After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 741773 - [PATCH] core: force disable_ipv6=0 when turning on userspace IPv6LL
[PATCH] core: force disable_ipv6=0 when turning on userspace IPv6LL
Status: RESOLVED FIXED
Product: NetworkManager
Classification: Platform
Component: IP and DNS config
unspecified
Other Linux
: Normal normal
: ---
Assigned To: NetworkManager maintainer(s)
NetworkManager maintainer(s)
Depends on:
Blocks:
 
 
Reported: 2014-12-19 17:17 UTC by Dan Williams
Modified: 2015-01-12 15:51 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
0001-core-force-disable_ipv6-0-when-turning-on-userspace-.patch (1.71 KB, patch)
2014-12-19 17:17 UTC, Dan Williams
reviewed Details | Review

Description Dan Williams 2014-12-19 17:17:31 UTC
Created attachment 293064 [details] [review]
0001-core-force-disable_ipv6-0-when-turning-on-userspace-.patch

If a device assumes a connection without activating a user-requested or
NM-requested connection, then disable_ipv6 is not touched.  When the device
is deactivated, it still isn't touched even though userspace IPv6LL
is enabled.  This could lead to an user-requested activation with
IPv6 configuration, but disable_ipv6=1.

Whenever userspace IPv6LL is turned on, we should also set disable_ipv6=0
to ensure IPv6 can function.  Userspace IPv6LL will ensure that the
interface does not have an address until the user/connection requests
it, which was the only reason that NM touched disable_ipv6 anyway.

fixes:NetworkManager_Test203_testcase_286589
fixes:NetworkManager_Test204_testcase_286590
Comment 1 Pavel Simerda 2014-12-30 09:01:51 UTC
The userspace IPv6LL sounds interesting. Is there documentation for it?
Comment 2 Pavel Simerda 2014-12-30 09:03:04 UTC
Also how well does that play with bug #683206?
Comment 3 Dan Winship 2015-01-05 20:44:00 UTC
Comment on attachment 293064 [details] [review]
0001-core-force-disable_ipv6-0-when-turning-on-userspace-.patch

I think I reviewed this over IRC before Christmas? Anyway, it looks right.
Comment 4 Dan Williams 2015-01-06 18:29:35 UTC
(In reply to comment #1)
> The userspace IPv6LL sounds interesting. Is there documentation for it?

Besides bc91b0f07ada5535427373a4e2050877bcc12218 (ipv6: addrconf: implement address generation modes) I don't think so.  It's literally just a setting to prevent the kernel from creating an IPv6LL address for the interface, so that userspace (eg NM) can handle it instead.
Comment 5 Dan Williams 2015-01-09 19:57:00 UTC
Pushed as 984b0763d95eb503a1694bb90b3c09b221b54a0e
Comment 6 Dan Williams 2015-01-09 23:17:29 UTC
Pushed to nm-1-0 as b652ac3cf3708f7b3efff40ab451340a1e759280