No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

classic Classic list List threaded Threaded
32 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

Bill Shirley-2
I don't know if it is affecting anything but you still have the bridge defined
in /etc/network/interfaces.  Also, you still have wlp2s0 on 10.1.1.0/24. Do
you want to define the wireless?

So 'blue' is to use 'yellow' to get to the internet.  Do you have iptables on
'yellow' set up to masquerade for 10.1.1.0/24

Bill

On 1/27/2018 7:10 PM, A wrote:

I meant to include the current /etc/network/interfaces.  Below.  However, in the meantime I'm getting different results.  Perhaps due to the fact that I changed "static" to "manuel".  I also deleted the bridge.  Now, both machines can ping each other on the wired interface, and  yellow can still ping the 'Net.  Blue cannot ping the Internet.

It's unclear if the bridge-utils is truly out of the picture, despite the tools reporting so.  I suspect this may still be an issue that will come back later when I retry.

PING blue (10.1.1.14) 56(84) bytes of data.
64 bytes from blue (10.1.1.14): icmp_seq=1 ttl=64 time=0.353 ms

ping google.com
PING google.com (172.217.11.78) 56(84) bytes of data.
64 bytes from lax17s34-in-f14.1e100.net (172.217.11.78): icmp_seq=1 ttl=56 time=23.4 ms

# ping yellow
PING yellow (127.0.1.1) 56(84) bytes of data.
64 bytes from yellow (127.0.1.1): icmp_seq=1 ttl=64 time=0.088 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.088/0.088/0.088/0.000 ms

# ping router
PING yellow (10.1.1.1) 56(84) bytes of data.
64 bytes from yellow (10.1.1.1): icmp_seq=1 ttl=64 time=0.077 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`

/etc/network/interfaces:


auto lo
iface lo inet loopback
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN

auto enp6s0

iface enp6s0 inet dhcp
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN


auto enp4s5
allow-hotplug enp4s5
iface enp4s5 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

    # Before we can change the address we have to bring the face down
    pre-up ip link set enp4s5 down

# I don't know if I can use two ups.  I don't know the right way to do this.
    up ip address 0.0.0.0 dev enp4s5
    up ip link set enp4s5 up
    down ip link set enp4s5 down


allow-hotplug wlp2s0
iface wlp2s0 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

# I don't know if I can use two pre-ups.  I don't know the right way to do this.
    pre-up ip link set wlp2s0 down
    pre-up ip address 0.0.0.0 dev wlp2s0
    up ip link set wlp2s0 up
    down ip link set wlp2s0 down

    wireless-mode master
    wireless-essid XXXXXXXXXXXXXX
    wireless-channel 1
    wpa-ssid XXXXXXXXXXXXXX
    wpa-psk XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    gateway 10.1.1.1



auto br0
iface br0 inet static
    address 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255
    bridge-ports enp4s5 wlp2s0




On 01/27/2018 03:20 PM, A wrote:



On 01/27/2018 01:28 PM, Simon Hobson wrote:
A <[hidden email]> wrote:

I did originally have two separate subnets with a /28 CIDR, but I was unable to reach the Internet from blue and someone suggested I have one subnet in order to act as a typical home router. So I reconfigured everything and it's now borked worse than it was.  Said person disappeared shortly after of course.

There is no commercial router.  Yellow is the router, gateway, access point, dhcp server, dns server, firewall (iptables) and more.

OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.

Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.

Thus devices on the wired network cannot talk to devices on the WiFi and vice-verca.

Bear in mind that I've not used WiFi in this manner (I'm used to using external APs), so I am unsure of some of the details. If you want to run a single unified network then you will need to create a bridge, and put the wired and wireless adapters into that bridge - and put your address 10.1.1.1/24 onto the bridge. You will then have one network, and the bridge software will pass packets between them, as well as keeping track of which clients are in which network segment.

I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up. 



As far as (almost) all software on your box is concerned, you just deal with one interface (the bridge, eg br0). What I am unsure about is how dhcpd behaves in this case - hopefully someone who's run this setup can comment ? I would hope that it would use the bridge interface in the same manner as it would use a "real" one, but there can be some subtle differences.

I had it working with the bridge at one point - each box could ping the other on both wired & wireless, but blue couldn't reach the Internet.  Lets see if we can put it back.

.... blue now receives an IP of .14; neither machine can ping the other, though each can ping its own assigned IP.

# brctl show
bridge name    bridge id        STP enabled    interfaces
br0        8000.7085c23b1324    no                   enp4s5
                                                                         enp6s0


$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp4s5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether c8:3a:35:da:42:72 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global enp4s5
       valid_lft forever preferred_lft forever
3: enp6s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet [xx.xx.xx.xx]/20 brd 255.255.255.255 scope global enp6s0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever
4: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether f0:7d:68:c1:b4:13 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.10/24 brd 10.1.1.255 scope global wlp2s0
       valid_lft forever preferred_lft forever
5: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global br0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever

# ip route
default via 174.xx.yy.1 dev enp6s0
10.1.1.0/24 dev enp4s5  proto kernel  scope link  src 10.1.1.1
174.xx.yy.0/20 dev enp6s0  proto kernel  scope link  src 174.xx.bb.zz




_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
A
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

A



On 01/27/2018 04:22 PM, Bill Shirley wrote:
I don't know if it is affecting anything but you still have the bridge defined
in /etc/network/interfaces.  Also, you still have wlp2s0 on 10.1.1.0/24. Do
you want to define the wireless?

Yes.  I just finished responding to your prior email. I need to run out for a few minutes before it gets dark.  Back in 30.


So 'blue' is to use 'yellow' to get to the internet.  Do you have iptables on
'yellow' set up to masquerade for 10.1.1.0/24

This is what I have at the moment so far.  I see an error with FOUR.  Back in 30.


