Bug 498553
Description
Antonio Hernandez
2009-05-01 02:47:23 UTC
Created attachment 342033 [details]
Attached traceback automatically from anaconda.
What version of anaconda is on the livecd you're using? katzj: bug#500991 is very similar (from rawhide and not a LiveCD) ... Does that have the information you're looking for? The traceback is similar, but a bit different so not sure if it's the same or not. Antonio -- more information on exactly what livecd you're using and what partitioning setup you tried to do would help here Created attachment 344987 [details]
Attached traceback automatically from anaconda.
Created attachment 345476 [details]
Attached traceback automatically from anaconda.
jmccann: can you describe how you are reproducing this failure? What are the steps you are following that lead up to the traceback? * Booted i686 live (converted to usb) from http://alt.fedoraproject.org/pub/alt/stage/f11-rc0.1/ on an Acer Aspire One netbook. * Clicked Install to HD. * Clicked Next * Selected US English & clicked Next * Set hostname & clicked Next * Clicked Next to set default timezone * Set root password & clicked Next * Selected "Use entire drive" from dropdown & clicked Next * Clicked "Write changes to disk" on popup dialog * Dialog went away * Clicked Next on same page that showed "use entire drive" * Crashed *** Bug 502771 has been marked as a duplicate of this bug. *** I worked around the issue with the following. --- storage/formats/fs.py.orig 2009-05-27 16:18:04.000000000 +0200 +++ storage/formats/fs.py 2009-05-27 16:17:25.000000000 +0200 @@ -349,8 +349,7 @@ rc = iutil.execWithPulseProgress(self.mkfsProg, argv, stdout="/dev/tty5", - stderr="/dev/tty5", - progress=w) + stderr="/dev/tty5") except Exception as e: raise FormatCreateError(e, self.device) finally: Its the pulse progress bar which does not appear on the small netbooks, could it be it needs the window to show correctly, on the netbooks anaconda appears in fullscreen mode. That workaround also allowed my install to continue. jmccann: Just to confirm, you used an updates.img to provide the patch posted to comment#10 to work around the bug you are reporting in comment#8? No, I vi'd the fs.py file directly (as in comment 10) in the running livecd and removed the .pyc and .pyo files then ran the installer again. I've completed many live image installs on many platforms and not seen this issue using the very clear steps described in comment#8. I finally read a comment from Traxtopel in the bug#502771 (cloned to this bug): >> When you have selected the partitioning of the harddisk, clicked next and >> waiting for the install to start. The NEXT button is still clickable, if you >> click on this by mistake or think the install is not running, it will fail. >> >> Perhaps this is only present on slower machines. >> >> Would it be possible to disable Next after paritioning starts, so this cannot >> fail? I've now reproduced the problem by feverishly clicking the "Next" button while file system creation is underway. I cannot reproduce this in normal installs, only live installs. Perhaps there is a window focus problem. so clumens and I looked into this a bit yesterday. We think the problem is this: 1) metacity is starting progress dialogs between pages below the fullscreen installer window even though they are transient to the installer window. or potentially a None parent is getting passed in so they aren't getting set transient properly. 2) anaconda isn't setting progress dialogs to modal so the next button still works when it shouldn't 3) the next button doesn't expect to get clicked while the dialog is up and behaves in all sorts of weird ways at that point The anaconda fixes are: 1) make the progress dialogs modal to the fullscreen window 2) verify that the parent getting passed in is not None 3) potentially explicitly raise the dialog to work around them showing up behind the fullscreen window if they're showing up behind because of a metacity bug and not 2) Thanks everyone! This really helps outline the failure conditions as well as the changes required to properly address the issue. I've discussed this issue with clumens and rstrode, and we've agreed to add the circumstances described into the https://fedoraproject.org/wiki/Common_F11_bugs page and address this post-F11. I've added the keyword CommonBugs. I, or another QA member will revisit this bug and document the issue in the coming days. However, if someone would like to help and draft a note ... please see https://fedoraproject.org/wiki/Common_F11_bugs#My_bug_is_not_listed This should be fixed in the next build of anaconda in Rawhide, post F11. Thanks for the bug report and testing. This is closed but running 11.5.0.59 the problem still exists. Do I need to move to anaconda-11.1.2.183-1 for this bug to be resolved? traxtopel: you got it, this won't be fixed until an anaconda build post-F11. I don't believe a version has been cut yet for this, but the fix is in the master git branch in anaconda (http://git.fedorahosted.org/git/?p=anaconda.git;a=commit;h=48ebe93bb6d76dcabc98fb4f1af6375f7464ff5f). But does this fix https://bugzilla.redhat.com/show_bug.cgi?id=502771 I am still seeing this issue, i.e. when installing from liveusb the harddisk does not get formatted. This issue has been documented at https://fedoraproject.org/wiki/Common_F11_bugs#498553 Created attachment 347197 [details]
Attached traceback automatically from anaconda.
*** Bug 504981 has been marked as a duplicate of this bug. *** Created attachment 348605 [details]
Attached traceback automatically from anaconda.
Created attachment 348926 [details]
Attached traceback automatically from anaconda.
I am a big newbie at Linux so any help understanding the processes to get this resolved would be appreciated. I am installing this in Sun's Virtual Box as a guest. The host is Windows Vista Premium. Created attachment 349076 [details]
Attached traceback automatically from anaconda.
Hi Dave, The problem is already resolved upstream, but not fixed in F11. The bug is caused by hitting the Next button when the installer is busy processing a previous next button click. To work around this bug, be very careful when you hit the Next button. Only hit the button one time after being asked for information and then wait until the installer switches pages before clicking again. This waiting period can be several minutes for some of the steps (normally there would be progress dialogs for the slow steps, but the bug prevents the dialogs from showing up). Created attachment 349999 [details]
Attached traceback automatically from anaconda.
Created attachment 355871 [details]
Attached traceback automatically from anaconda.
*** Bug 505861 has been marked as a duplicate of this bug. *** Created attachment 357488 [details]
Attached traceback automatically from anaconda.
Created attachment 358305 [details]
Attached traceback automatically from anaconda.
Comment on attachment 358305 [details]
Attached traceback automatically from anaconda.
Can't install F11 on Sun Virtual Box 2.2.4r47978
Exception when formatting /dev/sda1 when installing from fresh ISO CD.
Created attachment 361842 [details]
Attached traceback automatically from anaconda.
Created attachment 362153 [details]
Attached traceback automatically from anaconda.
Created attachment 362948 [details]
Attached traceback automatically from anaconda.
Created attachment 365121 [details]
Attached traceback automatically from anaconda.
Created attachment 365457 [details]
Attached traceback automatically from anaconda.
Created attachment 432392 [details]
Attached traceback automatically from anaconda.
|