Re: SCTP INITs failing :Incorrect CRC32C,should be

Dear all,

you can choose in Wireshark Edit/Preferences/Protocols/SCTP and
select Checksum type 'Automatic'.

Please note that Adler32 has been replaced by CRC32 years ago!

Best regards
Michael

On Apr 25, 2009, at 6:07 AM, Adnan Hasnain Alam wrote:

> Vivek,
>
> Have you checked if both sides are using CRC32 and not Adler-32.  
> (RFC 3309). Change the settings in Wireshark to determine if the  
> checksum is correct if it were Adler-32.
>
>
> Best regards,
>
> Adnan
>
> On Fri, Apr 24, 2009 at 7:56 PM, Kumar, Vivek (NSN - IN/Bangalore) <[email protected] 
> > wrote:
> Hello,
>
> We have a scenario in which we have a server running linux kernel  
> 2.6.28-7 on a MIPS64 architecture machine.
> The client on the other hand is an X86 machine.
>
> The issue being observed here now is that the When the client sends  
> and INIT message the server discards the packet as the packet shows  
> "Incorrect CRC32C,should be 0x98bfedc6".
>
> If we disable the CRC checking at the server side though it proceeds  
> one step further,i.e.,the server now responds back with an INIT ACK  
> (ignoring the checksum) but the client now discards the packet  
> giving the same reason,i.e.,"Incorrect CRC32C,should be "
>
> I have attached a capture file illustrating the same.
>
> Any comments/suggestions would be really helpful.
>
> <<capture.txt>>
>
> Thanks in advance.
> Regards,
> Vivek Kumar
> Ph: +918043633849
>
>
> _______________________________________________
> Sigtran mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/sigtran
>
>
>
>
> -- 
> Adnan Hasnain Alam
> Sent from Delhi, DL, India  
> _______________________________________________
> Sigtran mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/sigtran