[Gluster-devel] question on crypt xlator

[Edward Shishkin]

On Wed, 15 Oct 2014 04:22:37 +0200
manu at netbsd.org (Emmanuel Dreyfus) wrote:


Hello Emmanuel,


> Edward Shishkin <edward at redhat.com> wrote:
> 
> > Thanks for the report,
> > I am still not sure that symlinks are the culprit.
> 
> Um, you link on a symlink? That could be the answer: Linux links on
> the symlink while BSD links on the target (both behavior are POSIX
> compliant): Here is BSD hehavior:
> 
> # ls -l a b
> -rw-r--r--  1 root  wheel  0 Oct 15 04:19 a
> lrwxr-xr-x  1 root  wheel  1 Oct 15 04:20 b -> a
> # ln b c
> # ls -li a b c
> 579242 -rw-r--r--  2 root  wheel  0 Oct 15 04:19 a
> 579244 lrwxr-xr-x  1 root  wheel  1 Oct 15 04:20 b -> a
> 579242 -rw-r--r--  2 root  wheel  0 Oct 15 04:19 c
> 


This is a really interesting difference, that I didn't know about :)
However, in our crypt.t we don't operate on symlinks except symlink(2)
and open(2) after remount..


> 
> > Could you please try the attached version of crypt.t and report
> > results?
> 
> It passes.


I am sorry, but it seems that the fastest option would be to pass
me root permissions on your machine. So I can set gdb breakpoints
to find out what is going wrong. Please, explore such possibility,
and let me know the result.

Thanks,
Edward.