-A PREROUTING -i enp6s0 -p tcp -m multiport --dports 80,443 -j DNAT --to-destination 10.1.1.1
-A FORWARD -i enp6s0 -o wlp2s0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment THREE -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -i enp4s5 -o enp6s0 -m comment --comment FOUR -j ACCEPT
-A FORWARD ! -d 10.1.1.0/32 -o enp6s0 -m comment --comment SEVEN -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -o br0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment EIGHT -j ACCEPT


Bill

On 1/27/2018 7:10 PM, A wrote:

I meant to include the current /etc/network/interfaces.  Below.  However, in the meantime I'm getting different results.  Perhaps due to the fact that I changed "static" to "manuel".  I also deleted the bridge.  Now, both machines can ping each other on the wired interface, and  yellow can still ping the 'Net.  Blue cannot ping the Internet.

It's unclear if the bridge-utils is truly out of the picture, despite the tools reporting so.  I suspect this may still be an issue that will come back later when I retry.

PING blue (10.1.1.14) 56(84) bytes of data.
64 bytes from blue (10.1.1.14): icmp_seq=1 ttl=64 time=0.353 ms

ping google.com
PING google.com (172.217.11.78) 56(84) bytes of data.
64 bytes from lax17s34-in-f14.1e100.net (172.217.11.78): icmp_seq=1 ttl=56 time=23.4 ms

# ping yellow
PING yellow (127.0.1.1) 56(84) bytes of data.
64 bytes from yellow (127.0.1.1): icmp_seq=1 ttl=64 time=0.088 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.088/0.088/0.088/0.000 ms

# ping router
PING yellow (10.1.1.1) 56(84) bytes of data.
64 bytes from yellow (10.1.1.1): icmp_seq=1 ttl=64 time=0.077 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`

/etc/network/interfaces:


auto lo
iface lo inet loopback
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN

auto enp6s0

iface enp6s0 inet dhcp
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN


auto enp4s5
allow-hotplug enp4s5
iface enp4s5 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

    # Before we can change the address we have to bring the face down
    pre-up ip link set enp4s5 down

# I don't know if I can use two ups.  I don't know the right way to do this.
    up ip address 0.0.0.0 dev enp4s5
    up ip link set enp4s5 up
    down ip link set enp4s5 down


allow-hotplug wlp2s0
iface wlp2s0 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

# I don't know if I can use two pre-ups.  I don't know the right way to do this.
    pre-up ip link set wlp2s0 down
    pre-up ip address 0.0.0.0 dev wlp2s0
    up ip link set wlp2s0 up
    down ip link set wlp2s0 down

    wireless-mode master
    wireless-essid XXXXXXXXXXXXXX
    wireless-channel 1
    wpa-ssid XXXXXXXXXXXXXX
    wpa-psk XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    gateway 10.1.1.1



auto br0
iface br0 inet static
    address 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255
    bridge-ports enp4s5 wlp2s0




On 01/27/2018 03:20 PM, A wrote:



On 01/27/2018 01:28 PM, Simon Hobson wrote:
A <[hidden email]> wrote:

I did originally have two separate subnets with a /28 CIDR, but I was unable to reach the Internet from blue and someone suggested I have one subnet in order to act as a typical home router. So I reconfigured everything and it's now borked worse than it was.  Said person disappeared shortly after of course.

There is no commercial router.  Yellow is the router, gateway, access point, dhcp server, dns server, firewall (iptables) and more.

OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.

Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.

Thus devices on the wired network cannot talk to devices on the WiFi and vice-verca.

Bear in mind that I've not used WiFi in this manner (I'm used to using external APs), so I am unsure of some of the details. If you want to run a single unified network then you will need to create a bridge, and put the wired and wireless adapters into that bridge - and put your address 10.1.1.1/24 onto the bridge. You will then have one network, and the bridge software will pass packets between them, as well as keeping track of which clients are in which network segment.

I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up. 



As far as (almost) all software on your box is concerned, you just deal with one interface (the bridge, eg br0). What I am unsure about is how dhcpd behaves in this case - hopefully someone who's run this setup can comment ? I would hope that it would use the bridge interface in the same manner as it would use a "real" one, but there can be some subtle differences.

I had it working with the bridge at one point - each box could ping the other on both wired & wireless, but blue couldn't reach the Internet.  Lets see if we can put it back.

.... blue now receives an IP of .14; neither machine can ping the other, though each can ping its own assigned IP.

# brctl show
bridge name    bridge id        STP enabled    interfaces
br0        8000.7085c23b1324    no                   enp4s5
                                                                         enp6s0


$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp4s5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether c8:3a:35:da:42:72 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global enp4s5
       valid_lft forever preferred_lft forever
3: enp6s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet [xx.xx.xx.xx]/20 brd 255.255.255.255 scope global enp6s0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever
4: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether f0:7d:68:c1:b4:13 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.10/24 brd 10.1.1.255 scope global wlp2s0
       valid_lft forever preferred_lft forever
5: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global br0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever

# ip route
default via 174.xx.yy.1 dev enp6s0
10.1.1.0/24 dev enp4s5  proto kernel  scope link  src 10.1.1.1
174.xx.yy.0/20 dev enp6s0  proto kernel  scope link  src 174.xx.bb.zz




_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

Bill Shirley-2
You're looking in the wrong table.  You want 'iptables -t nat'.
You need an entry something like:
iptables -t nat -A POSTROUTING -o enp6s0 -s 10.1.1.0/24 -j MASQUERADE

I don't craft my iptables by hand; I use Shorewall.

Bill

On 1/27/2018 7:52 PM, A wrote:



On 01/27/2018 04:22 PM, Bill Shirley wrote:
I don't know if it is affecting anything but you still have the bridge defined
in /etc/network/interfaces.  Also, you still have wlp2s0 on 10.1.1.0/24. Do
you want to define the wireless?

Yes.  I just finished responding to your prior email. I need to run out for a few minutes before it gets dark.  Back in 30.


So 'blue' is to use 'yellow' to get to the internet.  Do you have iptables on
'yellow' set up to masquerade for 10.1.1.0/24

This is what I have at the moment so far.  I see an error with FOUR.  Back in 30.


