Red Hat Bugzilla – Bug 722466
creating 32 bit isos fails when there is 2 kernels
Last modified: 2013-07-04 08:57:00 EDT
Description of problem:
when building the rawhide qa acceptance iso's i hit a issue in lorax. with both kernel and kernel-PAE we get a traceback and not initrd
Traceback (most recent call last):
File "/usr/bin/pungi", line 218, in <module>
File "/usr/bin/pungi", line 122, in main
File "/usr/lib/python2.7/site-packages/pypungi/__init__.py", line 845, in doBuildinstall
File "/usr/lib/python2.7/site-packages/pylorax/__init__.py", line 447, in run
File "/usr/lib/python2.7/site-packages/pylorax/images.py", line 515, in create_initrd
self.installtree.compress(initrd, kernel, self.ctype, self.cargs)
File "/usr/lib/python2.7/site-packages/pylorax/installtree.py", line 517, in compress
self.make_squashfs_runtime(initrd, kernel, type, args)
File "/usr/lib/python2.7/site-packages/pylorax/installtree.py", line 633, in make_squashfs_runtime
logger.debug("initramfs.img size = %i", os.stat(self.initramfs).st_size)
OSError: [Errno 2] No such file or directory: '/srv/pungi/20110715/work/Fedora/i386/initramfs.img'
if i add --exclude=kernel-PAE to the repo line in the kickstart i can create the 32 bit isos just fine
Version-Release number of selected component (if applicable):
Steps to Reproduce:
Proposing as a F16 Alpha blocker. The impact of this issue is that kernel-PAE is not provided. Unclear what criteria would be affected by this issue.
Discussed at the 2011-07-15 Alpha blocker review meeting. Agreed that this is not an Alpha blocker as the impact is too small: lack of a PAE kernel would not stop anyone from installing and running the Alpha, it will only result in less than the full amount of system memory being accessible on some configurations. Re-proposing as a Final blocker as some present at the meeting believed it should be one.
What is the status of this? The i386 20110721 images included kernel-PAE, but the i386 TC1, RC1, and RC2 DVDs do not.
I remember Dennis telling me there was some bug with yum and two kernels. Maybe that's the problem?
Dennis do you have some more information? Or the bug #?
This particular bug is fixed, otherwise the composes would not even finish, so there must be some other problem now.
this particular issue is fixed, Im now getting a traceback from yum. i need to file a yum bug for it yet.