Re: more on proff crash building native

["Michael L. Hitch" <mhitch%lightning.msu.montana.edu@localhost>]

On Sat, 14 Feb 2009, David Brownlee wrote:

>         might vary. How hard is this error to trigger? Can someone who
>         has a repeatable test case try it over ssh localhost to eliminate
>         the network card, or possibly on a vax with a different network
>         chipset?

vlc1$ for x in 1 2 3 4 5;do scp localhost:/netbsd-generic 
/tmp/netbsd-$x;md5 /tmp/netbsd-$x;rm /tmp/netbsd-$x;done
netbsd-generic                                100% 2803KB 121.9KB/s 
286.2KB/s   00:23
MD5 (/tmp/netbsd-1) = 814d47aafafe88fffcf467277a6fcd9f
netbsd-generic                                100% 2803KB 121.9KB/s 
105.7KB/s   00:23
MD5 (/tmp/netbsd-2) = a12df2d62ec1b08833fc948daf7d766d
netbsd-generic                                100% 2803KB 116.8KB/s 
110.2KB/s   00:24
MD5 (/tmp/netbsd-3) = a12df2d62ec1b08833fc948daf7d766d
netbsd-generic                                100% 2803KB 116.8KB/s 
121.7KB/s   00:24
MD5 (/tmp/netbsd-4) = a12df2d62ec1b08833fc948daf7d766d
netbsd-generic                                100% 2803KB 116.8KB/s 
53.2KB/s   00:24
MD5 (/tmp/netbsd-5) = a12df2d62ec1b08833fc948daf7d766d

  This failure rate is lower than I saw copying from a remote host.  In 
that case, every copy was bad (and each difference varied quite a bit in 
number and location, i.e. nothing consistent).

--
Michael L. Hitch                        mhitch%montana.edu@localhost
Computer Consultant
Information Technology Center
Montana State University        Bozeman, MT     USA