-A PREROUTING -i enp6s0 -p tcp -m multiport --dports 80,443 -j DNAT --to-destination 10.1.1.1
-A FORWARD -i enp6s0 -o wlp2s0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment THREE -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -i enp4s5 -o enp6s0 -m comment --comment FOUR -j ACCEPT
-A FORWARD ! -d 10.1.1.0/32 -o enp6s0 -m comment --comment SEVEN -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -o br0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment EIGHT -j ACCEPT


Bill

On 1/27/2018 7:10 PM, A wrote:

I meant to include the current /etc/network/interfaces.  Below.  However, in the meantime I'm getting different results.  Perhaps due to the fact that I changed "static" to "manuel".  I also deleted the bridge.  Now, both machines can ping each other on the wired interface, and  yellow can still ping the 'Net.  Blue cannot ping the Internet.

It's unclear if the bridge-utils is truly out of the picture, despite the tools reporting so.  I suspect this may still be an issue that will come back later when I retry.

PING blue (10.1.1.14) 56(84) bytes of data.
64 bytes from blue (10.1.1.14): icmp_seq=1 ttl=64 time=0.353 ms

ping google.com
PING google.com (172.217.11.78) 56(84) bytes of data.
64 bytes from lax17s34-in-f14.1e100.net (172.217.11.78): icmp_seq=1 ttl=56 time=23.4 ms

# ping yellow
PING yellow (127.0.1.1) 56(84) bytes of data.
64 bytes from yellow (127.0.1.1): icmp_seq=1 ttl=64 time=0.088 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.088/0.088/0.088/0.000 ms

# ping router
PING yellow (10.1.1.1) 56(84) bytes of data.
64 bytes from yellow (10.1.1.1): icmp_seq=1 ttl=64 time=0.077 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`

/etc/network/interfaces:


auto lo
iface lo inet loopback
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN

auto enp6s0

iface enp6s0 inet dhcp
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN


auto enp4s5
allow-hotplug enp4s5
iface enp4s5 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

    # Before we can change the address we have to bring the face down
    pre-up ip link set enp4s5 down

# I don't know if I can use two ups.  I don't know the right way to do this.
    up ip address 0.0.0.0 dev enp4s5
    up ip link set enp4s5 up
    down ip link set enp4s5 down


allow-hotplug wlp2s0
iface wlp2s0 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

# I don't know if I can use two pre-ups.  I don't know the right way to do this.
    pre-up ip link set wlp2s0 down
    pre-up ip address 0.0.0.0 dev wlp2s0
    up ip link set wlp2s0 up
    down ip link set wlp2s0 down

    wireless-mode master
    wireless-essid XXXXXXXXXXXXXX
    wireless-channel 1
    wpa-ssid XXXXXXXXXXXXXX
    wpa-psk XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    gateway 10.1.1.1



auto br0
iface br0 inet static
    address 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255
    bridge-ports enp4s5 wlp2s0




On 01/27/2018 03:20 PM, A wrote:



On 01/27/2018 01:28 PM, Simon Hobson wrote:
A <[hidden email]> wrote:

I did originally have two separate subnets with a /28 CIDR, but I was unable to reach the Internet from blue and someone suggested I have one subnet in order to act as a typical home router. So I reconfigured everything and it's now borked worse than it was.  Said person disappeared shortly after of course.

There is no commercial router.  Yellow is the router, gateway, access point, dhcp server, dns server, firewall (iptables) and more.

OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.

Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.

Thus devices on the wired network cannot talk to devices on the WiFi and vice-verca.

Bear in mind that I've not used WiFi in this manner (I'm used to using external APs), so I am unsure of some of the details. If you want to run a single unified network then you will need to create a bridge, and put the wired and wireless adapters into that bridge - and put your address 10.1.1.1/24 onto the bridge. You will then have one network, and the bridge software will pass packets between them, as well as keeping track of which clients are in which network segment.

I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up. 



As far as (almost) all software on your box is concerned, you just deal with one interface (the bridge, eg br0). What I am unsure about is how dhcpd behaves in this case - hopefully someone who's run this setup can comment ? I would hope that it would use the bridge interface in the same manner as it would use a "real" one, but there can be some subtle differences.

I had it working with the bridge at one point - each box could ping the other on both wired & wireless, but blue couldn't reach the Internet.  Lets see if we can put it back.

.... blue now receives an IP of .14; neither machine can ping the other, though each can ping its own assigned IP.

# brctl show
bridge name    bridge id        STP enabled    interfaces
br0        8000.7085c23b1324    no                   enp4s5
                                                                         enp6s0


$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp4s5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether c8:3a:35:da:42:72 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global enp4s5
       valid_lft forever preferred_lft forever
3: enp6s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet [xx.xx.xx.xx]/20 brd 255.255.255.255 scope global enp6s0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever
4: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether f0:7d:68:c1:b4:13 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.10/24 brd 10.1.1.255 scope global wlp2s0
       valid_lft forever preferred_lft forever
5: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global br0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever

# ip route
default via 174.xx.yy.1 dev enp6s0
10.1.1.0/24 dev enp4s5  proto kernel  scope link  src 10.1.1.1
174.xx.yy.0/20 dev enp6s0  proto kernel  scope link  src 174.xx.bb.zz




_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

Bill Shirley-2
Also in the entries you've shown: FOUR, SEVEN, and EIGHT are
using /32 which I think should be /24 I would think.  EIGHT is a
rule for the bridge, do you need that?

Bill

On 1/27/2018 9:24 PM, Bill Shirley wrote:
You're looking in the wrong table.  You want 'iptables -t nat'.
You need an entry something like:
iptables -t nat -A POSTROUTING -o enp6s0 -s 10.1.1.0/24 -j MASQUERADE

I don't craft my iptables by hand; I use Shorewall.

