guan...@gyhuang:~/Desktop/tcpreplay-3.4.3$ sudo tcprewrite --dlt=enet
--skipbroadcast --enet-smac=00:22:19:17:68:94
--enet-dmac=00:22:19:17:AB:2F  --skip-soft-errors -i
/home/guanyao/Research_Works/OpenFlowMR/equinix-sanjose.dirB.20081218-045904.UTC.anon.pcap
-o /home/guanyao/Research_Works/OpenFlowMR/output.pcap
tcprewrite: ./plugins/dlt_en10mb/en10mb.c:346: dlt_en10mb_encode:
Assertion `pktlen >= 14' failed.
Aborted (core dumped)

So, I guess I should use fixlen option. I find I can only use
--fixlen=del, otherwise the same error occurs.

guan...@gyhuang:~/Desktop/tcpreplay-3.4.3$ sudo tcprewrite --dlt=enet
--skipbroadcast --fixlen=del --enet-smac=00:22:19:17:68:94
--enet-dmac=00:22:19:17:AB:2F  --skip-soft-errors -i
/home/guanyao/Research_Works/OpenFlowMR/equinix-sanjose.dirB.20081218-045904.UTC.anon.pcap
-o /home/guanyao/Research_Works/OpenFlowMR/output.pcap

Fatal Error: Error rewriting packets: From
edit_packet.c:untrunc_packet() line 314:
Invalid fixlen value: 0x3

>From the ouput I guess I should use mtu option. However, the same
error comes again:
guan...@gyhuang:~/Desktop/tcpreplay-3.4.3$ sudo tcprewrite --dlt=enet
--skipbroadcast --fixlen=del --mtu-trunc --enet-smac=00:22:19:17:68:94
--enet-dmac=00:22:19:17:AB:2F --skip-soft-errors -i
/home/guanyao/Research_Works/OpenFlowMR/equinix-sanjose.dirB.20081218-045904.UTC.anon.pcap
-o /home/guanyao/Research_Works/OpenFlowMR/output.pcap
tcprewrite: ./plugins/dlt_en10mb/en10mb.c:346: dlt_en10mb_encode:
Assertion `pktlen >= 14' failed.
Aborted (core dumped)


Any idea what happens here?



On Tue, Dec 1, 2009 at 4:31 PM, Aaron Turner <[email protected]> wrote:
> On Tue, Dec 1, 2009 at 3:12 PM, Guanyao Huang <[email protected]> wrote:
>> Hi, I am still confused. Below is my command:
>> myn...@xxxx:~/Desktop/tcpreplay-3.4.3$ sudo tcprewrite --user-dlt=1
>> --dlt=enet --enet-smac=00:22:19:17:68:94 --enet-dmac=00:22:19:17:AB:2F
>> --fixlen=del --skip-soft-errors -i
>> /home/myname/Research_Works/OpenFlowMR/equinix-sanjose.dirB.20081218-045904.UTC.anon.pcap
>> -o /home/myname/Research_Works/OpenFlowMR/output.pcap
>
> There is no reason to use --dlt=enet and --user-dlt.  --user-dlt is
> for when you use --dlt=user and you need to create a custom L2 header
> using --user-dlink.
>
>> Fatal Error: Error rewriting packets: From
>> edit_packet.c:untrunc_packet() line 314:
>> Invalid fixlen value: 0x3
>>
>> I donot know what does this stand for.
>
> Uh... that option never worked.  not sure why it's documented.  Sorry.
>
>> Did you mean I should first use the above commands to change smac and
>> dmac and also dlt type, then I can use --endpoint, right?
>
> You can do all of it at the same time if you wanted.  Often people
> break it up into two steps because the cli options gets pretty long
> and confusing.  But....
>
>> Also, regarding the --endpoints arg, I dont want to split the traffic
>> to client and server, I only want to send all of it from one host to
>> another host, why still I need cache file?
>
> Well actually you do want to use a cachefile with endpoints (and
> smac/dmac for that matter).  The reason is that tcprewrite is
> *stupid*.  It's stupid because it needs to be fast in order to edit
> packets as they're sent (via tcpreplay-edit) since both use the same
> engine/code base.
>
> It is allowed to be stupid because it can reference a cache file which
> has all the smarts- ie, the cache file tells tcprewrite which packets
> are sent by the server and which are sent by the client.  Without the
> cache file, tcprewrite will treat all the packets the same (ie: ALL
> packets will have the same source & destination MAC & IP's).
>
> Hence if you want your packets to look right, you need to use tcpprep
> to create a cachefile for tcprewrite.
>
> Now, there is a time when you can get away without doing a cache file-
> when all the packets are going in a single direction (server never
> talks back to the client).  SNMP traps or syslog traffic are good
> examples of that.  If that's the case, just use --srcipmap &
> --dstipmap like this:
>
> --srcipmap=0.0.0.0/0:5.5.5.5/32
>
> 0.0.0.0/0 == match all IP's
> 5.5.5.5/32 == replace with new IP is 5.5.5.5
>
> That will only apply to the source IP field.  For the destination IP
> field, use --dstipmap.
>
> Good luck!
> -Aaron
>
>
> --
> Aaron Turner
> http://synfin.net/
> http://tcpreplay.synfin.net/ - Pcap editing and replay tools for Unix & 
> Windows
> Those who would give up essential Liberty, to purchase a little temporary
> Safety, deserve neither Liberty nor Safety.
>    -- Benjamin Franklin
> "carpe diem quam minimum credula postero"
>
>
>
>> Thank you.
>>
>> On Tue, Dec 1, 2009 at 8:05 AM, yassine antir <[email protected]> wrote:
>>> Hi,
>>>
>>> you should rewrite your RAW IP first
>>>
>>> The following would cause all traffic to have a destination host MAC address
>>> (example 00:55:22:AF:C6:37) (the gateway MAC address if your destination is
>>> not in the same network) and your source MAC address (example
>>> 00:44:66:FC:29:AF):
>>>
>>> $ tcprewrite --enet-dmac=00:55:22:AF:C6:37 --enet-smac=00:44:66:FC:29:AF
>>> --infile=input.pcap --outfile=output.pcap
>>>
>>>
>>>
>>> then just relay the output file.
>>>
>>> For more informations see http://tcpreplay.synfin.net/trac/wiki/tcprewrite
>>>
>>>
>>>
>>>> Date: Tue, 1 Dec 2009 01:40:55 -0800
>>>> From: [email protected]
>>>> To: [email protected]
>>>> Subject: [Tcpreplay-users] Help regarding sending raw IP pcap file to
>>>> ethernet
>>>>
>>>> Hi,
>>>>
>>>> I now want to tcpreplay some raw IP on the DLT_EN10MB (Ethernet) . I
>>>> dont know what changes I should make on the pcap file. It seems to me
>>>> "Forcing Traffic Between Two Hosts" is enough.
>>>>
>>>> Do you have any ideas what else I should do. Currently, I can not
>>>> successfully send the pcap file. The pcap file was from some high
>>>> speed link. Even if I use a speed option, I can see from tcpdump that
>>>> the tcpreplaying of flows is not successful.
>>>>
>>>> For my case, if I Forcing Traffic Between Two Hosts, how can I get the
>>>> correct cache file?
>>>>
>>>> Thanks.
>>>>
>
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> Tcpreplay-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/tcpreplay-users
> Support Information: http://tcpreplay.synfin.net/trac/wiki/Support
>

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing. 
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcpreplay-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/tcpreplay-users
Support Information: http://tcpreplay.synfin.net/trac/wiki/Support

Reply via email to