For my application, everything would have to be automatically done out of the box. I have no issues with this personally, I even use FreeBSD as my main workstation, so I have the time and know-how.
I just can't see myself taking the time to train a bunch of admins nationally to deploy a technology that requires more than just scripting, but a background knowledge on what's going on.
I'd like to use Solaris/ZFS to replace a ton of SANs from EMC^2.
My issue with ZFS is that ZFS, according to the internet, is not auto-self-healing, an admin is required to trigger self-healing.
Drive drops, system continues automatically, (yes?). Drive gets replaced, drive requires system admin to trigger the new drive into the zpool again. That is the only issue right now.
Zfs is self healing. You can initiate it via zpool scrub command assuming you have redundancy.. You can then script it and put it in a cron job to run weekly/monthly.
If you use spares you can also avoid sys admin intervention to replace doggy disk, until its actually replaced. Then you just use the zpool replace command.
For my application, everything would have to be automatically done out of the box. I have no issues with this personally, I even use FreeBSD as my main workstation, so I have the time and know-how.
I just can't see myself taking the time to train a bunch of admins nationally to deploy a technology that requires more than just scripting, but a background knowledge on what's going on.
I'd like to use Solaris/ZFS to replace a ton of SANs from EMC^2.
pcfxer 2 years ago
Tough crowd ;).
My issue with ZFS is that ZFS, according to the internet, is not auto-self-healing, an admin is required to trigger self-healing.
Drive drops, system continues automatically, (yes?). Drive gets replaced, drive requires system admin to trigger the new drive into the zpool again. That is the only issue right now.
pcfxer 3 years ago
Zfs is self healing. You can initiate it via zpool scrub command assuming you have redundancy.. You can then script it and put it in a cron job to run weekly/monthly.
If you use spares you can also avoid sys admin intervention to replace doggy disk, until its actually replaced. Then you just use the zpool replace command.
damoos 2 years ago