Tested with HP 3500 switches and still getting the same error when using
pox as a controller. Also getting this error when using floodlight with
mininet. However when I use nox, it works fine. Don't know where the real
is.

Regards,
Shayan Saeed


On Mon, Nov 4, 2013 at 2:48 PM, Shayan Saeed <shayansaee...@gmail.com>wrote:

> For idle-timeout, I was doing some tests by sending some packets
> periodically after some variable delay to see if the flow for those packets
> times out.
>
> *Turns out that if I send the same packet after some short delays again
> and again and then wait a long while*, *the flow doesn't get removed*even 
> though I can see from dpctl dump-flows that no new packet has been
> received in that time period. I have put the results from dpctl dump-flows
> below:
>
> As you can see that 2 sec have passed since this flow rule has been
> created and n_packets = 6 with idle timeout value=1
>
> cookie=0, duration_sec=2s, duration_nsec=632000000s, table_id=0,
> priority=65535, n_packets=6, n_bytes=324,
> idle_timeout=1,hard_timeout=0,tcp,in_port=2,dl_vlan=0xffff,dl_vlan_pcp=0x00,dl_src=00:00:00:00:00:06,dl_dst=00:00:00:00:00:01,nw_src=10.0.0.6,nw_dst=10.0.0.1,nw_tos=0x00,tp_src=7000,tp_dst=7000,actions=mod_dl_dst:00:00:00:00:00:01,output:1
>
> Now after 4 sec the number of packets is still 6 but the entry has not
> timed out even though about 2 sec have passed which is greater than the
> idle_timeout duration
>
> cookie=0, duration_sec=4s, duration_nsec=516000000s, table_id=0,
> priority=65535, n_packets=6, n_bytes=324,
> idle_timeout=1,hard_timeout=0,tcp,in_port=2,dl_vlan=0xffff,dl_vlan_pcp=0x00,dl_src=00:00:00:00:00:06,dl_dst=00:00:00:00:00:01,nw_src=10.0.0.6,nw_dst=10.0.0.1,nw_tos=0x00,tp_src=7000,tp_dst=7000,actions=mod_dl_dst:00:00:00:00:00:01,output:1
>
> I am running the pox l3 learning switch here for some basic testing. The
> flow gets removed just after 5 seconds of receiving nothing even though I
> haven't set any hard timeout value and idle_timeout value of 1 sec.
>
> Regards,
> Shayan Saeed
>
_______________________________________________
openflow-discuss mailing list
openflow-discuss@lists.stanford.edu
https://mailman.stanford.edu/mailman/listinfo/openflow-discuss

Reply via email to