Bug 953329

Created attachment 737055 [details]
File: anaconda-tb

Created attachment 737056 [details]
File: anaconda.log

Created attachment 737057 [details]
File: backtrace

Created attachment 737058 [details]
File: environ

Created attachment 737059 [details]
File: ifcfg.log

Created attachment 737060 [details]
File: lsblk_output

Created attachment 737061 [details]
File: nmcli_dev_list

Created attachment 737062 [details]
File: packaging.log

Created attachment 737063 [details]
File: program.log

Created attachment 737064 [details]
File: storage.log

Created attachment 737065 [details]
File: syslog

Description of problem:
Install F19 Alpha RC4 DVD (on USB stick) to a blank SATA disk (was previously part of a RAID array, RAID array was destroyed immediately prior to install, so disk should have been left totally blank).

Version-Release number of selected component:
anaconda-19.20-1

Additional info:
cmdline:        /usr/bin/python  /sbin/anaconda
cmdline_file:   initrd=initrd.img inst.stage2=hd:LABEL=Fedora\x2019-Alpha\x20x86_64 xdriver=vesa nomodeset quiet slub_debug=- BOOT_IMAGE=vmlinuz 
executable:     /sbin/anaconda
hashmarkername: anaconda
kernel:         3.9.0-0.rc6.git2.3.fc19.x86_64
product:        Fedora
release:        Cannot get release name.
type:           anaconda
version:        19-Alpha

Truncated backtrace:
Traceback (most recent call last):
  File "/usr/lib64/python2.7/site-packages/pyanaconda/threads.py", line 141, in run
    threading.Thread.run(self, *args, **kwargs)
  File "/usr/lib64/python2.7/threading.py", line 504, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/usr/lib64/python2.7/site-packages/pyanaconda/install.py", line 119, in doInstall
    turnOnFilesystems(storage, mountOnly=flags.flags.dirInstall)
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 162, in turnOnFilesystems
    storage.doIt()
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 303, in doIt
    self.devicetree.processActions()
  File "/usr/lib/python2.7/site-packages/blivet/devicetree.py", line 237, in processActions
    action.execute()
  File "/usr/lib/python2.7/site-packages/blivet/deviceaction.py", line 272, in execute
    self.device.create()
  File "/usr/lib/python2.7/site-packages/blivet/devices.py", line 791, in create
    self._postCreate()
  File "/usr/lib/python2.7/site-packages/blivet/devices.py", line 1489, in _postCreate
    DeviceFormat(device=self.path, exists=True).destroy()
  File "/usr/lib/python2.7/site-packages/blivet/formats/__init__.py", line 326, in destroy
    raise FormatDestroyError(msg)
FormatDestroyError: error wiping old signatures from /dev/sda2: 1

Both times I hit this bug, I could reboot, try again, and install successfully. On the second attempt, it is apparent that the first attempt successfully created some partitions - the disk shows as fully occupied and I have to go through 'Reclaim Space'. But this second time, the formatting stage succeeds and installation proceeds.

bcl asks: " It would be interesting to see what parted -s /dev/vda2 u s p showed after a failure."

I guess I'll add the -f unconditionally so the world can keep on turning until wipefs gets its mind right. Pretty tired of this kind of stuff.

Description of problem:
clean installation. nothing special

Version-Release number of selected component:
anaconda-19.20-1

Additional info:
cmdline:        /usr/bin/python  /sbin/anaconda
cmdline_file:   initrd=initrd.img inst.stage2=hd:LABEL=Fedora\x2019-Alpha\x20x86_64 quiet BOOT_IMAGE=vmlinuz 
executable:     /sbin/anaconda
hashmarkername: anaconda
kernel:         3.9.0-0.rc6.git2.3.fc19.x86_64
product:        Fedora
release:        Cannot get release name.
type:           anaconda
version:        19-Alpha

Truncated backtrace:
Traceback (most recent call last):
  File "/usr/lib64/python2.7/site-packages/pyanaconda/threads.py", line 141, in run
    threading.Thread.run(self, *args, **kwargs)
  File "/usr/lib64/python2.7/threading.py", line 504, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/usr/lib64/python2.7/site-packages/pyanaconda/install.py", line 119, in doInstall
    turnOnFilesystems(storage, mountOnly=flags.flags.dirInstall)
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 162, in turnOnFilesystems
    storage.doIt()
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 303, in doIt
    self.devicetree.processActions()
  File "/usr/lib/python2.7/site-packages/blivet/devicetree.py", line 237, in processActions
    action.execute()
  File "/usr/lib/python2.7/site-packages/blivet/deviceaction.py", line 272, in execute
    self.device.create()
  File "/usr/lib/python2.7/site-packages/blivet/devices.py", line 791, in create
    self._postCreate()
  File "/usr/lib/python2.7/site-packages/blivet/devices.py", line 1489, in _postCreate
    DeviceFormat(device=self.path, exists=True).destroy()
  File "/usr/lib/python2.7/site-packages/blivet/formats/__init__.py", line 326, in destroy
    raise FormatDestroyError(msg)
FormatDestroyError: error wiping old signatures from /dev/sda2: 1

anaconda-19.24-1.fc19, python-blivet-0.12-1.fc19 has been submitted as an update for Fedora 19.
https://admin.fedoraproject.org/updates/python-blivet-0.12-1.fc19,anaconda-19.24-1.fc19

Package anaconda-19.24-1.fc19, python-blivet-0.12-1.fc19:
* should fix your issue,
* was pushed to the Fedora 19 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing anaconda-19.24-1.fc19 python-blivet-0.12-1.fc19'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2013-7403/python-blivet-0.12-1.fc19,anaconda-19.24-1.fc19
then log in and leave karma (feedback).

Description of problem:
I just tried to add a user during the install process.

Version-Release number of selected component:
anaconda-19.20-1

Additional info:
cmdline:        /usr/bin/python  /sbin/anaconda
cmdline_file:   initrd=/distrotrees/62656/initrd method=http://download.devel.redhat.com/released/F-19/Alpha/Fedora/i386/os/ repo=http://download.devel.redhat.com/released/F-19/Alpha/Fedora/i386/os/  BOOT_IMAGE=/distrotrees/62656/kernel 
executable:     /sbin/anaconda
hashmarkername: anaconda
kernel:         3.9.0-0.rc6.git2.3.fc19.i686
product:        Fedora
release:        Cannot get release name.
type:           anaconda
version:        19-Alpha

Truncated backtrace:
Traceback (most recent call last):
  File "/usr/lib/python2.7/site-packages/pyanaconda/threads.py", line 141, in run
    threading.Thread.run(self, *args, **kwargs)
  File "/usr/lib/python2.7/threading.py", line 504, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/usr/lib/python2.7/site-packages/pyanaconda/install.py", line 119, in doInstall
    turnOnFilesystems(storage, mountOnly=flags.flags.dirInstall)
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 162, in turnOnFilesystems
    storage.doIt()
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 303, in doIt
    self.devicetree.processActions()
  File "/usr/lib/python2.7/site-packages/blivet/devicetree.py", line 237, in processActions
    action.execute()
  File "/usr/lib/python2.7/site-packages/blivet/deviceaction.py", line 272, in execute
    self.device.create()
  File "/usr/lib/python2.7/site-packages/blivet/devices.py", line 791, in create
    self._postCreate()
  File "/usr/lib/python2.7/site-packages/blivet/devices.py", line 1489, in _postCreate
    DeviceFormat(device=self.path, exists=True).destroy()
  File "/usr/lib/python2.7/site-packages/blivet/formats/__init__.py", line 326, in destroy
    raise FormatDestroyError(msg)
FormatDestroyError: error wiping old signatures from /dev/sda2: 1

This problem persists in Fedora 19 Beta TC3.

dean: can you attach your /tmp/program.log from an install that reproduces this problem? Thanks.

I do not know how to retrieve files from a virtual machine whose build failed. Please advise.

I was left at a prompt that asks:

  What do you want to do now?
  1) Report Bug
  2) Debug
  3) Quit

  Please make your choice from above:

Created attachment 746374 [details]
screen shot of the anaconda main display

Created attachment 746376 [details]
screen shot of the program log display

Dean: you can hit ctrl-alt-f2 to get a console. If your VM has a network connection, you can fpaste the files out from there. Or you can scp them to the host system.

Created attachment 746473 [details]
/tmp/program.log

So that program.log shows wipefs was called with -f each time. bcl says that's the best anaconda can do, and if it still fails after that, it's wipefs' problem. So, re-assigning.

As I was trying to reproduce the bug to get program.log, the install was successful twice before it failed again?!?!?!

  cat >$path/anaconda-ks.cfg <<EOF
auth       --enableshadow --passalgo=sha512
autopart   --type=lvm
bootloader --location=mbr --boot-drive=$disk
cdrom
clearpart  --all --initlabel --drives=$disk
firewall   --enable --service=mdns,ssh
firstboot  --disable
ignoredisk --only-use=vda
keyboard   --vckeymap=us --xlayouts='us'
lang         en_US.UTF-8
network    --activate --bootproto=static --device=eth0 --gateway=192.168.1.254 --ip=192.168.1.$ip --nameserver=$nameserver --netmask=255.255.255.0 --noipv6
network    --hostname=$host.hunter.org
reboot
rootpw     --iscrypted \$6\$6YuUd3dm5L5ALg25\$mOot78r5I7kNFPhNRcOuFmHHS3FO.vfetR.z7K6LfGEmIk6mMlWh72coTpbiPna5puEDDPvNYngGnFL3KCvGo.
text
timezone     America/Chicago --isUtc --ntpserver=192.168.1.10
user       --gecos="Local Administrator" --groups=wheel --iscrypted --name=local --password=\$6\$6YuUd3dm5L5ALg25\$mOot78r5I7kNFPhNRcOuFmHHS3FO.vfetR.z7K6LfGEmIk6mMlWh72coTpbiPna5puEDDPvNYngGnFL3KCvGo.
xconfig    --startxonboot

%packages
@base-x
@core
@dial-up
@firefox
@fonts
@gnome-desktop
@hardware-support
@input-methods
@multimedia
@printing
@standard
%end
EOF

  virsh destroy $domain
  virsh undefine $domain
  virsh vol-delete $domain --pool Guests
  virsh vol-create-as --name $domain --pool Guests --capacity $capacity

  virt-install \
    --autostart \
    --channel unix,path=/var/lib/libvirt/qemu/guest.agent,mode=bind,target_type=virtio,name=org.qemu.guest_agent.0 \
    --connect qemu:///system \
    --disk vol=Guests/$domain \
    --extra-args "inst.ks=file:/anaconda-ks.cfg" \
    --graphics spice \
    --initrd-inject $path/anaconda-ks.cfg \
    --location $location \
    --name $domain \
    --network network=Host \
    --noautoconsole \
    --os-type "linux" \
    --os-variant "fedora18" \
    --ram 2048 \
    --vcpus 2

This problem persists in Fedora 19 Beta TC4.

(In reply to comment #27)
> So that program.log shows wipefs was called with -f each time. bcl says
> that's the best anaconda can do, and if it still fails after that, it's
> wipefs' problem. So, re-assigning.

According to comment #9 it seems that anaconda tries to erase actively used device (so you get EBUSY), the device /dev/vda2 is used by DM. In this case -f for wipefs(8) is workaround rather than a real bugfix.


I don't see any wipefs(8) error message in the program.log from comment #26, but I see:

Running... lvm pvcreate --config  devices { filter=["r|/loop3$|","r|/loop4$|","r|/loop5$|","r|/loop6$|","r|/loop7$|"] }  --dataalignment 1024k /dev/vda2
Can't initialize physical volume "/dev/vda2" of volume group "fedora_fedora19" without -ff

It would be also nice to call lsblk(8) here to get more infomation. I guess that lvm failed because the device is still used by another lvm instance.

Anyway, I need more information to fix anything in wipefs(8). From my point of view it works as expected. Sorry.

The problem occurs in an alternating (odd/even) pattern; every odd time it fails and every even time is succeeds.

I am sorry, but after further review I am now getting a different error:

03:59:36,249 INFO program: Running... lvm pvcreate --config  devices { filter=["r|/loop3$|","r|/loop4$|","r|/loop5$|","r|/loop6$|","r|/loop7$|"] }  --dataalignment 1024k /dev/vda2
03:59:36,314 INFO program:   Can't initialize physical volume "/dev/vda2" of volume group "fedora_fedora19" without -ff
03:59:36,317 DEBUG program: Return code: 5

Please disregard my comments from #20 on and I apologize for my lack of attention.

Thanks, Dean. Can you file a new bug for that problem? File against anaconda, with all logs attached. Thanks again!