summaryrefslogtreecommitdiffstats
path: root/perl-install/fs
Commit message (Collapse)AuthorAgeFilesLines
* - mount ntfs-3g as ntfs during installPascal Rigaux2008-03-141-0/+2
| | | | | (fixes adding "windows" entry in bootloader)
* - really fix partition device name for some dmraid (missing "p", cf #38363)Pascal Rigaux2008-03-071-1/+1
| | | | | (this is needed to complete commit r238726 | pixel | 2008-03-06)
* - fix resizing/formatting ntfs, and create "Windows" entry in bootloaderPascal Rigaux2008-03-051-0/+1
| | | | | (broken because of ntfs-3g switch in previous release)
* - use and install ntfs-3g by default for ntfs partitions (#37823)Pascal Rigaux2008-03-031-3/+4
|
* create fs::any::prepare_minimal_root() out of ↵Olivier Blin2008-02-281-0/+16
| | | | install::install2::formatPartitions() (to be used in draklive-install
* move create_minimal_files() from install::any to fs::anyOlivier Blin2008-02-281-0/+12
|
* remove spurious spacesOlivier Blin2008-02-281-1/+1
|
* - do not assign a mount point to partitions of type "Compaq diagnostics"Pascal Rigaux2008-02-271-1/+1
|
* (mount) handle 'auto' typeThierry Vignaud2008-02-261-1/+1
|
* - adapt to cooker:Pascal Rigaux2008-02-221-1/+1
| | | | | o jfsprogs is now jfsutils
* vol_id failures ("unknown volume type") are too verbosePascal Rigaux2008-01-291-1/+1
|
* - detect raid partitions based on either type 0xfd or vol_id detectingPascal Rigaux2008-01-281-3/+4
| | | | | | | linux_raid_member (#35684) for this, introduce field {type_name}
* - do not propose ext4dev filesystem during install (#37157)Pascal Rigaux2008-01-251-1/+1
|
* - create /dev/cdrom symlink for installer (#36703)Pascal Rigaux2008-01-181-0/+1
| | | | | | do create the symlink to have it during install (otherwise fs::wild_device::from_part will give a non accessible device)
* - don't check /proc/partitions for a partition_table::lvm Pascal Rigaux2007-12-031-0/+3
| | | | | (ie PV on non partitioned drive)
* add install cmdline option "force_atime" Pascal Rigaux2007-11-211-1/+1
| | | | | (to use when installing distros <= 2007.1 where relatime is unknown)
* remove debug code from previous commitPascal Rigaux2007-11-191-1/+1
|
* (umount) fix garbaged error messages when umounting fsThierry Vignaud2007-11-131-1/+1
|
* have a more readable commentPascal Rigaux2007-11-091-1/+1
|
* add support for ext4Thierry Vignaud2007-11-064-10/+17
|
* (part_raw) simplifyThierry Vignaud2007-11-051-1/+1
|
* - bootloader-config must not need network::network from libdrakx-netPascal Rigaux2007-10-241-1/+1
|
* - partitioning wizard: ensure existing_part and resize_fat are proposed beforePascal Rigaux2007-10-191-3/+3
| | | | | wipe_drive
* - use /dev/cdrom in fstab, but do not write corresponding udev rulePascal Rigaux2007-10-041-20/+3
| | | | | | | | | | (the rules will be generated at boot time otherwise it won't catch the hdX/srX switch) (ID_PATH is something like "pci-0000:00:1f.1-ide-1:0", so it can't catch the ide/scsi device name switch that occurs at first boot on a box with cdrom on pata and harddisk on sata)
* - use /dev/cdrom in fstab, and write corresponding udev rulePascal Rigaux2007-10-031-0/+29
| | | | | (the rules were generated at boot time but here we ensure consistent naming)
* - diskdrake:Pascal Rigaux2007-10-011-3/+4
| | | | | o bugfix 10.4.162: allow "LVM" on RAID (#34359)
* - ignore /proc/partitions when "solaris" extended partition is present (#33866)Pascal Rigaux2007-09-241-1/+5
|
* cleanupPascal Rigaux2007-09-241-1/+1
|
* create partition_table_initialize() and partition_table_clear_and_initialize()Pascal Rigaux2007-09-191-1/+1
| | | | | | which are wrappers around partition_table::initialize() but which also create a singleton VG automatically (so that it's easier for the user)
* - fix getting UUID on mdmadm (eg: md0) devicesPascal Rigaux2007-09-192-6/+4
| | | | | | | the fix is to stop doing UUID_from_magic and LABEL_from_magic, and correctly set device_UUID and device_LABEL in the type_subpart returned by type_subpart_from_magic()
* we really want to remove existing partitions first (via tell_kernel)Pascal Rigaux2007-09-181-1/+1
|
* create partition_table::readonly and use itPascal Rigaux2007-09-181-5/+3
|
* rename partition_table::raw::clear() into partition_table::initialize()Pascal Rigaux2007-09-181-1/+1
|
* rename ->zero_MBR into ->clear, and ->zero_MBR_and_dirty into ->clear_and_dirtyPascal Rigaux2007-09-182-2/+2
| | | | | (old name kept for compatibility)
* - also load dm-zero for dmraidPascal Rigaux2007-09-171-1/+1
|
* o fix typo breaking reading fstab with UUID= entriesPascal Rigaux2007-09-141-1/+1
|
* - use user_xattr by default on reiserfs (#15068)Pascal Rigaux2007-09-131-0/+1
|
* - library: have "defaults" mount option instead of empty string (for rescue)Pascal Rigaux2007-09-101-1/+1
|
* - do not configure "windows" bootloader entries for "hidden" partitionsPascal Rigaux2007-09-071-3/+3
|
* mkswap is now in util-linux-ngPascal Rigaux2007-09-061-1/+1
|
* better use $::o->{uuid_by_default} to allow easy setting in auto_installsPascal Rigaux2007-09-041-1/+1
|
* - add global variable settable via "use_uuid" on kernel cmdlinePascal Rigaux2007-09-041-1/+1
| | | | | (it sets global variable $::uuid_by_default)
* - support for UUID (including UUID=xxx in fstab)Pascal Rigaux2007-08-315-1/+14
| | | | | (not by default for now)
* - do not set relatime option for mounting nfsPascal Rigaux2007-08-311-2/+2
| | | | | (since relatime is not handled by mount for nfs)
* ensure we have only one of { noatime, relatime }Pascal Rigaux2007-08-281-0/+1
|
* drop Move support (unused for quite a long time and superceeded by GlobetrotterThierry Vignaud2007-08-272-2/+2
| | | | | which itself was obsoleted by Mdv One)
* - diskdrake:Pascal Rigaux2007-08-211-1/+2
| | | | | | | o fix displaying umount error message (#32273) (would need a little cleaning of stderr)
* - diskdrake:Pascal Rigaux2007-08-141-1/+3
| | | | | o in fstab, have "xxx" instead of "defaults,xxx"
* - fix handling relatime vs noatimePascal Rigaux2007-08-141-2/+3
| | | | | | (esp. relatime is not the opposite of atime, noatime is) - fix titypo: norelatime doesn't exist
* partially revert previous (titi) commit: we do want noatime on laptops, sincePascal Rigaux2007-08-141-3/+5
| | | | | it's not a performance issue, but really a way to not wake up the hd