5 Oct
2013
5 Oct
'13
7:01 p.m.
Alain,
I'm trying to perform a packet capture of the data transmitted by
udp-sender using Wireshark and am running into the exact same problem
encountered by Lance 8 years ago. (posting listed below.)
What information can I provide to help debug this issue?
Also, can you tell me if the multicast data sent includes a sequence
number and does the receiver have the ability to filter out duplicate
multicast frames?
Regards,
Chris
On Mon 8/15/2005 10:25 AM, Alain Knaff wrote:
*Try upgrading to a newer version of
ethereal.*>**>*I just tried it here: version 0.10.9 has the problem, but
0.10.12*>*works fine.*>**>*Or use tcpdump...*
I have used the lattest
version of udpcast (udpcast-20050226.tar.gz)
as well as used ethereal version 0.10.12 and
I still get the "Bogus IP header length (0, must be at least 20)" I
will continue to try this on different Platforms with different
network switches and see what I get.
-Lance
5 Oct
5 Oct
10:55 p.m.
I believe wireshark is trying to decode it as a different protocol which
encapsulates IP in UDP. Try disabling it.
On Oct 6, 2013 6:02 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
Alain,
I'm trying to perform a packet capture of the data transmitted by
udp-sender using Wireshark and am running into the exact same problem
encountered by Lance 8 years ago. (posting listed below.)
What information can I provide to help debug this issue?
Also, can you tell me if the multicast data sent includes a sequence
number and does the receiver have the ability to filter out duplicate
multicast frames?
Regards,
Chris
On Mon 8/15/2005 10:25 AM, Alain Knaff wrote:
*Try upgrading to a newer version of
ethereal.*>**>*I just tried it here:
version 0.10.9 has the problem, but
0.10.12*>*works fine.*>**>*Or use
tcpdump...*
I have used the lattest version of udpcast (udpcast-20050226.tar.gz)
as well as used ethereal version 0.10.12 and
I still get the "Bogus IP header length (0, must be at least 20)" I
will continue to try this on different Platforms with different
network switches and see what I get.
-Lance
_______________________________________________
Udpcast mailing list
Udpcast(a)udpcast.linux.lu
https://www.udpcast.linux.lu/cgi-bin/mailman/listinfo/udpcast
11:06 p.m.
Jayen,
Thank you. You were correct. I no longer have this problem. I also
deduced that there is a sequence number in the multicast frame.
I'm digging into this utility as I currently have a scenario where I get a
bunch of duplicate frames hitting the udp-receiver (network issue, not
application issue) and the application does not gracefully handle it. I
figured that the application did not have a sequence number. As it does,
there must be some other problem. More to come.
Chris
On Sat, Oct 5, 2013 at 3:55 PM, Jayen <jayenashar(a)gmail.com> wrote:
I believe wireshark is trying to decode it as a
different protocol which
encapsulates IP in UDP. Try disabling it.
On Oct 6, 2013 6:02 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
Alain,
I'm trying to perform a packet capture of the data transmitted by
udp-sender using Wireshark and am running into the exact same problem
encountered by Lance 8 years ago. (posting listed below.)
What information can I provide to help debug this issue?
Also, can you tell me if the multicast data sent includes a sequence
number and does the receiver have the ability to filter out duplicate
multicast frames?
Regards,
Chris
On Mon 8/15/2005 10:25 AM, Alain Knaff wrote:
*Try upgrading to a newer version of
ethereal.*>**>*I just tried it
here: version 0.10.9 has the problem, but
0.10.12*>*works fine.*>**>*Or use
tcpdump...*
I have used the lattest version of udpcast (udpcast-20050226.tar.gz)
as well as used ethereal version 0.10.12 and
I still get the "Bogus IP header length (0, must be at least 20)" I
will continue to try this on different Platforms with different
network switches and see what I get.
-Lance
_______________________________________________
Udpcast mailing list
Udpcast(a)udpcast.linux.lu
https://www.udpcast.linux.lu/cgi-bin/mailman/listinfo/udpcast
11:27 p.m.
is it over Wi-Fi? There may be retransmissions, but they haven't caused me
any issues.
On Oct 6, 2013 10:06 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
Jayen,
Thank you. You were correct. I no longer have this problem. I also
deduced that there is a sequence number in the multicast frame.
I'm digging into this utility as I currently have a scenario where I get a
bunch of duplicate frames hitting the udp-receiver (network issue, not
application issue) and the application does not gracefully handle it. I
figured that the application did not have a sequence number. As it does,
there must be some other problem. More to come.
Chris
On Sat, Oct 5, 2013 at 3:55 PM, Jayen <jayenashar(a)gmail.com> wrote:
I believe wireshark is trying to decode it as a
different protocol which
encapsulates IP in UDP. Try disabling it.
On Oct 6, 2013 6:02 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
Alain,
I'm trying to perform a packet capture of the data transmitted by
udp-sender using Wireshark and am running into the exact same problem
encountered by Lance 8 years ago. (posting listed below.)
What information can I provide to help debug this issue?
Also, can you tell me if the multicast data sent includes a sequence
number and does the receiver have the ability to filter out duplicate
multicast frames?
Regards,
Chris
On Mon 8/15/2005 10:25 AM, Alain Knaff wrote:
*Try upgrading to a newer version of
ethereal.*>**>*I just tried it
here: version 0.10.9 has the problem, but
0.10.12*>*works fine.*>**>*Or use
tcpdump...*
I have used the lattest version of udpcast (udpcast-20050226.tar.gz)
as well as used ethereal version 0.10.12 and
I still get the "Bogus IP header length (0, must be at least 20)" I
will continue to try this on different Platforms with different
network switches and see what I get.
