Hi group
We have a problem changing the provisioning server not working. We're in process to replace the provisioning server. First we probed version 4.1.1 ISC Dhcp in ubuntu 11.4. This is what we got:
option tftp-server-name "172.16.1.2";
option ntp-servers 172.16.1.2;
ddns-update-style none;
option time-servers 10.0.0.1;
authoritative;
default-lease-time 14400;
hosts.conf
#######
group {
next-server 172.16.1.2;
filename "1024down_512up_1pc.bin";
host 30_jesus_guerrero_gomez { hardware ethernet 00:20:40:53:b4:b8; }
host 65_ana_elvia_martinez { hardware ethernet 00:04:bd:c8:c6:0e; }
}
shared-network CM_CMTS1 {
option dhcp-lease-time 172800;
max-lease-time 345600;
deny client-updates;
ddns-update-style none;
subnet 10.0.0.0 netmask 255.255.248.0 {
option broadcast-address 10.0.7.255;
deny client-updates;
ddns-update-style none;
ddns-updates off;
option routers 10.0.0.1;
}
}
The dhcp server starts up without any issues. Then we set the following settings for NTP:
restrict 127.0.0.1 mask 255.255.255.255 nomodify
server 0.pool.ntp.org
server 1.pool.ntp.org
server 2.pool.ntp.org
# acceso de actualizacion de tiempo a los servidores
restrict 172.16.0.0 mask 255.255.255.0 nomodify
restrict 172.16.1.0 mask 255.255.255.0 nomodify
# acceso de actualizacion de tiempo a los clientes
restrict 192.168.8.0 mask 255.255.248.0 nomodify
restrict 10.0.0.1 mask 255.255.248.0 nomodify
TFTP and NTP works. We probed tftp with the cmts CISCO UBR 7246 and also NTP and OK. But some cable modem goes to init(i) state. Also we check that ARP table is refresh with new mac from new server and is OK. Also we tried and old version from ISC dhcp 3.0.5 (same version the old provisioning server has installed) We were trying to debug the registration process with the cisco link:
http://www.cisco.com/en/US/tech/tk86/tk89/technologies_tech_note09186a00...
with the command debug cable-modem mac log verbose but we couldn't find that command on the UBR 7246 IOS version 12.1(5)T9.
So we would like to ask you:
1- Is there any problem with dhcp version or config file, when we believe routing is OK?
2- How can we do debuggin process with an alternative command for that UBR? IOS version 12.1(5) T9
3- Can you help us :-) ?
I thought that changing provisioning server was very easy and only matter of changing IP address but I don't know what else to do.
Sorry for my english.
Thanks again.
We've been searching for a solution and I found some posts related:
http://www.linuxquestions.org/questions/linux-software-2/linux-dhcp-serv...
http://www.docsis.org/node/238
Both posts says about an error on subnet mask which could be the cause of init(i) state. We're using netmask 255.255.248.0 for 1022 hosts. with network address 10.0.0.0 and range 10.0.0.2 to 10.0.7.254 for dhcp lease for cable modems. Also is set on cmts CISCO uBR 7246.
I couldn't make any tests due to all devices are out of town and I have to travel and make some lab tests, there's no remote access. So I want to ask you if that subnetting could cause the init(i) state to some of the cable modems? This is to confirm is this could be the most reasonable cause and have some ideas before going to that lab place.
Thanks in advance
Maybe its just the isc flavor of config, but I don't see any pools assigned.
Your interface in the cmts will have the 10.0.0.1 ip assigned to it, and you also need to make sure that ip is *not* included in the pool. for example
pool {
range 10.0.0.2 10.0.0.254;
allow "modemcm";
}
Also, I've found this to be of great help:
http://docsis.beckitrue.com/documents/cisco/initialization_wallchart.pdf
Hi group, hi capm
Thanks for your attention. The issue was a wrong subnet. We got 255.255.255.248 and the correct one was 255.255.248.0
That thing was driving me crazy and I checked at least three times until another person saw the error.
Thanks in advanced.