Description of problem: I was just trying custom partitioning with vfat on /boot, I was warned that /boon cannot be vfat so i wanted to review (and edit) my partitioning. Names/mount-points were 'Unknown' and then I got traceback. Version-Release number of selected component: anaconda-19.30.12-1 The following was filed automatically by anaconda: anaconda 19.30.12-1 exception report Traceback (most recent call first): File "/usr/lib/python2.7/site-packages/blivet/devices.py", line 2368, in size size += max(0, self.align(pv.size - pv.format.peStart)) File "/usr/lib/python2.7/site-packages/blivet/devicefactory.py", line 408, in <lambda> containers.sort(key=lambda c: getattr(c, "freeSpace", c.size), File "/usr/lib/python2.7/site-packages/blivet/devicefactory.py", line 409, in get_container reverse=True) File "/usr/lib/python2.7/site-packages/pyanaconda/ui/gui/spokes/custom.py", line 1221, in _save_right_side container = factory.get_container() File "/usr/lib/python2.7/site-packages/pyanaconda/ui/gui/spokes/custom.py", line 2363, in _save_current_selector self._save_right_side(self._current_selector) File "/usr/lib/python2.7/site-packages/pyanaconda/ui/gui/spokes/custom.py", line 2374, in on_selector_clicked self._save_current_selector() File "/usr/lib/python2.7/site-packages/pyanaconda/ui/gui/spokes/lib/accordion.py", line 218, in _onSelectorClicked cb(selector) AttributeError: 'DeviceFormat' object has no attribute 'peStart' Additional info: cmdline: /usr/bin/python /sbin/anaconda cmdline_file: initrd=initrd.img inst.stage2=hd:LABEL=Fedora\x2019\x20i386 quiet BOOT_IMAGE=vmlinuz executable: /sbin/anaconda hashmarkername: anaconda kernel: 3.9.5-301.fc19.i686 product: Fedora release: Cannot get release name. type: anaconda version: 19
Created attachment 765486 [details] File: anaconda-tb
Created attachment 765488 [details] File: anaconda.log
Created attachment 765489 [details] File: environ
Created attachment 765490 [details] File: lsblk_output
Created attachment 765491 [details] File: nmcli_dev_list
Created attachment 765492 [details] File: os_info
Created attachment 765493 [details] File: program.log
Created attachment 765494 [details] File: storage.log
Created attachment 765496 [details] File: syslog
Created attachment 765497 [details] File: ifcfg.log
Created attachment 765498 [details] File: packaging.log
I was able to reproduce it with default partitioning. It appears after selecting 'I want to review/modify my disk partitions before continuing', clicking 'Continue' and then clicking on one of 'Unknown' partitions.
Created attachment 765521 [details] Video with bug reproducing
Bug appears only when radio button 'I want to review/modify my disk partitions before continuing' is selected and dialog with this option is shown only when disk is unselected, clicking 'Done' (no disks selected warning) then selected again and clicking 'Done'. When there is no disk unselecting, shown dialog does not contain such radio button option and button 'Custom partitioning' at the bottom of dialog works as expected.
I'd definitely give +1 FE to this one; if we have something else to fix in RC3, I'd be great to have this fixed as well. But I wouldn't block F19 on this single issue alone - the usual workflow (using "Custom Partitioning" button) works and this radio button workflow is hard to discover.
Steps to reproduce: 1. selected disk (or keep selected) 2. go to Custom Partitioning 3. remove all the existing partitions ("delete all...") 4. click to create the default layout 5. click Done 6. go to disk selection and unselect the disk 7. click Done 8. go back to disk selection and select the disk 9. click Done and choose 'I want to review/modify my disk partitions before continuing' 10. click on something on the weird Custom partitioning spoke that appears I think the problem is with unselecting and reselecting the disk that has some partitioning requests created. They are displayed in a weird way and obviously cause problems. I don't think this would be an easy fix. Though I wish I was wrong.
Probably -1 at this point too, if it's taken this long for someone to hit this it's unlikely to be too prevalent.
fyi, dlehman is testing fix now. I'm more inclined -1 blocker too, in case we would need RC3, I'd be +1 FE to include this one. It should be simple fix.
(In reply to Jaroslav Reznik from comment #18) > fyi, dlehman is testing fix now. I'm more inclined -1 blocker too, in case > we would need RC3, I'd be +1 FE to include this one. It should be simple fix. A complete fix would be quite complex -- I would say too complex to consider at this point for f19.
I tried to modify disk partitions for second time.. cmdline: /usr/bin/python /sbin/anaconda cmdline_file: initrd=initrd.img inst.stage2=hd:LABEL=Fedora\x2019\x20x86_64 rd.live.check quiet BOOT_IMAGE=vmlinuz hashmarkername: anaconda kernel: 3.9.5-301.fc19.x86_64 package: anaconda-19.30.13-1 product: Fedora reason: AttributeError: 'DeviceFormat' object has no attribute 'peStart' release: Cannot get release name. version: 19
This bug is still relevant in Fedora 20 Alpha RC1
Trying to create partitions during installation on dual-boot Windows 8 / Fedora machine cmdline: /usr/bin/python /sbin/anaconda cmdline_file: BOOT_IMAGE=/images/pxeboot/vmlinuz inst.stage2=hd:LABEL=Fedora\x2019\x20x86_64 quiet hashmarkername: anaconda kernel: 3.9.5-301.fc19.x86_64 package: anaconda-19.30.13-1 product: Fedora reason: AttributeError: 'DeviceFormat' object has no attribute 'peStart' release: Cannot get release name. version: 19
This bug is still relevant in Fedora 20 Beta RC5, proposing as beta blocker (alongside with bug 1027846) - violation of criterion: Custom partitioning (Reject or disallow invalid disk and volume configurations without crashing).
Discussed in the 2013-11-07 Go/No-Go meeting [1]. Voted as a RejectedBlocker as this use case was deemed too much of a corner case to justify blocking the release of Fedora 20 beta. [1] http://meetbot.fedoraproject.org/meetbot/meetbot/fedora-meeting-2/2013-11-07/
I wasn't able to reproduce this bug anymore with anaconda-20.25.8-1. In https://bugzilla.redhat.com/show_bug.cgi?id=978298#c16 , step 10 shows empty disk (as after step 3). This bug seems to be fixed if this is desired behavior.
(In reply to Lukas Brabec from comment #25) > empty disk (as after step 3). This bug seems to be fixed if this is desired > behavior. The desired behavior is that partitions you removed will still be gone but any new devices you had scheduled creation of prior to hiding the disk will be gone as well.
Ok, it behaves as expected.
anaconda 20.25.9-1.fc20 is stable, closing