> On Dec 2, 2015, at 10:15 AM, Dan McDonald <[email protected]> wrote:
>
> The change I have *INSIDE* zfs_receive_one() can be discarded (around lines
> 2929-2960) and replaced with suppression of errors for "refreservation"
> around lines 3276 or so for the non-finalsnap snapshots. Does that make
> sense to you? I will code it up in the interim.
First off, I meant to say "refquota" not "refreservation".
Second off, I'm running zfs-test on a new set of changes which I hope address
your concerns in libzfs, Matt. Here are the relevant webrevs:
The original pre-review webrev:
http://kebe.com/~danmcd/webrevs/4986-orig/
The new, improved overall webrev:
http://kebe.com/~danmcd/webrevs/4986/
And the incremental diffs between the original and the overall:
http://kebe.com/~danmcd/webrevs/4986-reviewonly/
Hopefully zfs-test will bear out I've broken nothing else. That'll take an
hour, however, and I've an appointment in an hour, so my latency will be a bit
high until later this afternoon.
Dan
_______________________________________________
developer mailing list
[email protected]
http://lists.open-zfs.org/mailman/listinfo/developer
