Yeah, this make sense. We might also need to double check that bectl does the thing thing here when you clone the deep BE

In D35531#805896, @markj wrote:

There are some other thresholds that are derived from the free target. zfs_arc_free_target, vmd_pageout_wakeup_thresh and vmd_background_launder_target, at least. It's not obvious to me that the system will behave sanely if those values are left unchanged. Though, today it's possible to modify the per-domain free targets, and the sysctl handler doesn't take any of this into account. So I don't object to this diff, but it would be good to know why your customer wants to set this sysctl in the first place.

In D35509#805364, @lwhsu wrote:

Renaming to G sounds a possible way, what's the suitable character for doc and ports?

Try to trigger a build

Poke

Update

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Update manpage date

In D35248#798944, @markj wrote:

In D35248#798943, @imp wrote:

Shouldn't you reach over into the boot loader to get the shared files?
If there's some reason you can't do this, it should be documented and/or discussed.

Yes, there's no real reason not to. I just did copied them because I had to modify some of the shared files and wanted to keep things self-contained.

I also wonder if there's precedent for sharing code with the loader like that. Should it perhaps be moved to some common area, or should I just set .PATH=stand/libsa/zfs and not worry about it?

reviewed by: allanjude

Reviewed by: allanjude

reviewed by: allanjude

Reviewed by: allanjude

My only concern here is a user not understanding why the disks they know exist are not present in the list.
dialog doesn't really provide a way to present the 'disabled' disks, so I don't know there is much we can do about this.
Since it is likely the commands will fail for the disk that is in use, I think this is fine.

Reviewed by: allanjude

In D21939#568632, @jilles wrote:

This change may cause breakage in setups where the chroot does not have /etc/pwd.db and related files. Given that chroot(8) has worked like this since 2003, why change it now?

The text that used to be in the manual page may be a bit misleading but it is still correct, since chroot(2) requires root privileges and therefore the actual privilege change occurs after it. The lookup of UIDs and GIDs is not explicitly mentioned.

Swapping the order can be done with the existing tools by executing su(1) or another chroot(8) in the chroot. The former has the advantage of also setting various other items such as the login class.

Reviewed By: allanjude

Reviewed by: allanjude

yeah, --msgbox complains "Expected no more than 3 tokens for --msgbox, have 9." or however many words there were in your error message, plus the "0 0" parameters.

Don't sleep longer than required

Fix commit message, broken tail -F is on 13.0 not 12.2

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed By: allanjude

Reviewed by: allanjude

Reviewed By: allanjude

This came up again via: https://blog.guillaume-gomez.fr/articles/2021-12-14+sysinfo%3A+version+0.22+and+FreeBSD+support

Approved By: allanjude

Make sure to MFC this to stable/13 as well.

I think indenting the line continuations would help with the readability.

Reviewed By: allanjude

Nice improvement

Update to fix build error

In D33061#746999, @brnrd wrote:

Awaiting comments from upstream before committing. I'm assuming this will be incorporated in next releases too.

Committing version 1.1.1l_1,1 is too beautiful a chance to let is pass me by!

Update to the cleaner style patch

Update with cleaner patch and add configure scripts to enable armcap.c

In D28737#729430, @dan.kotowski_a9development.com wrote:

Ping - is there any work left to do here or can this be merged and closed?

Address review feedback from markj

Reviewed by: allanjude