Bill

On 1/27/2018 7:52 PM, A wrote:



On 01/27/2018 04:22 PM, Bill Shirley wrote:
I don't know if it is affecting anything but you still have the bridge defined
in /etc/network/interfaces.  Also, you still have wlp2s0 on 10.1.1.0/24. Do
you want to define the wireless?

Yes.  I just finished responding to your prior email. I need to run out for a few minutes before it gets dark.  Back in 30.


So 'blue' is to use 'yellow' to get to the internet.  Do you have iptables on
'yellow' set up to masquerade for 10.1.1.0/24

This is what I have at the moment so far.  I see an error with FOUR.  Back in 30.


-A PREROUTING -i enp6s0 -p tcp -m multiport --dports 80,443 -j DNAT --to-destination 10.1.1.1
-A FORWARD -i enp6s0 -o wlp2s0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment THREE -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -i enp4s5 -o enp6s0 -m comment --comment FOUR -j ACCEPT
-A FORWARD ! -d 10.1.1.0/32 -o enp6s0 -m comment --comment SEVEN -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -o br0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment EIGHT -j ACCEPT


Bill

On 1/27/2018 7:10 PM, A wrote:

I meant to include the current /etc/network/interfaces.  Below.  However, in the meantime I'm getting different results.  Perhaps due to the fact that I changed "static" to "manuel".  I also deleted the bridge.  Now, both machines can ping each other on the wired interface, and  yellow can still ping the 'Net.  Blue cannot ping the Internet.

It's unclear if the bridge-utils is truly out of the picture, despite the tools reporting so.  I suspect this may still be an issue that will come back later when I retry.

PING blue (10.1.1.14) 56(84) bytes of data.
64 bytes from blue (10.1.1.14): icmp_seq=1 ttl=64 time=0.353 ms

ping google.com
PING google.com (172.217.11.78) 56(84) bytes of data.
64 bytes from lax17s34-in-f14.1e100.net (172.217.11.78): icmp_seq=1 ttl=56 time=23.4 ms

# ping yellow
PING yellow (127.0.1.1) 56(84) bytes of data.
64 bytes from yellow (127.0.1.1): icmp_seq=1 ttl=64 time=0.088 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.088/0.088/0.088/0.000 ms

# ping router
PING yellow (10.1.1.1) 56(84) bytes of data.
64 bytes from yellow (10.1.1.1): icmp_seq=1 ttl=64 time=0.077 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`

/etc/network/interfaces:


auto lo
iface lo inet loopback
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN

auto enp6s0

iface enp6s0 inet dhcp
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN


auto enp4s5
allow-hotplug enp4s5
iface enp4s5 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

    # Before we can change the address we have to bring the face down
    pre-up ip link set enp4s5 down

# I don't know if I can use two ups.  I don't know the right way to do this.
    up ip address 0.0.0.0 dev enp4s5
    up ip link set enp4s5 up
    down ip link set enp4s5 down


allow-hotplug wlp2s0
iface wlp2s0 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

# I don't know if I can use two pre-ups.  I don't know the right way to do this.
    pre-up ip link set wlp2s0 down
    pre-up ip address 0.0.0.0 dev wlp2s0
    up ip link set wlp2s0 up
    down ip link set wlp2s0 down

    wireless-mode master
    wireless-essid XXXXXXXXXXXXXX
    wireless-channel 1
    wpa-ssid XXXXXXXXXXXXXX
    wpa-psk XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    gateway 10.1.1.1



auto br0
iface br0 inet static
    address 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255
    bridge-ports enp4s5 wlp2s0




On 01/27/2018 03:20 PM, A wrote:



On 01/27/2018 01:28 PM, Simon Hobson wrote:
A <[hidden email]> wrote:

I did originally have two separate subnets with a /28 CIDR, but I was unable to reach the Internet from blue and someone suggested I have one subnet in order to act as a typical home router. So I reconfigured everything and it's now borked worse than it was.  Said person disappeared shortly after of course.

There is no commercial router.  Yellow is the router, gateway, access point, dhcp server, dns server, firewall (iptables) and more.

OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.

Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.

Thus devices on the wired network cannot talk to devices on the WiFi and vice-verca.

Bear in mind that I've not used WiFi in this manner (I'm used to using external APs), so I am unsure of some of the details. If you want to run a single unified network then you will need to create a bridge, and put the wired and wireless adapters into that bridge - and put your address 10.1.1.1/24 onto the bridge. You will then have one network, and the bridge software will pass packets between them, as well as keeping track of which clients are in which network segment.

I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up. 



As far as (almost) all software on your box is concerned, you just deal with one interface (the bridge, eg br0). What I am unsure about is how dhcpd behaves in this case - hopefully someone who's run this setup can comment ? I would hope that it would use the bridge interface in the same manner as it would use a "real" one, but there can be some subtle differences.

I had it working with the bridge at one point - each box could ping the other on both wired & wireless, but blue couldn't reach the Internet.  Lets see if we can put it back.

.... blue now receives an IP of .14; neither machine can ping the other, though each can ping its own assigned IP.

# brctl show
bridge name    bridge id        STP enabled    interfaces
br0        8000.7085c23b1324    no                   enp4s5
                                                                         enp6s0


$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp4s5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether c8:3a:35:da:42:72 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global enp4s5
       valid_lft forever preferred_lft forever
3: enp6s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet [xx.xx.xx.xx]/20 brd 255.255.255.255 scope global enp6s0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever
4: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether f0:7d:68:c1:b4:13 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.10/24 brd 10.1.1.255 scope global wlp2s0
       valid_lft forever preferred_lft forever
5: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global br0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever

# ip route
default via 174.xx.yy.1 dev enp6s0
10.1.1.0/24 dev enp4s5  proto kernel  scope link  src 10.1.1.1
174.xx.yy.0/20 dev enp6s0  proto kernel  scope link  src 174.xx.bb.zz




_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
A
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

A
In reply to this post by Bill Shirley-2



On 01/27/2018 06:24 PM, Bill Shirley wrote:
You're looking in the wrong table.  You want 'iptables -t nat'.
You need an entry something like:
iptables -t nat -A POSTROUTING -o enp6s0 -s 10.1.1.0/24 -j MASQUERADE

Forgot that one, sorry.  I've been showing you grepped output.  This is from the nat table.

