Documentation > Miscellaneous > FAQ

Troubleshooting/FAQ

This sums up problems we’ve seen users run into.

I modprobed Jool but it doesn’t seem to be doing anything.

Modprobing Jool without enough arguments is legal. It will assume you intend to finish configuring using the userspace app, and sit idle until you’ve done so.

Use the userspace app’s --global flag to figure out Jool’s status:

$ jool_siit --global
  Status: Disabled
$ jool --global
  Status: Disabled

SIIT Jool’s minimum configuration requirements are

NAT64 Jool’s minimum configuration requirements are

If that doesn’t seem to be the problem, try the logs.

Jool is intermitently unable to translate traffic.

Did you run something in the lines of

ip addr flush dev eth1

?

Then you might have deleted the interface’s Link address.

Link addresses are used by several relevant IPv6 protocols. In particular, they are used by the Neighbor Discovery Protocol, which means if you don’t have them, the translating machine will have trouble finding its IPv6 neighbors.

Check the output of ip addr.

user@T:~$ /sbin/ip address
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 08:00:27:83:d9:40 brd ff:ff:ff:ff:ff:ff
    inet6 2001:db8:aaaa::1/64 scope global 
       valid_lft forever preferred_lft forever
    inet6 fe80::a00:27ff:fe83:d940/64 scope link 
       valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 08:00:27:c6:01:48 brd ff:ff:ff:ff:ff:ff
    inet6 2001:db8:bbbb::1/64 scope global tentative 
       valid_lft forever preferred_lft forever

Interface eth0 is correctly configured; it has both a “scope global” address (used for typical traffic) and a “scope link” address (used for internal management). Interface eth1 lacks a link address, and is therefore a headache inducer.

The easiest way to restore scope link addresses, we have found, is to just reset the interface:

ip link set eth1 down
ip link set eth1 up

Yes, I’m serious:

user@T:~$ /sbin/ip address
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 08:00:27:83:d9:40 brd ff:ff:ff:ff:ff:ff
    inet6 2001:db8:aaaa::1/64 scope global 
       valid_lft forever preferred_lft forever
    inet6 fe80::a00:27ff:fe83:d940/64 scope link 
       valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 08:00:27:c6:01:48 brd ff:ff:ff:ff:ff:ff
    inet6 fe80::a00:27ff:fec6:148/64 scope link 
       valid_lft forever preferred_lft forever

(Note, you need to add the global address again)

Also, for future reference, keep in mind that the “correct” way to flush an interface is

ip addr flush dev eth1 scope global

IPv4 doesn’t need link addresses that much.

The throughput is terrible!

Turn offloads off!

If you’re running Jool in a guest virtual machine, something important to keep in mind is that you might rather or also have to disable offloads in the VM host’s uplink interface.