Problem with AMPDU and RtsCtsThreshold

[jaasilva]

Hi all,

I've encountered a problem when using wifi 802.11n (2.4GHZ), in ad-hoc mode, and setting the RtsCtsThreshold to a value like 100 (for instance).

In my case, I'm doing routing at the application level and exchanging small packets (50-200 bytes).

When I set RtsCtsThreshold to a very low value (like 20 or less) or when I set it to a higher value (like 400 or more) the problem doesn't appear.

But when I set RtsCtsThreshold to values in between (like 100), an assert is triggered in mac-low:

assert failed. cond="remainingAmpduDuration > 0", file=../src/wifi/model/mac-low.cc, line=1725

I checked and remainingAmpduDuration has a negative value.

My theory is that the problem only appears when there are some packets that use the rts/cts mechanism and others that don't use it at the same time.

Does that makes sense?

Can anyone understand why this problem is happening?

Is it supposed to happen?

Many thanks,

JS

[Sebastien Deronne]

Could you provide your simulation script?

This could be a bug.

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to the Google Groups "ns-3-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+unsubscribe@googlegroups.com.
To post to this group, send email to ns-3-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ns-3-users.
For more options, visit https://groups.google.com/d/optout.

[jaasilva]

Hi Sebastien,

Here is the simulation script.

In the example, I'm using a module that I'm developing (a dht).

On Friday, October 21, 2016 at 11:43:40 AM UTC+1, Sebastien Deronne wrote:

Could you provide your simulation script?

This could be a bug.

2016-10-21 12:33 GMT+02:00 jaasilva <jaa....@campus.fct.unl.pt>:

Hi all,

I've encountered a problem when using wifi 802.11n (2.4GHZ), in ad-hoc mode, and setting the RtsCtsThreshold to a value like 100 (for instance).

In my case, I'm doing routing at the application level and exchanging small packets (50-200 bytes).

When I set RtsCtsThreshold to a very low value (like 20 or less) or when I set it to a higher value (like 400 or more) the problem doesn't appear.

But when I set RtsCtsThreshold to values in between (like 100), an assert is triggered in mac-low:

assert failed. cond="remainingAmpduDuration > 0", file=../src/wifi/model/mac-low.cc, line=1725

I checked and remainingAmpduDuration has a negative value.

My theory is that the problem only appears when there are some packets that use the rts/cts mechanism and others that don't use it at the same time.

Does that makes sense?

Can anyone understand why this problem is happening?

Is it supposed to happen?

Many thanks,

JS

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to the Google Groups "ns-3-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+...@googlegroups.com.

[sebastie...@gmail.com]

Your script can not compile:

../scratch/example4.cc:9:10: fatal error: 'ns3/dht-helper.h' file not found
#include "ns3/dht-helper.h"
         ^
1 error generated.

Could you please isolate the issue and keep all your own changes away?

Thanks!

<example4.cc><experiment.h>

[Yiannis]

Hi,

As far as I remember, I had encountered a similar problem (ns3.25)
with 11ac and RTS/CTS. I recall that this assert had something
to do with missed ACKs or CTS or after a retransmission where
the aggregated frame was below the rts threshold. Nevertheless,
I didn't investigate it further.

yiannis

[sebastie...@gmail.com]

Which version are you using?

Le 6 janv. 2017 à 10:52, jaasilva <jaa....@campus.fct.unl.pt> a écrit :

Hi Sebastien,

I made minimal changes to the wifi-adhoc example (to use 802.11n 2.4GHz) and it makes the problem appear.

But now there is something strange.

When I set the control mode of the remote station manager to the same value as the data value, the assert appear at the beginning of the experiment.

When I set the control mode to the lowest value possible (6mbps), the experiment runs until the end but never finishes.

Do you understand what is happening?

Regards,

JS

<wifi-adhoc.cc>

[jaasilva]

Hi,

I'm using the last version, ns-3.26.

[Gmail]

Could you try with ns-3-dev?