-A POSTROUTING -s 10.1.1.0/32 -o enp6s0 -j MASQUERADE



I don't craft my iptables by hand; I use Shorewall.
Bill

On 1/27/2018 7:52 PM, A wrote:



On 01/27/2018 04:22 PM, Bill Shirley wrote:
I don't know if it is affecting anything but you still have the bridge defined
in /etc/network/interfaces.  Also, you still have wlp2s0 on 10.1.1.0/24. Do
you want to define the wireless?

Yes.  I just finished responding to your prior email. I need to run out for a few minutes before it gets dark.  Back in 30.


So 'blue' is to use 'yellow' to get to the internet.  Do you have iptables on
'yellow' set up to masquerade for 10.1.1.0/24

This is what I have at the moment so far.  I see an error with FOUR.  Back in 30.


-A PREROUTING -i enp6s0 -p tcp -m multiport --dports 80,443 -j DNAT --to-destination 10.1.1.1
-A FORWARD -i enp6s0 -o wlp2s0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment THREE -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -i enp4s5 -o enp6s0 -m comment --comment FOUR -j ACCEPT
-A FORWARD ! -d 10.1.1.0/32 -o enp6s0 -m comment --comment SEVEN -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -o br0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment EIGHT -j ACCEPT


Bill

On 1/27/2018 7:10 PM, A wrote:

I meant to include the current /etc/network/interfaces.  Below.  However, in the meantime I'm getting different results.  Perhaps due to the fact that I changed "static" to "manuel".  I also deleted the bridge.  Now, both machines can ping each other on the wired interface, and  yellow can still ping the 'Net.  Blue cannot ping the Internet.

It's unclear if the bridge-utils is truly out of the picture, despite the tools reporting so.  I suspect this may still be an issue that will come back later when I retry.

PING blue (10.1.1.14) 56(84) bytes of data.
64 bytes from blue (10.1.1.14): icmp_seq=1 ttl=64 time=0.353 ms

ping google.com
PING google.com (172.217.11.78) 56(84) bytes of data.
64 bytes from lax17s34-in-f14.1e100.net (172.217.11.78): icmp_seq=1 ttl=56 time=23.4 ms

# ping yellow
PING yellow (127.0.1.1) 56(84) bytes of data.
64 bytes from yellow (127.0.1.1): icmp_seq=1 ttl=64 time=0.088 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.088/0.088/0.088/0.000 ms

# ping router
PING yellow (10.1.1.1) 56(84) bytes of data.
64 bytes from yellow (10.1.1.1): icmp_seq=1 ttl=64 time=0.077 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`

/etc/network/interfaces:


auto lo
iface lo inet loopback
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN

auto enp6s0

iface enp6s0 inet dhcp
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN


auto enp4s5
allow-hotplug enp4s5
iface enp4s5 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

    # Before we can change the address we have to bring the face down
    pre-up ip link set enp4s5 down

# I don't know if I can use two ups.  I don't know the right way to do this.
    up ip address 0.0.0.0 dev enp4s5
    up ip link set enp4s5 up
    down ip link set enp4s5 down


allow-hotplug wlp2s0
iface wlp2s0 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

# I don't know if I can use two pre-ups.  I don't know the right way to do this.
    pre-up ip link set wlp2s0 down
    pre-up ip address 0.0.0.0 dev wlp2s0
    up ip link set wlp2s0 up
    down ip link set wlp2s0 down

    wireless-mode master
    wireless-essid XXXXXXXXXXXXXX
    wireless-channel 1
    wpa-ssid XXXXXXXXXXXXXX
    wpa-psk XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    gateway 10.1.1.1



auto br0
iface br0 inet static
    address 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255
    bridge-ports enp4s5 wlp2s0




On 01/27/2018 03:20 PM, A wrote:



On 01/27/2018 01:28 PM, Simon Hobson wrote:
A <[hidden email]> wrote:

I did originally have two separate subnets with a /28 CIDR, but I was unable to reach the Internet from blue and someone suggested I have one subnet in order to act as a typical home router. So I reconfigured everything and it's now borked worse than it was.  Said person disappeared shortly after of course.

There is no commercial router.  Yellow is the router, gateway, access point, dhcp server, dns server, firewall (iptables) and more.

OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.

Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.

Thus devices on the wired network cannot talk to devices on the WiFi and vice-verca.

Bear in mind that I've not used WiFi in this manner (I'm used to using external APs), so I am unsure of some of the details. If you want to run a single unified network then you will need to create a bridge, and put the wired and wireless adapters into that bridge - and put your address 10.1.1.1/24 onto the bridge. You will then have one network, and the bridge software will pass packets between them, as well as keeping track of which clients are in which network segment.

I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up. 



As far as (almost) all software on your box is concerned, you just deal with one interface (the bridge, eg br0). What I am unsure about is how dhcpd behaves in this case - hopefully someone who's run this setup can comment ? I would hope that it would use the bridge interface in the same manner as it would use a "real" one, but there can be some subtle differences.

I had it working with the bridge at one point - each box could ping the other on both wired & wireless, but blue couldn't reach the Internet.  Lets see if we can put it back.

.... blue now receives an IP of .14; neither machine can ping the other, though each can ping its own assigned IP.

# brctl show
bridge name    bridge id        STP enabled    interfaces
br0        8000.7085c23b1324    no                   enp4s5
                                                                         enp6s0


$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp4s5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether c8:3a:35:da:42:72 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global enp4s5
       valid_lft forever preferred_lft forever
3: enp6s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet [xx.xx.xx.xx]/20 brd 255.255.255.255 scope global enp6s0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever
4: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether f0:7d:68:c1:b4:13 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.10/24 brd 10.1.1.255 scope global wlp2s0
       valid_lft forever preferred_lft forever
5: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global br0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever

# ip route
default via 174.xx.yy.1 dev enp6s0
10.1.1.0/24 dev enp4s5  proto kernel  scope link  src 10.1.1.1
174.xx.yy.0/20 dev enp6s0  proto kernel  scope link  src 174.xx.bb.zz




_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
A
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

A
In reply to this post by Bill Shirley-2
It apparently defaults to /32 when one neglects to specify a netmask.

