Please could I ask if the attached screen shot shows that an IPV6 range has been added to our system.
If so, we have not added them, and if you look carefully, the options are greyed out (i.e edit or delete) and as the ipv6 are not ours, I don’t want to activate them.
They are also shown in DNS, but the logs shows failures as follows:
/usr/bin/tail: cannot open `/var/djbdns/tinydns-fec0:21d:9ff:fe6d:3376/log/main/current’ for reading: No such file or directory
/usr/bin/tail: cannot open `/var/djbdns/tinydns-2002:54ea:1afb:b:21d:9ff:fe6d:3376/log/main/current’ for reading: No such file or directory
The interworx was updated to release candidate, so is running the latest version, and the server is centos 6 64 bit, with all updates installed.
I would be grateful if someone could let me know how and why they appeared in out system, and how to clear them out of our system.
The fec0::* IPv6 address is what’s known as a link-local address - an address used for local routing only.
This address simply comes from your operating system at boot. It’s a little odd, because the fec0::/10 range was deprecated in favor of fe80::/10 at some point. That’s why this particular address is appearing in your control panel. We filter out fe80, but it appears we didn’t consider the deprecated ranges might appear also.
The other address, beginning 2002:, is also a special range. 2002::/16 is used in tunnelling IPv6 requests over IPv4 addresses.
To be perfectly honest, I’m not sure why that one exists on your server. It appears that someone set up an IPv6-to-4 tunnel on this system, and I can only promise that it wasn’t InterWorx that did it
It’s unlikely that address would be issued via DHCP, so it’s probably going to be in a file like /etc/sysconfig/network-scripts/ifcfg-eth0, though it may be in an alias file instead.
Also of note - both addresses end with “b:21d:9ff:fe6d:3376”, which is likely the MAC address of your NIC. That indicates to me that both addresses are being automatically generated on your behalf.
Please could I ask though, why only 1 ipv6 is listed, even though we have 2 nics on different nodes, using 2 different gateways and both fully working.
The other 2002: I’ll have to look onto as they are only 2 of us with access to this, one is in hospital (a motorbike accident) and me, so I know I didn’t set it, and don’t think steve would have from his hospital bed, but I’ll look into it.
I think iw-tim and iw-Paul maybe correct, and the ipv6 address appeared because of an upstream system. This may also explain why the 6 to 4 also appears, if it was dynamically setup from upstream.
In any event it caused a medium issue with email due to no rdns on ipv6 but was very quickly resolved by iw-Paul by turning off ipv6.
I think I will have to order some ipv6 ranges, so we can test them but I think we will test them first on our cluster test were about to start.
Once again, many thanks to interworx for their help
21d:9ff:fe6d:3376/log/main/current’ for reading: No such file or directory
