On 06/05/2015 08:44 AM, Michael Felt wrote: [your top-posting is still making it harder to reply]
> Well, in terms of compatibility, for people like me who have used the AIX > option for years (it goes back to 1996-1997 when ACLs became mainstream in > AIX (4.2 at least, if not 4.1.X). > > Further, I not comprehend why coreutils shows a + on a directory (with -ld) > but not on a file. Makes me wonder what it is actually reporting on. Most likely because you have a case where the directory has a non-defalut ACL but the file does not. In fact, it's quite common for directories to have more ACLs than files, because some of those ACLs are used for setting the default permissions on files created in the directory. > > Guess I need to dig into what (linux) xattrs are. I am assuming something > not in AIX - under that label at least. xattrs can include more than ACLs; and meanwhile, while ACLs are often implemented by xattrs they can also be implemented in other means. Which is why libvirt shows '.' for the presence of xattrs that don't affect ACL. On a Linux system, look at the output of 'getfacl' on a directory and file, where the directory shows with '+' but the file does not, to compare the two different ACL settings. >>> root@x064:[/usr/bin]/usr/bin/ls -l /usr/bin/ls >>> -r-xr-xr-x 1 bin bin 28256 Mar 10 13:44 /usr/bin/ls >>> root@x064:[/usr/bin]/usr/bin/ls -e /usr/bin/ls >>> -r-xr-xr-x- 1 bin bin 28256 Mar 10 13:44 /usr/bin/ls >> >> Note that GNU ls already uses the 11th character as '.' (xattrs present) >> or '+' (ACLs present), adding '-' as (ACLs present but disabled) could >> indeed be a possible extension, even without needing to burn '-e' to get >> it. But '-e' is still available, so we could indeed use it. >> -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