We seem to have moved past dhcp... Take this private so as not to bother others with non-dhcp related issues?

Many thanks to all!!!  Much appreciated!!!

On 01/27/2018 06:31 PM, Bill Shirley wrote:
Also in the entries you've shown: FOUR, SEVEN, and EIGHT are
using /32 which I think should be /24 I would think.  EIGHT is a
rule for the bridge, do you need that?

Bill

On 1/27/2018 9:24 PM, Bill Shirley wrote:
You're looking in the wrong table.  You want 'iptables -t nat'.
You need an entry something like:
iptables -t nat -A POSTROUTING -o enp6s0 -s 10.1.1.0/24 -j MASQUERADE

I don't craft my iptables by hand; I use Shorewall.

Bill

On 1/27/2018 7:52 PM, A wrote:



On 01/27/2018 04:22 PM, Bill Shirley wrote:
I don't know if it is affecting anything but you still have the bridge defined
in /etc/network/interfaces.  Also, you still have wlp2s0 on 10.1.1.0/24. Do
you want to define the wireless?

Yes.  I just finished responding to your prior email. I need to run out for a few minutes before it gets dark.  Back in 30.


So 'blue' is to use 'yellow' to get to the internet.  Do you have iptables on
'yellow' set up to masquerade for 10.1.1.0/24

This is what I have at the moment so far.  I see an error with FOUR.  Back in 30.


-A PREROUTING -i enp6s0 -p tcp -m multiport --dports 80,443 -j DNAT --to-destination 10.1.1.1
-A FORWARD -i enp6s0 -o wlp2s0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment THREE -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -i enp4s5 -o enp6s0 -m comment --comment FOUR -j ACCEPT
-A FORWARD ! -d 10.1.1.0/32 -o enp6s0 -m comment --comment SEVEN -j ACCEPT
-A FORWARD -d 10.1.1.0/32 -o br0 -m conntrack --ctstate RELATED,ESTABLISHED -m comment --comment EIGHT -j ACCEPT


Bill

On 1/27/2018 7:10 PM, A wrote:

I meant to include the current /etc/network/interfaces.  Below.  However, in the meantime I'm getting different results.  Perhaps due to the fact that I changed "static" to "manuel".  I also deleted the bridge.  Now, both machines can ping each other on the wired interface, and  yellow can still ping the 'Net.  Blue cannot ping the Internet.

It's unclear if the bridge-utils is truly out of the picture, despite the tools reporting so.  I suspect this may still be an issue that will come back later when I retry.

PING blue (10.1.1.14) 56(84) bytes of data.
64 bytes from blue (10.1.1.14): icmp_seq=1 ttl=64 time=0.353 ms

ping google.com
PING google.com (172.217.11.78) 56(84) bytes of data.
64 bytes from lax17s34-in-f14.1e100.net (172.217.11.78): icmp_seq=1 ttl=56 time=23.4 ms

# ping yellow
PING yellow (127.0.1.1) 56(84) bytes of data.
64 bytes from yellow (127.0.1.1): icmp_seq=1 ttl=64 time=0.088 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.088/0.088/0.088/0.000 ms

# ping router
PING yellow (10.1.1.1) 56(84) bytes of data.
64 bytes from yellow (10.1.1.1): icmp_seq=1 ttl=64 time=0.077 ms
^C
--- yellow ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`

/etc/network/interfaces:


auto lo
iface lo inet loopback
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN

auto enp6s0

iface enp6s0 inet dhcp
    dns-nameservers 10.1.1.1 8.8.8.8 8.8.4.4 50.23.197.95
    dns-search FQDN


auto enp4s5
allow-hotplug enp4s5
iface enp4s5 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

    # Before we can change the address we have to bring the face down
    pre-up ip link set enp4s5 down

# I don't know if I can use two ups.  I don't know the right way to do this.
    up ip address 0.0.0.0 dev enp4s5
    up ip link set enp4s5 up
    down ip link set enp4s5 down


allow-hotplug wlp2s0
iface wlp2s0 inet manuel
    gateway 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255

# I don't know if I can use two pre-ups.  I don't know the right way to do this.
    pre-up ip link set wlp2s0 down
    pre-up ip address 0.0.0.0 dev wlp2s0
    up ip link set wlp2s0 up
    down ip link set wlp2s0 down

    wireless-mode master
    wireless-essid XXXXXXXXXXXXXX
    wireless-channel 1
    wpa-ssid XXXXXXXXXXXXXX
    wpa-psk XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    gateway 10.1.1.1



auto br0
iface br0 inet static
    address 10.1.1.1
    network 10.1.1.0
    netmask 255.255.255.0
    broadcast 10.1.1.255
    bridge-ports enp4s5 wlp2s0




On 01/27/2018 03:20 PM, A wrote:



On 01/27/2018 01:28 PM, Simon Hobson wrote:
A <[hidden email]> wrote:

I did originally have two separate subnets with a /28 CIDR, but I was unable to reach the Internet from blue and someone suggested I have one subnet in order to act as a typical home router. So I reconfigured everything and it's now borked worse than it was.  Said person disappeared shortly after of course.

There is no commercial router.  Yellow is the router, gateway, access point, dhcp server, dns server, firewall (iptables) and more.

OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.

Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.

Thus devices on the wired network cannot talk to devices on the WiFi and vice-verca.

Bear in mind that I've not used WiFi in this manner (I'm used to using external APs), so I am unsure of some of the details. If you want to run a single unified network then you will need to create a bridge, and put the wired and wireless adapters into that bridge - and put your address 10.1.1.1/24 onto the bridge. You will then have one network, and the bridge software will pass packets between them, as well as keeping track of which clients are in which network segment.

