On Tue, 2015-09-22 at 15:06 -0400, Samantha N. Bueno wrote:
[I feel like this was fixed before, but I think it was something
else
that I had to rip out later since it caused problems.
Anyway.]
Wasn't it the call of the execute() method?
Otherwise the installer attempts to work with disks it thinks have a size
of 0B (size reported of an unformatted disk), which causes problems.
Resolves: rhbz#1264411
---
pyanaconda/ui/gui/spokes/lib/dasdfmt.py | 5 +++++
pyanaconda/ui/gui/spokes/storage.py | 1 +
pyanaconda/ui/tui/spokes/storage.py | 7 ++++++-
3 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/pyanaconda/ui/gui/spokes/lib/dasdfmt.py
b/pyanaconda/ui/gui/spokes/lib/dasdfmt.py
index 1b94705..e1d3df2 100644
--- a/pyanaconda/ui/gui/spokes/lib/dasdfmt.py
+++ b/pyanaconda/ui/gui/spokes/lib/dasdfmt.py
@@ -27,6 +27,7 @@ from pyanaconda import constants
from pyanaconda.i18n import _
import threading
+from blivet import storageInitialize
from blivet.devicelibs.dasd import format_dasd
from blivet.errors import DasdFormatError
@@ -112,6 +113,10 @@ class DasdFormatDialog(GUIObject):
threadMgr.add(AnacondaThread(name=constants.THREAD_DASDFMT,
target=self.run_dasdfmt, args=(self._epoch,)))
+ # Need to make devicetree aware of storage change
+ threadMgr.add(AnacondaThread(name=constants.THREAD_STORAGE,
target=storageInitialize,
+ args=(self.storage, self.data,
self.storage.devicetree.protectedDevNames)))
+
@gtk_action_wait
def update_dialog(self, epoch_started):
"""
diff --git a/pyanaconda/ui/gui/spokes/storage.py b/pyanaconda/ui/gui/spokes/storage.py
index ca5d30b..155fe7e 100644
--- a/pyanaconda/ui/gui/spokes/storage.py
+++ b/pyanaconda/ui/gui/spokes/storage.py
@@ -314,6 +314,7 @@ class StorageSpoke(NormalSpoke, StorageChecker):
hubQ.send_not_ready(self.__class__.__name__)
# on the off-chance dasdfmt is running, we can't proceed further
threadMgr.wait(constants.THREAD_DASDFMT)
+ threadMgr.wait(constants.THREAD_STORAGE)
hubQ.send_message(self.__class__.__name__, _("Saving storage
configuration..."))
I think the wait should go after the status is changed
because otherwise
users will see something like "No disks selected" while we are waiting
for the thread and then the "Saving storage configuration" status will
show for a hardly-noticeable moment.
Looks good to me otherwise.
--
Vratislav Podzimek
Anaconda Rider | Red Hat, Inc. | Brno - Czech Republic