds like an interesting privacy enhancement and worth adding.  (The currently cited case of shops monitoring all visitors is not as applicable to  laptops/desktops as to phones, but the same kind of monitoring can be done for global tracking, e.g. from workplace to mall or airport to airport, or for targeting a specific individual.)

(Filing against NetworkManager as the top-level component because I don’t know the underlying architecture well enough, feel free to reassign.)

Comment 1 Dan Williams 2014-06-09 19:03:11 UTC

Yeah, I've seen the posts on the wpa_supplicant lists about this, and it's much more a driver/supplicant thing than NM at this point.  Upstream wpa_supplicant has indicated that they would like to default to this behavior once it has kernel support.  So I'll move the bug over to the kernel for that.

http://lists.shmoo.com/pipermail/hostap/2014-June/030405.html

Comment 2 Josh Boyer 2014-06-09 19:19:31 UTC

Are you using this as a tracking bug?  This really needs to happen upstream.

Comment 3 John W. Linville 2014-06-09 19:27:57 UTC

At first glance, this seems interesting...I'll Cc Johannes to see what he thinks -- he probably already has an answer in mind... :-)

Comment 4 Johannes Berg 2014-06-10 10:32:11 UTC

As Dan said, there was a discussion on the hostap list, which was probably not the appropriate venue but IMHO neither is RH's bugzilla, so I'll not track it here.

In general, I see no big issues with this, but programming the hardware for it might be tricky and will likely need driver support.