June OpenZFS Leadership Meeting
Matthew Ahrens
mahrens at delphix.com
Thu Jun 27 21:21:57 UTC 2019
At this week's meeting we discussed FreeBSD/ZoL integration; error
reporting infrastructure; and using pyzfs to get zpool config information.
The video recording is now posted:
https://www.youtube.com/watch?v=TJwykiJmH0M
The next meeting will be at 11am US/Pacific on July 23rd.
Full notes below:
-
FreeBSD / ZoL integration update
-
Goal: Single repo that builds for both platforms
-
https://github.com/zfsonfreebsd/ZoF/tree/projects/zfsbsd/module
-
Goal is for FreeBSD 13 to use ZoF by default (but may slip to FreeBSD
14)
-
Goal to get it into current repo around end of September?
-
Jorgen: Where should I direct Windows/OSX questions? (#openzfs or ZoL
Developer list)
-
Do a PR for additional platforms/tree for additional platforms
-
Change zfsonlinux/zfs => openzfs/zfs. Brian & Matt will come up with
a proposal for naming and structure that reflects what developers do/use
-
Announcements of big, long-standing projects being available:
-
Linux - Redacted send/recv merged!
-
Illumos - ZFS crypto landed!
-
Discuss error reporting (Tom)
-
Do we want to move to something more general (e.g., have the kernel
return strings)?
-
Continue to extend zfs_errno_t, though this doesn’t help provide
additional context/information, and then add new strings for user
-
pyzfs PR <https://github.com/zfsonlinux/zfs/pull/8956> (Richard E)
-
Specifically how to get the pool config info programmatically
-
Programmatic “zpool status” that is retrievable, and any time you add
something to zpool you need to make sure status continues to be
retrievable. (some mistake that Joshua pointed out...)
-
Translation of the config to a stable nvlist? Not sure if this should
be in libzfs_core or the kernel. Kernel is better but might
ignore backward
compatibility.
-
Newer version of userland continue to work without rebooting the
kernel.
-
Could a spanning mode option be added for multi-vdev zpools instead of
the default striping data across all vdevs? The intent is to prioritize
data integrity over access speed for large archival / backup pools, and a
workaround with multiple pools joined with mergerfs seems suboptimal if the
LOE of adding support for this is low. (Guirara DaiKaiju)
-
Matt: I think you’re suggesting that `zpool create disk1 disk2 …`
create the pool as a mirror, as though you typed `zpool create
mirror disk1
disk 2 …`. And perhaps that `zpool add…` should instead do `zpool attach
…`.
On Tue, Jun 25, 2019 at 9:56 AM Matthew Ahrens <mahrens at delphix.com> wrote:
> The next OpenZFS Leadership meeting will be held today, June 25, 1pm-2pm
> Pacific time. We don't have much on the agenda for this month, so if there
> are things you'd like to discuss, this is a good opportunity - send me an
> email or edit the doc :)
>
> Everyone is welcome to attend and participate, and we will try to keep the
> meeting on agenda and on time. The meetings will be held online via Zoom,
> and recorded and posted to the website and YouTube after the meeting.
>
> The agenda for the meeting will be a discussion of the projects listed in
> the agenda doc.
>
> For more information and details on how to attend, as well as notes and
> video from the previous meeting, please see the agenda document:
>
>
> https://docs.google.com/document/d/1w2jv2XVYFmBVvG1EGf-9A5HBVsjAYoLIFZAnWHhV-BM/edit
>
>
> <https://docs.google.com/document/d/1w2jv2XVYFmBVvG1EGf-9A5HBVsjAYoLIFZAnWHhV-BM/edit>
> --matt
>
More information about the freebsd-fs
mailing list