-Lance
_______________________________________________
Udpcast mailing list
Udpcast(a)udpcast.linux.lu
https://www.udpcast.linux.lu/cgi-bin/mailman/listinfo/udpcast
11:34 p.m.
No, it's a wired network with a known flaw that leads to temporary packet
storms of up to a second or so. (We are working on addressing this network
flaw, but it seems to have exposed a vulnerability within this application.)
On Sat, Oct 5, 2013 at 4:27 PM, Jayen <jayenashar(a)gmail.com> wrote:
is it over Wi-Fi? There may be retransmissions, but
they haven't caused me
any issues.
On Oct 6, 2013 10:06 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
> Jayen,
>
> Thank you. You were correct. I no longer have this problem. I also
> deduced that there is a sequence number in the multicast frame.
>
> I'm digging into this utility as I currently have a scenario where I get
> a bunch of duplicate frames hitting the udp-receiver (network issue, not
> application issue) and the application does not gracefully handle it. I
> figured that the application did not have a sequence number. As it does,
> there must be some other problem. More to come.
>
> Chris
>
>
> On Sat, Oct 5, 2013 at 3:55 PM, Jayen <jayenashar(a)gmail.com> wrote:
>
>> I believe wireshark is trying to decode it as a different protocol which
>> encapsulates IP in UDP. Try disabling it.
>> On Oct 6, 2013 6:02 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
>>
>>> Alain,
>>>
>>> I'm trying to perform a packet capture of the data transmitted by
>>> udp-sender using Wireshark and am running into the exact same problem
>>> encountered by Lance 8 years ago. (posting listed below.)
>>>
>>> What information can I provide to help debug this issue?
>>>
>>> Also, can you tell me if the multicast data sent includes a sequence
>>> number and does the receiver have the ability to filter out duplicate
>>> multicast frames?
>>>
>>> Regards,
>>>
>>> Chris
>>>
>>>
>>>
>>> On Mon 8/15/2005 10:25 AM, Alain Knaff wrote:
>>> >*Try upgrading to a newer version of ethereal.*>**>*I just tried
it
>>> here: version 0.10.9 has the problem, but 0.10.12*>*works
fine.*>**>*Or use
>>> tcpdump...*
>>>
>>> I have used the lattest version of udpcast (udpcast-20050226.tar.gz)
>>> as well as used ethereal version 0.10.12 and
>>> I still get the "Bogus IP header length (0, must be at least 20)"
I
>>> will continue to try this on different Platforms with different
>>> network switches and see what I get.
>>>
>>> -Lance
>>> _______________________________________________
>>> Udpcast mailing list
>>> Udpcast(a)udpcast.linux.lu
>>> https://www.udpcast.linux.lu/cgi-bin/mailman/listinfo/udpcast
>>>
>>
>
6 Oct
6 Oct
12:03 a.m.
Do you have a switching loop? (I guess if you found it, you'd have already
addressed it.)
On Oct 6, 2013 10:34 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
No, it's a wired network with a known flaw that
leads to temporary packet
storms of up to a second or so. (We are working on addressing this network
flaw, but it seems to have exposed a vulnerability within this application.)
On Sat, Oct 5, 2013 at 4:27 PM, Jayen <jayenashar(a)gmail.com> wrote:
is it over Wi-Fi? There may be retransmissions,
but they haven't caused
me any issues.
On Oct 6, 2013 10:06 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
> Jayen,
>
> Thank you. You were correct. I no longer have this problem. I also
> deduced that there is a sequence number in the multicast frame.
>
> I'm digging into this utility as I currently have a scenario where I get
> a bunch of duplicate frames hitting the udp-receiver (network issue, not
> application issue) and the application does not gracefully handle it. I
> figured that the application did not have a sequence number. As it does,
> there must be some other problem. More to come.
>
> Chris
>
>
> On Sat, Oct 5, 2013 at 3:55 PM, Jayen <jayenashar(a)gmail.com> wrote:
>
>> I believe wireshark is trying to decode it as a different protocol
>> which encapsulates IP in UDP. Try disabling it.
>> On Oct 6, 2013 6:02 AM, "Chris Healy" <cphealy(a)gmail.com> wrote:
>>
>>> Alain,
>>>
>>> I'm trying to perform a packet capture of the data transmitted by
>>> udp-sender using Wireshark and am running into the exact same problem
>>> encountered by Lance 8 years ago. (posting listed below.)
>>>
>>> What information can I provide to help debug this issue?
>>>
>>> Also, can you tell me if the multicast data sent includes a sequence
>>> number and does the receiver have the ability to filter out duplicate
>>> multicast frames?
>>>
>>> Regards,
>>>
>>> Chris
>>>
>>>
>>>
>>> On Mon 8/15/2005 10:25 AM, Alain Knaff wrote:
>>> >*Try upgrading to a newer version of ethereal.*>**>*I just tried
it
>>> here: version 0.10.9 has the problem, but 0.10.12*>*works
fine.*>**>*Or use
>>> tcpdump...*
>>>
>>> I have used the lattest version of udpcast (udpcast-20050226.tar.gz)
>>> as well as used ethereal version 0.10.12 and
>>> I still get the "Bogus IP header length (0, must be at least 20)"
I
>>> will continue to try this on different Platforms with different
>>> network switches and see what I get.
>>>
>>> -Lance
>>> _______________________________________________
>>> Udpcast mailing list
>>> Udpcast(a)udpcast.linux.lu
>>> https://www.udpcast.linux.lu/cgi-bin/mailman/listinfo/udpcast
>>>
>>
>
3460
days inactive
3461
days old
5 comments
2 participants
participants (2)
-
Chris Healy -
Jayen