summaryrefslogtreecommitdiffstats
path: root/storage/formats
Commit message (Collapse)AuthorAgeFilesLines
...
* Import _ped so it can be used for _ped.DiskLabelException.Peter Jones2009-08-201-0/+1
| | | | | Somebody added an exception handler for _ped.DiskLabelException without bringing _ped in to get it from. Oopsie.
* Add support for the reiserfs filesystem (#504401)David Cantrell2009-08-191-0/+37
| | | | | | | | | | | | As the bug indicates, the kernel module is still available. I'm not really interested in supporting installation to reiserfs, but I think we should be able to recognize the filesystems in the partitioning UI as well as let people shrink existing reiserfs filesystems to make room for Fedora on more sane filesystems. Detection works as is, but users have to pass 'reiserfs' on the command line to enable installation to reiserfs volumes or to enable resize support.
* Call udev_settle from DiskLabel.commit to ensure it happens.David Lehman2009-08-171-0/+3
|
* Don't try to get the size of fstypes w/ no infofsProg defined.David Lehman2009-08-171-1/+1
|
* Create a DiskLabel format class for partition tables.David Lehman2009-08-171-0/+261
|
* Honor nodmraid commandline option (#499733)Hans de Goede2009-08-102-2/+9
|
* Recognize mpath devices when we see them.Peter Jones2009-08-061-0/+88
| | | | | This identifies that a device is part of a multipath, and builds an mpath device for partitioning.
* Show locked LUKS devices as "Encrypted (LUKS)", not "LUKS".David Lehman2009-07-211-0/+11
|
* Edit user-visible dialogs for style.Peter Jones2009-07-131-3/+3
| | | | | We have lots of strings that we display to users which are unclear or contain poor usage and grammar. That's bad.
* Default to /boot on ext4Jeremy Katz2009-06-301-1/+1
|
* Allow /boot on ext4 now that we have a grub that allows itJeremy Katz2009-06-301-16/+0
|
* New iswmd parameter for kernel cmdlineJacek Danecki2009-06-222-0/+9
|
* Convert storage/ to Makefile.amDavid Cantrell2009-06-092-33/+23
|
* Make sure the right _isMigratable gets used for Ext3FS (#501585).Chris Lumens2009-05-201-0/+2
| | | | | | | It appears that the migratable property in the FS class was using FS._isMigratable when we'd overridden that method in Ext3FS. This fix prevents the upgrade migrate fs screen from showing up unless ext4migrate is on the cmdline, like intended.
* Get existing fs size for xfs.David Cantrell2009-05-141-0/+4
|
* Get existing fs size for ntfs.David Cantrell2009-05-141-0/+3
|
* Get existing fs size for jfs.David Cantrell2009-05-141-0/+3
|
* Get existing fs size for ext2, ext3, and ext4.David Cantrell2009-05-141-0/+3
|
* Compute existing filesystem size using fs info utility.David Cantrell2009-05-141-13/+82
| | | | | | | The _getExistingSize() method in class FS will using the filesystem info/dump utility to figure out the size of the filesystem. Each filesystem must define _infofs, _defaultInfoOptions, and _existingSizeFields.
* Do not allow users to migrate ext4 to ext4.David Cantrell2009-05-141-5/+5
| | | | | | | | Noticed that ext4 filesystems were presenting users with a 'migrate to' combo box with 'ext4' as the only entry. Tracked down the problem to the migratable property override in Ext3FS. Changed it around to use helper functions with property() and the UI is working more sanely now. Users cannot select ext4 as the migration target for ext4.
* Add writeKS methods to all the format objects.Chris Lumens2009-05-126-1/+31
|
* Run tune2fs on newly formatted ext[34] filesystems. (#495476)David Lehman2009-05-081-0/+22
|
* Verify with fsck after resizing filesystemsJeremy Katz2009-05-061-0/+2
| | | | | Resizing can introduce errors too, so for paranoia, we should check afterwards
* The PS3 bootloader allows booting from ext4 filesystems (#498539).Chris Lumens2009-05-051-1/+16
|
* Correct setting the SELinux context on mountpoints (#494995).Chris Lumens2009-05-011-8/+8
| | | | | | We need to pass the mountpoint and the chroot in separately so everything doesn't end up under /mnt/sysimage. Doing that results in the contexts being set to mnt_t, which causes all sorts of problems on reboot.
