Wireshark-users: Re: [Wireshark-users] end-to-end delay calculation
From: "Martin Mathieson" <martin.r.mathieson@xxxxxxxxxxxxxx>
Date: Thu, 13 Mar 2008 15:06:19 +0000
You could of course do a ping.  And see how close to the middle the far side sees something happen...

On Thu, Mar 13, 2008 at 2:49 PM, Fabiana moreno <fvmoreno@xxxxxxxxx> wrote:
Well i did this...but i dont know if it's ideal...i think it would work though....

I synchronized the two computers to a server on the internet before every capture so both have the exact same time.

i have my sniffer in the two computers, the one that sends and the ones that receive.

I created a function in matlab that calculates the difference between those two times...

I am being really carefull when i have a packet lost or things like that....

But probably this can work???? ...

On 13/03/2008, juan.wortley@xxxxxxx <juan.wortley@xxxxxxx> wrote:
Hi Fabiana,
 
the only way to do that is if you know the time difference between the machines.
It´s not so easy however if you use windows you could try synchronizing both endpoints by using:
w32tm /config /manualpeerlist:[IP to synchronize to] /update /syncfromflags:MANUAL
 
and then monitor the time difference with:
w32tm /stripchart /computer:[IP to synchronize to]
 
,you always can have a delay reference by pinging and dividing by 2 the RTT.
 
HTH
Juan
 


From: wireshark-users-bounces@xxxxxxxxxxxxx [mailto:wireshark-users-bounces@xxxxxxxxxxxxx] On Behalf Of ext Fabiana moreno
Sent: Miércoles, 12 de Marzo de 2008 09:38 a.m.
To: Community support list for Wireshark
Subject: Re: [Wireshark-users] end-to-end delay calculation

so ... it is not reliable if i substract the time i hav in the client minus the time i have in the server to get my end the end delay? 

On 12/03/2008, Hansang Bae <hbae@xxxxxxxxxx> wrote:
Fabiana moreno wrote:
> Hello there,
>
> I know wireshark is not able to calculate the end-to-end delay of a
> packet when streaming. I was just wondering if adjusting the clocks of
> my two computers(sender and receiver) to the network time protocol and
> having the sniffer at both ends i could calculate the end-to-end-delay
> tracing each packet. Does this sound logical?
>


It is within the limits of ntp.  Unless there is a WAN involved, the
packets are flying around at an order of magnitude faster than what ntp
can provide (ms resolution)

--

Thanks,
Hansang
_______________________________________________
Wireshark-users mailing list
Wireshark-users@xxxxxxxxxxxxx
http://www.wireshark.org/mailman/listinfo/wireshark-users


_______________________________________________
Wireshark-users mailing list
Wireshark-users@xxxxxxxxxxxxx
http://www.wireshark.org/mailman/listinfo/wireshark-users



_______________________________________________
Wireshark-users mailing list
Wireshark-users@xxxxxxxxxxxxx
http://www.wireshark.org/mailman/listinfo/wireshark-users