sudo zpool create -O casesensitivity=insensitive -O normalization=formD -O ashift=12 POOLNAME mirror /dev/diskXsX /dev/diskXsX
..do you think I can mount that pool as read/writeable in 10.8/ZEVO to copy the data over?
Does anybody have any experience with this?
I have an extra machine on which I guess I can install any OS and try to import the ZEVO pool and send across network to the new 10.9/macZFS but I fear it will be a LOT more work and very slow transfer speeds.
Thanks in advance for any input.
Regards.
Robert
--
---
You received this message because you are subscribed to the Google Groups "zfs-macos" group.
To unsubscribe from this group and stop receiving emails from it, send an email to zfs-macos+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
<RobertRehnmarkPublicKey.gpgkey>
Hey, this is ilovezfs.I think there may be some confusion here as to what each of these versions are. For example, you referenced the command:
sudo zpool create -O casesensitivity=insensitive -O normalization=formD -O ashift=12 POOLNAME mirror /dev/diskXsX /dev/diskXsXwhich is not a command that would work on MacZFS 74.x.x.
MacZFS 74.3.x is the same as MacZFS 74.2.x, but simply upgraded to work with Mavericks.MacZFS 74.2.x/74.3.x is pool version 8, zfs file system version 2.My "zfsadm" script is not related to MacZFS 74. It is for the newer, OpenZFS based version of ZFS for OS X, which has been referred to variously as "ZFS-OSX," and "OSX-ZFS," "MacZFS 99," "MacZFS prototype generation," and "MacZFS experimental generation." The GitHub repository https://github.com/zfs-osx has the code for this version.https://github.com/zfs-osx (like every implementation of OpenZFS) is pool version 5000, file system version 5. For reference, ZEVO is pool version 28, file system version 5. There is a comparison table here: http://bitly.com/osxzfs
If you want to use MacZFS 74, I suggest that you run the commandsudo nvram boot-args="keepsyms=y"and reboot. Then you should install MacZFS 74.3.2b and reboot. If you get a kernel panic, post your panic report here. The panic reports are in /Library/Logs/DiagnosticReports.In my opinion, using MacZFS 74.3.1 is not a good answer, given that it has been fully superseded by MacZFS 74.3.2b. If there is a kernel panic in 74.3.2b that you have discovered, that must be dealt with not avoided. However, the kernel panic is likely attributable to a problem with your setup given that others are using 74.3.2b without a panic, in which case that problem in your setup needs to be resolved, not avoided.
If instead you want to use https://github.com/zfs-osx/zfs, then yes I'd recommend using https://gist.github.com/ilovezfs/7713854. A tutorial explaining one way it can be used is here: http://zerobsd.tumblr.com/post/62586498252/os-x-with-zfs.If you have questions, the IRC channel is #mac-zfs on freenode. Colloquy http://colloquy.info/ is a user-friendly client.
I understand that this might be hard or impossible to answer but:Can I use your script to compile and install OpenZFS, maybe even mount the ZEVO pools… and expect it to be stable?I would REALLY like to have features like case insensitivity and UTF8 normalization. Also it would be nice if the mounted filesystems were actually visible in volumes. :)So.. how far has it come?Can I use it for my iPhoto, iTunes and general file storage?Can it be scanned by Spotlight?
--
Can I use your script to compile and install OpenZFS
maybe even mount the ZEVO pools…
and expect it to be stable?
I would REALLY like to have features like case insensitivity and UTF8 normalization.
Also it would be nice if the mounted filesystems were actually visible in volumes.
Can I use it for my iPhoto, iTunes and general file storage?
Can it be scanned by Spotlight?
Thank you, if OpenZFS is not an option for everyday use then I will go for this.
Thank you!
... not be able to export the pool with a forced shutdown ...
--
--
No kissie, no kissie!!zpool create -f -O compression=lz4 -O casesensitivity=insensitive -O normalization=formD -O atime=off -o ashift=12 pool raidz disk1 disk2 disk3
--
Should I use the slices or just the whole disk when creating a new pool or attaching/adding disks?
I have another question though.Is it possible (and good practice) to turn off compression for a child filesystem?It will probably just make it slower instead of faster when I'm storing only pictures, highly compressed videos and mp3's on it, right?I have a hexa core Xeon X5680 running at 4,1 GHz so processing power is not a problem though.