Hey,
I wrote once a script which runs `tc` commands in order to create `tc` on specific port.
The `tc` commands ran on `redhat5.6` perfectly.
Now, when I upgraded to `redhat6.2` I've noticed that `tc` doesn't work well on `dev lo` (loopback) and btw still works for `dev eth0`;
When I ran `tc` command on `dev lo` for specific port, let's say 8999, all of the ports on the `lo` get stuck, doesn't matter how much I limit the port. When I delete `tc` structure my process (which use another port or the same, let's say 8998) doesn't return to normal.
My process can run normally only if I don't use the `tc` command on the `lo` at all. At the moment I start `tc` on any port on the `lo`, on any limitation, my process get stuck.
Is there any change in `redhat6.2` that may affect this strange behavior?
My `tc` commands:
initialization: delete `tc` stracture on `lo`:
Code:
tc qdisc del dev lo root
for `lo` put `mtu` to 1500
Code:
ifconfig lo mtu 1500 up
create the root and 1:1
Code:
tc qdisc add dev lo root handle 1: htb default 1
limit free for all ports
Code:
tc class add dev lo parent 1: classid 1:1 htb rate 1024mbit
limit bit rate for specific port to 500kbit
Code:
tc class add dev lo parent 1: classid 1:10 htb rate 500kbit
filter which leads packets to 1:10 (limit-class)
Code:
tc filter add dev lo parent 1: prio 1 protocol ip u32 match ip sport 8999 0xffff flowid 1:10
Thanks.
