Re: [developer] raidz overhead with ashift=12

For the archives, a fix has been integrated into illumos-gate via illumos#9318. https://github.com/illumos/illumos-gate/commit/b73ccab03ec36581b1ae5945ef1fee1d06c79ccf Thanks to all that helped me with this. Mike -- openzfs: openzfs-developer Permalink:

Re: [developer] raidz overhead with ashift=12

On Fri, Jun 7, 2019 at 11:06 AM Eric Borisch wrote: > On Fri, Jun 7, 2019 at 12:03 PM Matthew Ahrens > wrote: > >> The spreadsheet shows how much space will be allocated, which is >> reflected in the zpool `allocated` property. However, you are looking at >> the zfs `used` and `referenced`

Re: [developer] raidz overhead with ashift=12

> On Jun 7, 2019, at 12:15 PM, Mike Gerdts wrote: > > On Fri, Jun 7, 2019 at 12:03 PM Matthew Ahrens > wrote: > On Thu, Jun 6, 2019 at 10:56 PM Mike Gerdts > wrote: > I'm motivated to make zfs set refreservation=auto do the right

Re: [developer] raidz overhead with ashift=12

On Fri, Jun 7, 2019 at 12:03 PM Matthew Ahrens wrote: > On Thu, Jun 6, 2019 at 10:56 PM Mike Gerdts > wrote: > >> I'm motivated to make zfs set refreservation=auto do the right thing in >> the face of raidz and 4k physical blocks, but have data points that provide >> inconsistent data.

Re: [developer] raidz overhead with ashift=12

On Fri, Jun 7, 2019 at 12:03 PM Matthew Ahrens wrote: > The spreadsheet shows how much space will be allocated, which is reflected > in the zpool `allocated` property. However, you are looking at the zfs > `used` and `referenced` properties. These properties (as well as > `available` and all

Re: [developer] raidz overhead with ashift=12

On Thu, Jun 6, 2019 at 10:56 PM Mike Gerdts wrote: > I'm motivated to make zfs set refreservation=auto do the right thing in > the face of raidz and 4k physical blocks, but have data points that provide > inconsistent data. Experimentation shows raidz2 parity overhead that > matches my

Re: [developer] raidz overhead with ashift=12

> On Jun 6, 2019, at 10:54 PM, Mike Gerdts wrote: > > I'm motivated to make zfs set refreservation=auto do the right thing in the > face of raidz and 4k physical blocks, but have data points that provide > inconsistent data. Experimentation shows raidz2 parity overhead that matches > my

[developer] raidz overhead with ashift=12

I'm motivated to make zfs set refreservation=auto do the right thing in the face of raidz and 4k physical blocks, but have data points that provide inconsistent data. Experimentation shows raidz2 parity overhead that matches my expectations for raidz1. Let's consider the case of a pool with 8