Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 29 Jul 2017 08:25:41 +0000
From:      bugzilla-noreply@freebsd.org
To:        freebsd-bugs@FreeBSD.org
Subject:   [Bug 221084] Loader fails to boot from vanilla ZFS version 28 pools (regression caused by MFC of r314112)
Message-ID:  <bug-221084-8@https.bugs.freebsd.org/bugzilla/>

next in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D221084

            Bug ID: 221084
           Summary: Loader fails to boot from vanilla ZFS version 28 pools
                    (regression caused by MFC of r314112)
           Product: Base System
           Version: 11.1-STABLE
          Hardware: Any
                OS: Any
            Status: New
          Keywords: regression
          Severity: Affects Some People
          Priority: ---
         Component: bin
          Assignee: freebsd-bugs@FreeBSD.org
          Reporter: fk@fabiankeil.de
                CC: mav@FreeBSD.org, tsoome@me.com

Created attachment 184823
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=3D184823&action=
=3Dedit
loader: Ignore check_mos_features()'s return code but log it

After updating a system based on r321478 to one based on r321624
the loader rejected vanilla ZFS version 28 pools with the message:

   ZFS: pool bpool is not supported

Pools that have been upgraded to contain features are accepted as expected.

The offending commits seems to be r321525 + r321526 (which MFC'd r314112).

For now I'm using the attached patch which works around the problem by
ignoring the check_mos_features() return code.

When using vanilla ZFS version 28 pools it results in the loader message:

   ZFS: check_mos_features() returned 2 for pool bpool. Continuing

Probably  check_mos_features() should only be called when features are pres=
ent.

I ran into the problem using bhyveload which relies on /boot/userboot.so.
While I expect the problem to be reproducible with the other loaders as wel=
l,
I haven't verified this.

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-221084-8>