Bug#651719: marked as done (new ZFS installs completely broken in Wheezy/Sid)
Your message dated Sun, 11 Dec 2011 18:43:40 +0100
with message-id <CAOfDtXPdZvyHEcAi+4fLB49n0xAHC2U57kEPJWmwNLOWW93UfA@mail.gmail.com>
and subject line not a bug
has caused the Debian Bug report #651719,
regarding new ZFS installs completely broken in Wheezy/Sid
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)
--
651719: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=651719
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message ---
Package: zfsutils
Version: 8.3~svn226546-6
2011/12/11 Robert Millan <rmh@debian.org>:
> 2.2- zfsutils 8.2 in /target with a kfreebsd-9 runtime breaks
> grub-install (even if zfsutils-udeb is 8.3). Will be fixed when
> zfsutils 8.3 migrates.
Actually, this is much worse. Even 8.3 zfsutils (or zfsutils-udeb)
has the same problem (it confuses a partition with its container), the
only difference is that zfsutils 8.2 prints broken "zpool status"
output, whereas zfsutils 8.3 doesn't. Breakage in 8.3 can be exposed
with:
$ zpool create test /dev/ada0s1
$ zpool export test
$ zpool import
pool: test
....
state: FAULTED
status: One or mode devices contains corrupted data.
...
config:
test FAULTED corrupted data
ada0 UNAVAIL corrupted data
--
Robert Millan
--- End Message ---
--- Begin Message ---
- To: 651719-done@bugs.debian.org
- Subject: not a bug
- From: Robert Millan <rmh@debian.org>
- Date: Sun, 11 Dec 2011 18:43:40 +0100
- Message-id: <CAOfDtXPdZvyHEcAi+4fLB49n0xAHC2U57kEPJWmwNLOWW93UfA@mail.gmail.com>
It seems that the "zpool import" hack I introduced in the process of
fixing #651720 was picking up remnants of a previous pool with the
same name, which was indeed broken. I worked around it by creating a
clean pool on ada0 and then destroying it.
The fix I'm committing to partman-zfs will use GUIDs, avoiding this problem.
--
Robert Millan
--- End Message ---
Reply to: