Re: udpTable byte order issues (again)

[Thomas A. <tho...@bl...>]

Robert Story wrote:
> I'd be more impressed w/tests on Solaris x86 vs Solaric Sparc.
> 
> I can't see how just reversing the patch, which fixed some platform, is going
> to work w/out breaking that platform again.

Well, I think the original patch didn't fix that platform. It didn't
even change the behaviour on that platform at all (if you look at the
patch). IIRC it was based on a problem report on IRC and went in untested.

Now, here are my testing results on Solaris/x86:

Before and after the original patch:
UDP-MIB::udpLocalAddress.1.0.0.127.123 = IpAddress: 1.0.0.127
UDP-MIB::udpLocalAddress.1.0.0.127.8765 = IpAddress: 1.0.0.127

After my proposed patch:
UDP-MIB::udpLocalAddress.1.0.0.127.123 = IpAddress: 127.0.0.1
UDP-MIB::udpLocalAddress.1.0.0.127.8765 = IpAddress: 127.0.0.1

So after all my proposed patch doesn't promise to fix everything
(because a single simple change cannot), but is definitely an
improvement over 5.3.1.rc1 on all platforms.

Then, still, it'd need your votes.


+Thomas

-- 
Thomas Anders (thomas.anders at blue-cable.de)