Edward,
/etc/zpool.cache contains data pointing to devices involved in a
zpool. Changes to ZFS datasets are reflected in the actual zpool so
destroying a zfs dataset should not change zpool.cache.
zfs destroy is the correct command to destroy a file system.
It will be easier if we can know
- the output of ''zfs list'' at various stages
- the command he executed that failed.
--
Just me,
Wire ...
On 9/26/06, Edward Wetmore <Ed.Wetmore at sun.com>
wrote:> Good day all. Please respond to me directly as I am not on this alias.
> I have a customer who is develping his site''s implementation
of zfs,
> my case come to me because he is using Solaris 10 6/06 x86 on a Sun Fire
> V40z (an x86 unit). He had no problem assembling and mounting a zfs
> volume, the command set he got worked fine. He then then began
> preperation of documenting and scripting, which meant he wanted to take
> the zfs volume apart and then recreate it.
> He ran "zfsdestroy <filesystem_name>" (that was all he
was informed
> to do), but when he tried to build it again, the system indicated the
> file system was in use. He has found he has a /etc/zfs/zpool.cache.
> My question is (pardon my ignorance), what steps does he need to
> take to completely eliminate evidence that he previously had a zfs file
> system so that he can then build one again?
> Thanks for your time,
> Ed Wetmore
> AltPlat/Install/OS Tech Support Engineer
> Sun Microsystems
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
>