Centos ipv6 secondaries not working

[nrucel]

I added server ipv6 / 64. Ping6 works for google.com.
But as IPV6ADDR_SECONDARIES, almost all of what I have added does not work.

Code: Select all

    [root@server ~]# ping6 google.com
    PING google.com(fra15s10-in-x0e.1e100.net) 56 data bytes
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=1 ttl=57 time=47.9 ms
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=2 ttl=57 time=47.9 ms
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=3 ttl=57 time=47.9 ms
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=4 ttl=57 time=48.0 ms
    ^C
    --- google.com ping statistics ---
    4 packets transmitted, 4 received, 0% packet loss, time 3945ms
    rtt min/avg/max/mdev = 47.943/47.980/48.019/0.220 ms

I randomly created ipv6.
Ex: ipv6adress:XXXX:XXXX:XXXX:XXXX
I also created 3K. Only 110 worked
I created the IPv6 address: ipv6adress::X - ipv6adress::XXXX
It worked no more than 150

working IPV6ADDR_SECONDARIES:

Code: Select all

    [root@server ~]# ping6 -I 2a03:2100:0:12::7b google.com
    PING google.com(fra15s10-in-x0e.1e100.net) from 2a03:2100:0:12::7b : 56 data bytes
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=1 ttl=57 time=50.9 ms
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=2 ttl=57 time=52.2 ms
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=3 ttl=57 time=50.6 ms
    64 bytes from fra15s10-in-x0e.1e100.net: icmp_seq=4 ttl=57 time=58.3 ms
    ^C
    --- google.com ping statistics ---
    4 packets transmitted, 4 received, 0% packet loss, time 3399ms
    rtt min/avg/max/mdev = 50.661/53.061/58.308/3.097 ms

not working IPV6ADDR_SECONDARIES: (I expect, but I cannot get results.)

Code: Select all

   [root@server ~]# ping6 -I 2a03:2100:0:12:3515:fd11:5c82:2d7e google.com
    PING google.com(fra15s10-in-x0e.1e100.net) from 2a03:2100:0:12:3515:fd11:5c82:2d7e : 56 data bytes
    ^C
    --- google.com ping statistics ---
    26 packets transmitted, 0 received, 100% packet loss, time 25534ms

But it is pinging like this.

Code: Select all

   [root@server ~]# ping6 2a03:2100:0:12:3515:fd11:5c82:2d7e
    PING 2a03:2100:0:12:3515:fd11:5c82:2d7e(2a03:2100:0:12:3515:fd11:5c82:2d7e) 56 data bytes
    64 bytes from 2a03:2100:0:12:3515:fd11:5c82:2d7e: icmp_seq=1 ttl=64 time=0.034 ms
    64 bytes from 2a03:2100:0:12:3515:fd11:5c82:2d7e: icmp_seq=2 ttl=64 time=0.083 ms
    64 bytes from 2a03:2100:0:12:3515:fd11:5c82:2d7e: icmp_seq=3 ttl=64 time=0.050 ms
    64 bytes from 2a03:2100:0:12:3515:fd11:5c82:2d7e: icmp_seq=4 ttl=64 time=0.041 ms
    ^C
    --- 2a03:2100:0:12:3515:fd11:5c82:2d7e ping statistics ---
    4 packets transmitted, 4 received, 0% packet loss, time 3493ms
    rtt min/avg/max/mdev = 0.034/0.052/0.083/0.018 ms

Those who do not work are very heavy. Not connected to the sites. I do not get a result.
My system Centos 6.9(final) 12 cpu 32gb ram
/etc/sysconfig/network-scripts/ifcfg-eth0:

Code: Select all

 DEVICE=eth0
    TYPE=Ethernet
    UUID=15edbe41-c655-4bee-a309-6973bff8e6bb
    ONBOOT=yes
    NM_CONTROLLED=yes
    BOOTPROTO=none
    HWADDR=00:0C:29:F7:46:61
    IPADDR=89.**.**.***
    PREFIX=27
    GATEWAY=89.**.**.***
    DNS1=8.8.8.8
    DNS2=8.8.4.4
    DEFROUTE=yes
    IPV4_FAILURE_FATAL=yes
    IPV6INIT=yes
    NAME="System eth0"
    IPV6ADDR=2a03:2100:0:12::/64
    IPV6_DEFAULTGW=2a03:2100:0:12::1
    IPV6FORWARDING=no
    IPV6_AUTOCONF=no
    IPV6ADDR_SECONDARIES="2a03:2100:0:12:XXX:XXXX:XXXX:XXXX 2a03:2100:0:12:XXX:XXXX:XXXX:XXXX 2a03:2100:0:12:XXX:XXXX:XXXX:XXXX"

My settings are that way

Code: Select all

IPV6ADDR_SECONDARIES="2a03:2100:0:12:XXX:XXXX:XXXX:XXXX/64 2a03:2100:0:12:XXX:XXXX:XXXX:XXXX/64"

or

Code: Select all

 IPV6ADDR_SECONDARIES="2a03:2100:0:12:XXX:XXXX:XXXX:XXXX/64 \
    2a03:2100:0:12:XXX:XXXX:XXXX:XXXX/64"

I have tried them but the result is the same.

[jlehtone]

https://access.redhat.com/documentation ... faces.html
says only "space separated" and "IPV6INIT=yes", both of which you do have. No mention of any limit.

1. Does your interface have all the addresses that you want?

Code: Select all

ip -6 ad

2. Are they all within your prefix? If not, does the router (and upstream) know that the other prefix is "here"?

3. You cannot reach out via every address?

4. Outside devices cannot reach all addresses?

5. If you do listen the interface with 'tcpdump' while testing the 3 and 4, what do you see?
No "working" and "nonworking" differ?