b10-dhcp6 doesn't not bind on Scope:Global inet6 address.
Angelo Failla
pallotron at fb.com
Fri Mar 14 12:15:10 UTC 2014
On a side note I think you guys should make the v6 server binding by default to ALL configured v6 addresses…
At least just to make it consistent with the way the v4 server works.
The decision is on you, just providing some user feedback :)
--
Angelo Failla
Site Reliability Operations Dublin
Pallotron at fb.com
On 3/14/14, 12:04 PM, "Marcin Siodelski" <marcin at isc.org<mailto:marcin at isc.org>> wrote:
Angelo,
Thanks for your email.
Kea has support for binding socket to unicast address of your choice. It is documented in bind10-guide - section: 18.2.3. Unicast traffic support.
Also, the corresponding ticket about implementing this: bind10.isc.org/ticket/3195
Please, let me know if documentation is not sufficient or it doesn't work as expected and doesn't solve the problem in your network.
Marcin
On 14/03/14 12:30, Angelo Failla wrote:
Hi,
I’m writing to report a problem with bind10-dhcp6 component.
It seems like b10-dhcp6 only binds to ff02::1:2 (the local dhcp multicast address) and the Link IPv6 address but *NOT* the Global one.
See output of `ifconfig eth0` on my machine:
$ ifconfig eth0
eth0 Link encap:Ethernet HWaddr 00:02:C9:DD:0D:5E
inet addr:10.35.139.79 Bcast:10.35.139.255 Mask:255.255.255.0
inet6 addr: fe80::202:c9ff:fedd:d5e/64 Scope:Link
inet6 addr: 2401:db00:3011:b:face:0:3b:0/64 Scope:Global
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:21598559 errors:0 dropped:0 overruns:0 frame:0
TX packets:18807156 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:27172706152 (25.3 GiB) TX bytes:18380499794 (17.1 GiB)
Here you can see it’s not binding on 2401:db00:3011:b:face:0:3b:0/64 (Scope:Global) as it should:
$ sudo netstat -tunap | grep dhcp6
udp 0 0 ff02::1:2:547 :::* 31292/b10-dhcp6
udp 0 0 fe80::202:c9ff:fedd:d5e:547 :::* 31292/b10-dhcp6
The rack switches in our network are configured to relay dhcpv6 traffic to the DHCP servers in our network via their global IPv6 addresses.
Is this a known problem?
PS: I’m running a version of bind10 pulled from your git repo HEAD on the 11th of February 2014.
--
Angelo Failla
Site Reliability Operations Dublin
Pallotron at fb.com<mailto:Pallotron at fb.com>
_______________________________________________
bind10-dhcp mailing list
bind10-dhcp at lists.isc.org<mailto:bind10-dhcp at lists.isc.org>https://lists.isc.org/mailman/listinfo/bind10-dhcp<https://urldefense.proofpoint.com/v1/url?u=https://lists.isc.org/mailman/listinfo/bind10-dhcp&k=ZVNjlDMF0FElm4dQtryO4A%3D%3D%0A&r=ScT1tCv0IwhfW9O3zK6EHw%3D%3D%0A&m=%2Fm6jMGu0l3U0Dn2vKqc1Cwqm1P5a6N1hfZ9yoJyvPDA%3D%0A&s=2e5b51ab22c64e9ea6520a37b2f682392053bd23e646c0e7c3c2f5555eb35311>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/bind10-dhcp/attachments/20140314/2fcfc8d3/attachment-0001.html>
More information about the bind10-dhcp
mailing list