[Nfd-dev] [Operators] broadcast storm of malformed packets

Dehart, John jdd at wustl.edu
Thu Sep 5 08:51:07 PDT 2019


Right. I’m just trying to isolate it so the Testbed can survive. Its killing some nodes.
John


On Sep 5, 2019, at 11:26 AM, Junxiao Shi <shijunxiao at email.arizona.edu<mailto:shijunxiao at email.arizona.edu>> wrote:

Hi NDNOPS/John

I injected the latest one when I wrote the bug report. I don't know where does the earlier packets come from.
It doesn't matter which app is transmitting. Forwarder must not malfunction regardless of traffic.

Yours, Junxiao

On Thu, Sep 5, 2019, 11:22 Dehart, John <jdd at wustl.edu<mailto:jdd at wustl.edu>> wrote:
All:

Anyone have any ideas where this might be coming from?

John

On Sep 5, 2019, at 5:15 AM, Junxiao Shi <shijunxiao at email.arizona.edu<mailto:shijunxiao at email.arizona.edu>> wrote:

Dear folks

I've reported a bug here:
https://redmine.named-data.net/issues/4998

Yours, Junxiao

On Wed, Sep 4, 2019, 22:49 Junxiao Shi <shijunxiao at email.arizona.edu<mailto:shijunxiao at email.arizona.edu>> wrote:
Dear folks

It seems that there has been a broadcast storm going around the testbed.
I first noticed this yesterday, and it's still going on as of an hour ago.

I'm attaching a packet sample captured on my server.
The packets contain malformed Selectors, such as: (note that you can't have MetaInfo inside Selectors)
<image.png>

I ran a program to respond to these packets, and the traffic seems to have disappeared.
I suspect there are some sort of software bugs, causing the traffic to loop around instead of getting dropped.

Yours, Junxiao
_______________________________________________
Operators mailing list
Operators at alpha.netsec.colostate.edu<mailto:Operators at alpha.netsec.colostate.edu>
https://lists.netsec.colostate.edu/mailman/listinfo/operators


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20190905/2c6cdeb5/attachment-0001.html>


More information about the Nfd-dev mailing list