getting a little carried away?

[Poul-Henning Kamp]

In message <86y8dxiavz.fsf at dwp.des.no>, =?iso-8859-1?q?Dag-Erling_Sm=F8rgrav?= 
writes:
>this box has a RAID-1 array with a single slice containing six
>partitions:
>
>des at dwp ~% ls /dev/ar0*
>/dev/ar0        /dev/ar0s1a     /dev/ar0s1c     /dev/ar0s1e     /dev/ar0s1g
>/dev/ar0s1      /dev/ar0s1b     /dev/ar0s1d     /dev/ar0s1f
>
>the components of ar0 are ad4 and ad6.  I realize that a component in
>a RAID-1 array "tastes the same" to GEOM as a standalone disk, but I
>think this is a bit much, especially the adMcsN stuff:
>
>des at dwp ~% ls /dev/ad[46]*
>/dev/ad4        /dev/ad4cs1f    /dev/ad4s1e     /dev/ad6cs1c    /dev/ad6s1b
>/dev/ad4a       /dev/ad4cs1g    /dev/ad4s1f     /dev/ad6cs1d    /dev/ad6s1c
>/dev/ad4b       /dev/ad4d       /dev/ad4s1g     /dev/ad6cs1e    /dev/ad6s1d
>/dev/ad4c       /dev/ad4e       /dev/ad6        /dev/ad6cs1f    /dev/ad6s1e
>/dev/ad4cs1     /dev/ad4f       /dev/ad6a       /dev/ad6cs1g    /dev/ad6s1f
>/dev/ad4cs1a    /dev/ad4s1      /dev/ad6b       /dev/ad6d       /dev/ad6s1g
>/dev/ad4cs1b    /dev/ad4s1a     /dev/ad6c       /dev/ad6e
>/dev/ad4cs1c    /dev/ad4s1b     /dev/ad6cs1     /dev/ad6f
>/dev/ad4cs1d    /dev/ad4s1c     /dev/ad6cs1a    /dev/ad6s1
>/dev/ad4cs1e    /dev/ad4s1d     /dev/ad6cs1b    /dev/ad6s1a
>
>is this normal?  more to the point, is it desirable?

yes and yes.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.