I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up. 



As far as (almost) all software on your box is concerned, you just deal with one interface (the bridge, eg br0). What I am unsure about is how dhcpd behaves in this case - hopefully someone who's run this setup can comment ? I would hope that it would use the bridge interface in the same manner as it would use a "real" one, but there can be some subtle differences.

I had it working with the bridge at one point - each box could ping the other on both wired & wireless, but blue couldn't reach the Internet.  Lets see if we can put it back.

.... blue now receives an IP of .14; neither machine can ping the other, though each can ping its own assigned IP.

# brctl show
bridge name    bridge id        STP enabled    interfaces
br0        8000.7085c23b1324    no                   enp4s5
                                                                         enp6s0


$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp4s5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether c8:3a:35:da:42:72 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global enp4s5
       valid_lft forever preferred_lft forever
3: enp6s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet [xx.xx.xx.xx]/20 brd 255.255.255.255 scope global enp6s0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever
4: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether f0:7d:68:c1:b4:13 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.10/24 brd 10.1.1.255 scope global wlp2s0
       valid_lft forever preferred_lft forever
5: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 70:85:c2:3b:13:24 brd ff:ff:ff:ff:ff:ff
    inet 10.1.1.1/24 brd 10.1.1.255 scope global br0
       valid_lft forever preferred_lft forever
    inet6 fe80::7285:c2ff:fe3b:1324/64 scope link
       valid_lft forever preferred_lft forever

# ip route
default via 174.xx.yy.1 dev enp6s0
10.1.1.0/24 dev enp4s5  proto kernel  scope link  src 10.1.1.1
174.xx.yy.0/20 dev enp6s0  proto kernel  scope link  src 174.xx.bb.zz




_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

Simon Hobson
In reply to this post by A
A <[hidden email]> wrote:

> I meant to include the current /etc/network/interfaces.  Below.


> auto enp4s5
> allow-hotplug enp4s5
> iface enp4s5 inet manuel
>     gateway 10.1.1.1
>     network 10.1.1.0
>     netmask 255.255.255.0
>     broadcast 10.1.1.255
>
>     # Before we can change the address we have to bring the face down
>     pre-up ip link set enp4s5 down
>
> # I don't know if I can use two ups.  I don't know the right way to do this.
>     up ip address 0.0.0.0 dev enp4s5
>     up ip link set enp4s5 up
>     down ip link set enp4s5 down

Where on earth has this come from, it's an extremely unusual setup. By the way, it should be manual not manuel.

> allow-hotplug wlp2s0
> iface wlp2s0 inet manuel
>     gateway 10.1.1.1
>     network 10.1.1.0
>     netmask 255.255.255.0
>     broadcast 10.1.1.255

> auto br0
> iface br0 inet static
>     address 10.1.1.1
>     network 10.1.1.0
>     netmask 255.255.255.0
>     broadcast 10.1.1.255
>     bridge-ports enp4s5 wlp2s0

I would suggest trying this :

auto enp4s5
iface enp4s5 inet manual

auto wlp2s0
iface wlp2s0 inet manual
(add wifi settings here)

auto br0
iface br0 inet static
    address 10.1.1.1
    netmask 255.255.255.0
    bridge-ports enp4s5 wlp2s0

That on it's own should give you a working setup with your wired and wireless bridged. You then use br0 for your networking. The bridge code takes care of tracking which devices are on which bridge port.

Not that there is no default gateway here - it would be pointing to itself. The default gateway will be set when you WAN interface is up and will point to your ISP's router.

However, I'm still not sure exactly how dhcpd will work in this setup as it's not a config I've used.


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

Simon Hobson
In reply to this post by A
A <[hidden email]> wrote:

>> OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.
>
> Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.

Key thing there is BRIDGED - ONE bridge interface with ONE address in the subnet, more than one port assigned to the bridge. What you have is NOT a bridged network, it is two SEPARATE networks with the same subnet - it's is fundamentally broken and will NOT work (at least without a lot of fudging around to work around the brokenness.


> I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up.

OK, so there's a different issue there, and it'll be to do with your masq setup.

You probably need to take a step back and get the basics working first. Forget your DHCP for the moment and statically configure some clients - that way you can work on one issue at a time.

Get the internal networking going (devices can connect between wired and wireless).

Get the internet connection working - ie setup your masq rules.

THEN work on dhcp etc.

As it is, I think you are trying to troubleshoot multiple issues at once which makes it much harder - especially when, as it appears here, networking isn't you specialist subject.

BTW - you MUST also configure filters on inbound traffic, while a GNU/Linux system is fairly secure out of the box, it is likely to be running services that you don't want to expose. So configure rules that permit only the traffic inbound to the firewall that you want, and drop everything else.
I'm another that doesn't normally work direct with iptables - I use Shorewall too.

_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
A
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

A



On 01/28/2018 01:38 AM, Simon Hobson wrote:
A [hidden email] wrote:

OK, so this box is your gateway, AP, etc, etc. In that case I believe that your setup is fundamentally broken - you have TWO SEPARATE networks (one wired, one wireless) running the same subnet.
Yes, that's how I was told to set it up by a helpful individual.  I was told since it was one subnet, no routing would be needed.  The wireless & wired interfaces would be bridged.  Seemed reasonable.  It sounds like you are suggesting exactly the same thing so "fundamentally broken" seems a bit harsh.
Key thing there is BRIDGED - ONE bridge interface with ONE address in the subnet, more than one port assigned to the bridge. What you have is NOT a bridged network, it is two SEPARATE networks with the same subnet - it's is fundamentally broken and will NOT work (at least without a lot of fudging around to work around the brokenness.


I've never setup a bridge before and got the syntax for brctl wrong and that's what caused the config you saw.  I thought I was telling it to bridge wired & wireless faces, when apparently it was doing something else without telling me so.  I was doing: brctl addif br0 enp4s5 wlp2s0 which I thought linked the two faces.  It apparently added the enp4s5 and ignored the wlp2s0.

