Routing

Hi Expert,

Refer following KB,

http://kb.juniper.net/InfoCenter/index?page=content&id=KB22499&actp=RSS

It mentioned there that the solution is to use only the default routing table and source the NTP update traffic from the default routing table.

My question is how do I configure the NTP update traffic from the default routing table?

 rgds. 

You dont have to do nothing.

  For default, your ntp configuration will use the inet.0 table and any interface available in this table to reach the ntp server. Any source address should belong to the inet.0 table, so cant use routing-instances with a minimal ntp configuration.

  If you dont do this, you will get error messages as "address unavailable" or something else. For example, if you try to use a source address that belong to a routing instance, then this will not be present in the inet.0 table and you will get this messages or something like.

  If you want to use any source address that belong to a routing-instance, then will have to use rib groups, this way you could leak some of your routing instance interface address to the inet.0 table. And the opposite, the route to the ntp server from the inet.0 table to the routing-instance. I got this working on MX series, not sure if you could do the same in SRX.

Hi Alex,

Thanks for the prompt reply.

Actually, I'm having a problem with my jseries router but I only found this KB so I assumed the configuration would be the same.

My networks are throught MPLS VPN which I got routers at remote site and connected to SSG at my office.

My SSG will be the NTP server and others(jseries) as clients.

I've configured NTP setting but the state showing as INIT.

I also can see error logs on the log messages "NTP server unreachable".

Any idea how to solve this problem?

TQ.

Hi,

i have a problem about ntp associations.

I have 2 server ntp

ntp {
        boot-server 113.210.8.81;
        server 113.210.8.81 prefer;
        server 113.210.8.85;
        source-address 172.24.139.236

and a route static to reach both IP address

router> show configuration routing-options

static {
    route 172.24.139.238/32 next-table DCN.inet.0;
    route 113.210.8.81/32 next-table DCN.inet.0;
    route 113.210.8.85/32 next-table DCN.inet.0;

}

DCN {
        instance-type virtual-router;
        interface ge-0/2/0.470;
        interface ge-0/2/5.471;
        interface ge-0/3/1.0;
        interface lo0.2;
                protocols {
            ospf {
                 area 0.0.0.0 {
                    interface lo0.2 {
                        passive;
                    }
                    interface ge-0/2/0.470;
                    interface ge-0/2/5.471;
                    interface ge-0/3/1.0 {
                        passive;
                    }
                }
            }
        }
    }
}

router> show ntp associations
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
*113.210.8.81    113.210.1.95     2 -   26   64   77   16.054   14.603   2.342
 113.210.8.85    .INIT.          16 -    -  512    0    0.000    0.000 4000.00

with second IP addrees remain in INIT.

I don't understand why.

TKS

Raffaella

Hello Alex,

can you please give us the configuration that you did working for MX. We have similar problem with Radius, Syslog and NTP servers which are in a different O&M VRF in MX platforms.