* Fix an error when raising FormatCreateException.Chris Lumens2009-04-281-1/+1
|
* Support vfat filesystems in the partitioning UI (#496351).Chris Lumens2009-04-241-0/+1
|
* Define the fd variable before it can ever be referenced (#496930).Chris Lumens2009-04-211-2/+7
|
* Don't try to get size for nodev and bind filesystems.David Lehman2009-04-211-0/+6
|
* Determine minimum size for filesystems once, from constructor.David Lehman2009-04-141-39/+48
| | | | | | | | | | This way we won't have to setup complex devices every time the maxSize attribute is accessed. The device's minimum size should never change without the user doing something that triggers a storage reset/rescan, so there is no reason to expect that caching the minimum size will lead to new failures. We know that the device is already setup when the constructor is called, so it is a perfect time to obtain the minimum size and be done with it.
* Default to AES-XTS cipher mode with 512 bit key for new LUKS devices.David Lehman2009-04-141-0/+6
|
* Close file descriptors when an error occurs.Joel Granados Moreno2009-04-101-0/+2
|
* Fix handling of priority option from swap fstab entries. (#494992)David Lehman2009-04-091-2/+5
|
* Some fs types cannot be passed to programs (#495061, #493075).Chris Lumens2009-04-091-6/+16
| | | | | For fake filesystems types like EFI that are really vfat underneath, we want to pass "vfat" to all the commands that expect a type.
* When a new module is loaded, update the kernel_filesystems list.Chris Lumens2009-04-091-0/+8
|
* Try not to raise exceptions from minSize calculators.David Lehman2009-04-071-5/+2
| | | | | These should either get a min size or return a reasonable default. Crashing the installer for something like this is not reasonable.
* Load filesystem modules on demand (#490795, #494108).Chris Lumens2009-04-071-0/+37
| | | | | | | Instead of loading filesystem modules all at once in loader, load them as needed from the filesystem format __init__ methods. The intention here is to remove a lot of the special code from loader and avoid kernel errors in modules that the user never even wants to have involved.
* Don't traceback from failure finding minimum fs size. (#494070)David Lehman2009-04-061-1/+3
|
* Remove temporary directory used in _getExistingSize()David Cantrell2009-04-021-0/+1
|
* put xfs back where it belongsEric Sandeen2009-04-021-1/+1
| | | | | | | | Ow, turned xfs to unsupported *and* no cmdline override, bummer. -Eric Signed-off-by: Eric Sandeen <sandeen@redhat.com>
* Fixes of errors shown by pylint that didn't get into the beta build.Martin Gracik2009-03-314-11/+6
|
* Fix the supported property of filesystems and prepboot format.David Lehman2009-03-302-2/+10
| | | | | This will make the fstype combo contain only the filesystems it should.
* Code fixes of errors shown by pylint (mgracik).Chris Lumens2009-03-252-4/+2
| | | | | | This only fixes the serious errors that may affect F11 Beta. There's still a lot of patch left that fixes some more nitpicky stuff which can come in later.
* Device is sometimes None.Joel Granados Moreno2009-03-251-1/+1
| | | | If its none there is no need to parse it.
* Register the NoDevFS class.Chris Lumens2009-03-241-0/+2
|
* Override _setDevice and _getDevice in NFS.Chris Lumens2009-03-241-0/+11
| | | | | The base class expects a device to start with "/dev/", which obviously an NFS device does not.
* PReP formats can never be active. (#491865)David Lehman2009-03-241-0/+4
|
* Add a format for ISO9660 filesystems.Chris Lumens2009-03-241-0/+17
|
* Add EFI, Apple Bootstrap, and PPC PReP Boot formats.David Lehman2009-03-232-12/+84
|
* Bind mount formats are mountable.Chris Lumens2009-03-231-0/+4
|