Once I finally realized that yesterday and corrected the syntax it will not add the wireless interface as a port.

# brctl addif br0 wlp2s0
can't add wlp2s0 to bridge br0: Operation not supported

And just to be complete:

# brctl addif br0 enp4s5
device enp4s5 is already a member of a bridge; can't enslave it to bridge br0.

# brctl addif br0 enp6s0
#
That one succeeds, because like 4s5, 6s0 is a wired NIC.


That's after changing "manuel" to "manual" just in case that was a relevant issue.

Therefore as I can't add the wireless face as a port, I'm unable to bridge using brctl from the bridge-utils package.  So unless there is another virtual bridge package, bridging won't work.  Unless of course I'm still doing it wrong which is always a possibility.




I removed the bridge because I was unable to reach the Internet from yellow (nor blue).  And that is how things stand now.  Bridge up, Internet down.  Bridge down, Internet up.
OK, so there's a different issue there, and it'll be to do with your masq setup.

You probably need to take a step back and get the basics working first. Forget your DHCP for the moment and statically configure some clients - that way you can work on one issue at a time.

I had everything working including DHCP except blue->Internet.  After getting that far by myself, someone redirected me to a different setup that as you've pointed out, is broken, starting with the bridging package itself (no wireless support).

So proceeding further down this path is not possible afaik.  Therefore, also considering the fact that bridge-utils is considered deprecated, I'm going to go back to my original setup of 2 separate subnets and route them as I originally intended. I got sent down the garden path, thanks for helping bring me back!

There is apparently another way to do this using systemd I stumbled on, but I have not looked into it, bridge-utils seemed to be the fastest way forward.  Sadly, it was not.

Thank you for all your insight and help, it is greatly appreciated!

By the way... do you (or anyone) know of a general networking mailing list?  Thanks again!


_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
A
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

A

It suddenly occurred to me that it's probably my wireless card that does not support the operation, however it's ambiguous.

It's moot though because dhcpd won't listen on a virtual interface (unless I messed something up there too!)

In any event, I was able to get both blue and yellow onto the 'Net today with Bill Shirley's help and of course everyone else that pointed out my various mistakes.

Thanks again to everyone that participated, much appreciated!!!


On 01/28/2018 10:19 AM, A wrote:

# brctl addif br0 wlp2s0
can't add wlp2s0 to bridge br0: Operation not supported



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

Simon Hobson
In reply to this post by A
A <[hidden email]> wrote:

> I've never setup a bridge before and got the syntax for brctl wrong and that's what caused the config you saw.  I thought I was telling it to bridge wired & wireless faces, when apparently it was doing something else without telling me so.  I was doing: brctl addif br0 enp4s5 wlp2s0 which I thought linked the two faces.  It apparently added the enp4s5 and ignored the wlp2s0.
>
> Once I finally realized that yesterday and corrected the syntax it will not add the wireless interface as a port.
>
> # brctl addif br0 wlp2s0
> can't add wlp2s0 to bridge br0: Operation not supported
>
> ...
> That's after changing "manuel" to "manual" just in case that was a relevant issue.
>
> Therefore as I can't add the wireless face as a port, I'm unable to bridge using brctl from the bridge-utils package.  So unless there is another virtual bridge package, bridging won't work.  Unless of course I'm still doing it wrong which is always a possibility.

After a quick search, I came across these :

https://wiki.debian.org/BridgeNetworkConnections#Bridging_with_a_wireless_NIC
> Just like you can bridge two wired ethernet interfaces, you can bridge between an ethernet interface and a wireless interface. However, most Access Points (APs) will reject frames that have a source address that didn’t authenticate with the AP ...


Note that as you are operating in access point mode, you do NOT need to do all the ebtables stuff it then carries on to describe.

And then :
https://wiki.archlinux.org/index.php/Network_bridge#Wireless_interface_on_a_bridge
> To add a wireless interface to a bridge, you first have to assign the wireless interface to an access point or start an access point with hostapd. Otherwise the wireless interface will not be added to the bridge.


So this seems to imply that the wireless interface must be up and configured BEFORE it can be added to the bridge.

Next test is to make sure hostapd is running, and then see if the wireless interface can be added. Both these pages are clear that a wireless NIC can be added to a bridge.

_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
A
Reply | Threaded
Open this post in threaded view
|

Re: No subnet declaration; Can't open /var/lib/dhcp/dhcpd.leases for append

A

So this seems to imply that the wireless interface must be up and configured BEFORE it can be added to the bridge.

No doubt that's another place where I went wrong.  wireless face is still not up.  Had it working at one time, no longer.  Thanks for all your input Simon!  I think I'm going to stick with what's working right now (as mentioned in another email, blue & yellow can both ping the 'Net over wired) and focus on figuring out what's wrong with hostapd, then on to remaining DNS issues.  I greatly appreciate all your input!!!


On 01/28/2018 12:28 PM, Simon Hobson wrote:

After a quick search, I came across these :

https://wiki.debian.org/BridgeNetworkConnections#Bridging_with_a_wireless_NIC
Just like you can bridge two wired ethernet interfaces, you can bridge between an ethernet interface and a wireless interface. However, most Access Points (APs) will reject frames that have a source address that didn�t authenticate with the AP ...

Note that as you are operating in access point mode, you do NOT need to do all the ebtables stuff it then carries on to describe.

And then :
https://wiki.archlinux.org/index.php/Network_bridge#Wireless_interface_on_a_bridge
To add a wireless interface to a bridge, you first have to assign the wireless interface to an access point or start an access point with hostapd. Otherwise the wireless interface will not be added to the bridge.

So this seems to imply that the wireless interface must be up and configured BEFORE it can be added to the bridge.

Next test is to make sure hostapd is running, and then see if the wireless interface can be added. Both these pages are clear that a wireless NIC can be added to a bridge.

_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users



_______________________________________________
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users
12