Files created with vsphere on a seperate ZFS dataset are not visable from cli within FreeBSD itself

Johan Hendriks joh.hendriks at gmail.com
Fri Aug 16 15:22:07 UTC 2013


Home Email schreef:
>> Or is it that the files are in ESXishare but masked by the machine1 
>> mount.
>
> Yes, that is exactly the issue.
>
> On the server:
>
> Writing to /storage/ESXishare/ will place files on the 
> storage/ESXishare filesystem
> Writing to /storage/ESXishare/machine1/ will place files on the 
> storage/ESXishare/machine1 filesystem (which is mounted over the top 
> of a machine1 dir on the parent fs)
>
> On ESXi:
>
> Writing to ESXishare will write to the mounted filesystem 
> (storage/ESXishare)
> Writing to ESXishare/machine1 will write to the machine1 subdirectory 
> on this filesystem
>
> NFS mounts will not cross filesystem mount points.
>
> It would be nice to be able to do what you are trying to - it would 
> also allow the ability to use zfs clones to duplicate VMs. 
> Unfortunately, you would need to export and  mount each machine's ZFS 
> filesystem separately in ESXi.
>
>> One more thing.
>> I use thin provisioning.
>> But if i use cp to copy files they grow to the actual size.
>> How do i overcome that?
>
> As far as I'm aware FreeBSD cp still doesn't have a sparse option. The 
> way I do it is to use rsync with the --sparse option. There may be an 
> easiest way but I like rsync and use it for various other tasks so 
> generally have it installed anyway.
>
> --Matt
>
> _______________________________________________
> freebsd-fs at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-fs
> To unsubscribe, send any mail to "freebsd-fs-unsubscribe at freebsd.org"
I did some testing today, and after reading and browsing the internet it 
is not possible to use a per machine zfs dataset with one single mount 
and export.
It looks like NFSv4 does support a single mount.

I go another route and use two mounts, one wich will be snapshotted 
daily and one less often.
This way i have two datasets that i can work with. But it would be 
really awesome if it did work.
It would give even more flexibility to the whole system.

Thanks all for your time..

gr
Johan


More information about the freebsd-fs mailing list