Description of problem: Crash occurred during the openQA install_delete_pata test on BIOS - not UEFI - when testing Fedora-Rawhide-20160921.n.0 . That test uses a PATA disk (that is, qemu's emulation of a PATA disk) containing an existing partition, and uses guided partitioning to wipe the entire disk; you can see a video of the install process at https://openqa.fedoraproject.org/tests/35228/file/video.ogv . The exact same test is run on both BIOS and UEFI, and both variants run on both production and staging openQA. On both production and staging the BIOS variant of the test failed but the UEFI one passed, so it seems to be a real difference between BIOS and UEFI somehow, not some openQA randomness... Version-Release number of selected component: anaconda-26.3-1 The following was filed automatically by anaconda: anaconda 26.3-1 exception report Traceback (most recent call first): File "/usr/lib/python3.5/site-packages/blivet/formats/__init__.py", line 558, in _pre_destroy raise DeviceFormatError("device path does not exist or is not writable") File "/usr/lib/python3.5/site-packages/blivet/threads.py", line 45, in run_with_lock return m(*args, **kwargs) File "/usr/lib/python3.5/site-packages/blivet/formats/__init__.py", line 545, in destroy self._pre_destroy(**kwargs) File "/usr/lib/python3.5/site-packages/blivet/threads.py", line 45, in run_with_lock return m(*args, **kwargs) File "/usr/lib/python3.5/site-packages/blivet/deviceaction.py", line 700, in execute self.format.destroy() File "/usr/lib/python3.5/site-packages/blivet/threads.py", line 45, in run_with_lock return m(*args, **kwargs) File "/usr/lib/python3.5/site-packages/blivet/actionlist.py", line 325, in process action.execute(callbacks) File "/usr/lib/python3.5/site-packages/blivet/actionlist.py", line 48, in wrapped_func return func(obj, *args, **kwargs) File "/usr/lib/python3.5/site-packages/blivet/blivet.py", line 163, in do_it self.devicetree.actions.process(callbacks=callbacks, devices=self.devices) File "/usr/lib/python3.5/site-packages/blivet/threads.py", line 45, in run_with_lock return m(*args, **kwargs) File "/usr/lib/python3.5/site-packages/blivet/osinstall.py", line 1112, in turn_on_filesystems storage.do_it(callbacks) File "/usr/lib64/python3.5/site-packages/pyanaconda/install.py", line 204, in doInstall turn_on_filesystems(storage, mount_only=flags.flags.dirInstall, callbacks=callbacks_reg) File "/usr/lib64/python3.5/threading.py", line 862, in run self._target(*self._args, **self._kwargs) File "/usr/lib64/python3.5/site-packages/pyanaconda/threads.py", line 251, in run threading.Thread.run(self, *args, **kwargs) blivet.errors.DeviceFormatError: device path does not exist or is not writable Additional info: addons: com_redhat_kdump, com_redhat_docker cmdline: /usr/bin/python3 /sbin/anaconda cmdline_file: BOOT_IMAGE=vmlinuz initrd=initrd.img inst.stage2=hd:LABEL=Fedora-S-dvd-x86_64-rawh quiet executable: /sbin/anaconda hashmarkername: anaconda kernel: 4.8.0-0.rc7.git1.1.fc26.x86_64 product: Fedora release: Cannot get release name. type: anaconda version: rawhide
Created attachment 1203393 [details] File: anaconda-tb
Created attachment 1203394 [details] File: anaconda.log
Created attachment 1203395 [details] File: dnf.librepo.log
Created attachment 1203396 [details] File: environ
Created attachment 1203397 [details] File: hawkey.log
Created attachment 1203398 [details] File: lsblk_output
Created attachment 1203399 [details] File: nmcli_dev_list
Created attachment 1203400 [details] File: os_info
Created attachment 1203401 [details] File: storage.log
Created attachment 1203402 [details] File: syslog
Created attachment 1203403 [details] File: ifcfg.log
Created attachment 1203404 [details] File: packaging.log
Created attachment 1203405 [details] File: program.log
Proposing as a Beta blocker per criterion "When using the guided partitioning flow, the installer must be able to: ... Remove existing storage volumes to free up space, at the user's direction".
Updates image with proposed fix: https://dlehman.fedorapeople.org/updates/updates-1378162.0.img Please test and provide feedback at your convenience.
test should run as https://openqa.stg.fedoraproject.org/tests/44129 , when openQA grinds around to it.
Test passed, no other anaconda-ish test that passed before starting failing (except the other updates tests, because the updates image they were expecting wasn't used), so this fix looks good.
New updates image with v2 of proposed fix: https://dlehman.fedorapeople.org/updates/updates-1378162.1.img Please test and provide feedback at your convenience. The first version caused a failure when removing a second format on a disk that was scheduled for reformatting with a new disklabel.
Tested: https://openqa.stg.fedoraproject.org/tests/overview?distri=fedora&version=Rawhide&build=Fedora-Rawhide-20160921.n.0-EXTRA&groupid=1 https://openqa.stg.fedoraproject.org/tests/44497 looks good. The failures are not related to the commit (some base repo setup failures for tests that use the default repos, which just happens sometimes thanks to mirror sync problems, some existing unrelated fails, and some fails - the other updates.img tests, and the freeipa kickstart test that doesn't have working name resolution when it tries to download the updates image).
Upstream pull request: https://github.com/rhinstaller/blivet/pull/512
Whoops - this should be nominated as an F26 Beta blocker as it's affecting Rawhide, not F25. My bad.
Hah, and right after I posted #c21, it seems the new anaconda/blivet update *does* pull this bug into F25. I ran openQA on an image built with that update - https://bodhi.fedoraproject.org/updates/FEDORA-2016-bdc5976992 - and it hit this bug in one test: https://openqa.stg.fedoraproject.org/tests/46007 so looks like we will need this fix for F25 too, now. https://bodhi.fedoraproject.org/updates/FEDORA-2016-a8a52958b8 should fix it, I guess. I'm gonna do a bit of cheeky process short-circuiting here and make this an accepted blocker, because we discussed it at the blocker meeting this morning: https://meetbot-raw.fedoraproject.org/fedora-blocker-review/2016-09-26/f25-blocker-review.2016-09-26-16.04.log.html and it was going to be accepted as a blocker, right up until I noticed it wasn't affecting F25 at the time. So I think it's OK to take it as a blocker without re-voting.
python-blivet-2.1.5-1.fc25 has been pushed to the Fedora 25 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2016-a8a52958b8
python-blivet-2.1.5-1.fc25 has been pushed to the Fedora 25 stable repository. If problems still persist, please make note of it in this bug report.