[arch-projects] [netctl][PATCH] Disable IPv6 on IP6=no
--- It was proposed to disable IPv6 entirely in case of "IP6=no". Currently we only disable processing of router advertisements. This looks OK to me, but I would really like to get some feedback on this one, as I cannot extensively test the consequences. src/lib/ip | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/lib/ip b/src/lib/ip index 66aab56..9600504 100644 --- a/src/lib/ip +++ b/src/lib/ip @@ -47,11 +47,12 @@ ip_set() { case "$IP6" in dhcp*|stateless|static) [[ -d "/proc/sys/net/ipv6" ]] || modprobe ipv6 + sysctl -q -w "net.ipv6.conf.$interface_sysctl.disable_ipv6=0" [[ $IP6 == "static" ]] sysctl -q -w "net.ipv6.conf.$interface_sysctl.accept_ra=$?" ;; no) - [[ -d "/proc/sys/net/ipv6" ]] && sysctl -q -w "net.ipv6.conf.$interface_sysctl.accept_ra=0" + [[ -d "/proc/sys/net/ipv6" ]] && sysctl -q -w "net.ipv6.conf.$interface_sysctl.disable_ipv6=0" ;; "") # undefined IP6 does not prevent RA's from being received -> nop ;; -- 2.0.1
On Sat, Jul 5, 2014 at 4:01 PM, Jouke Witteveen <j.witteveen@gmail.com> wrote:
--- It was proposed to disable IPv6 entirely in case of "IP6=no". Currently we only disable processing of router advertisements. This looks OK to me, but I would really like to get some feedback on this one, as I cannot extensively test the consequences. src/lib/ip | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/lib/ip b/src/lib/ip index 66aab56..9600504 100644 --- a/src/lib/ip +++ b/src/lib/ip @@ -47,11 +47,12 @@ ip_set() { case "$IP6" in dhcp*|stateless|static) [[ -d "/proc/sys/net/ipv6" ]] || modprobe ipv6 + sysctl -q -w "net.ipv6.conf.$interface_sysctl.disable_ipv6=0" [[ $IP6 == "static" ]] sysctl -q -w "net.ipv6.conf.$interface_sysctl.accept_ra=$?" ;; no) - [[ -d "/proc/sys/net/ipv6" ]] && sysctl -q -w "net.ipv6.conf.$interface_sysctl.accept_ra=0" + [[ -d "/proc/sys/net/ipv6" ]] && sysctl -q -w "net.ipv6.conf.$interface_sysctl.disable_ipv6=0"
That would be disable_ipv6=1, of course.
;; "") # undefined IP6 does not prevent RA's from being received -> nop ;; -- 2.0.1
participants (1)
-
Jouke Witteveen