Changeset 2052 in MondoRescue for branches/2.2.8/mondo/src/mondorestore/mondo-prep.c
- Timestamp:
- Oct 23, 2008, 2:55:59 PM (15 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
branches/2.2.8/mondo/src/mondorestore/mondo-prep.c
r2037 r2052 1 1 /*************************************************************************** 2 mondo-prep.c - description 3 ----------------- 4 5 begin: Fri Apr 19 16:40:35 EDT 2002 6 copyright : (C) 2002 Mondo Hugo Rabson 7 email : Hugo Rabson <hugorabson@msn.com> 8 edited by : by Stan Benoit 4/2002 9 email : troff@nakedsoul.org 10 cvsid : $Id$ 2 $Id$ 11 3 ***************************************************************************/ 12 4 … … 19 11 * * 20 12 ***************************************************************************/ 21 22 /* mondo-prep.c Hugo Rabson23 24 25 26 07/2027 - when starting RAID, don't do it twice :)28 - moved Joshua's new get_phys_size_of_drive() code29 from resize_drive... into get_phys_size_of_drive()30 31 06/2932 - make sure software RAID devices are formatted IF user says they're to be33 - drivelist is struct now, not char[][]34 35 06/2636 - drop make_relevant_partition_bootable(); do it yourself in C (mostly)37 - offer to reboot if partition table is locked up by the kernel38 39 06/2240 - be sure not to resize non-NTFS images when restoring41 - set non-NTFS images' partition types properly42 43 06/1944 - shut down all LVMs and VGs before prepping45 46 05/0747 - usage of parted2fdisk instead of fdisk alone (ia32/ia64 compatibility)48 BCO49 50 03/3151 - rewrote partitioning and formatting code to call fdisk once per disk52 53 10/21/200354 - suspend/resume Newt gui before/after calling do_my_funky_lvm_stuff()55 56 10/2057 - rewrote format_everything() - what a mess it was.58 It now does things in three phases:-59 - formats software RAID devices (/dev/md0, etc.)60 - formats and configures LVM devices61 - formats regular partitions (/dev/hda1, /dev/sdb2, etc.)62 and any LVMs recently prepped63 64 10/0765 - use strstr(format, "raid") instead of strcmp(format,"raid") to determin66 if partition is a RAID component67 68 09/2369 - better comments70 71 09/1872 - better logging of RAID activity73 74 05/0575 - added Joshua Oreman's FreeBSD patches76 77 04/3078 - added textonly mode79 80 04/2481 - added lots of assert()'s and log_OS_error()'s82 83 04/2184 - format_everything() --- don't let bar go too far85 - mkfs -c to check for bad blocks when formatting86 87 04/0488 - misc clean-up (Tom Mortell)89 90 01/15/200391 - added code for LVM and SW Raid (Brian Borgeson)92 93 12/10/200294 - line 1238: friendlier output95 96 11/2097 - when wiping a given device in preparation for partitioning + formatting98 it, don't wipe the MBR; just the partition table. That allows for99 stupid-ass Compaq users who like to play with their MBR's.100 - disable mountlist.txt-->mountlist.txt.pre-resize copying (superfluous)101 102 09/09103 - run_program_and_log_output() now takes boolean operator to specify104 whether it will log its activities in the event of _success_105 106 07/01 - 07/31107 - added hooks to libmondo108 - RAID enhancements (Philippe de Muyter)109 110 01/01 - 03/31111 - partition_device() will refuse to partition /dev/mdX now (cos it's a112 virtual drive); however, it will return 0 (i.e. not an error)113 - make_list_of_drives() will exclude /dev/md* from list114 - mkreiserfs -ff instead of -q (Andy Glass)115 - don't resize drive if drive not found (or if its size cannot be det'd)116 - when generating list of drives from mountlist, skip the 'p' at the end117 of drive name if /dev/ida/ or /dev/cciss/; just do it (Michael Eisenberg)118 - don't specify journal size when formatting ext3119 (used to have -Jsize=10 in the call to mkfs)120 - handles files >2GB in size121 - format interactively, if Interactive Mode122 - removed reference to g_tape_size123 - when executing /tmp/i-want-my-lvm, only record the error# if the command124 was _not_ an 'insmod' command125 - pass partition size to fdisk in Kilobytes now, not Megabytes126 - log fdisk's output to /tmp/mondo-restore.log if it fails127 - don't try to format partitions of type 'image'128 - don't type to set types of 'image' partitions129 - if format code is 1 or 2 chars then assume it is a hex string130 - took out all '/ /' comments131 - don't extrapolate/add partition from RAID dev to mountlist if it's already132 present in mountlist133 - less repetitive logging in the event of vacuum-packing of last part'n134 - no extrapolation at all: RAID partitions should be listed in mountlist135 already, thanks to either Mindi v0.5x or the mountlist editor itself136 - no longer say, 'and logging to...' when setting a partition's type137 - don't run mkfs on RAID partitions (/dev/hd*, /dev/sd*); just set type138 - the setting of a partition's type now takes place in a separate subroutine139 from the subroutine that actually creates the partition140 - no need to set type if 83: 83 is the default (under fdisk)141 - turned on '-Wall'; cleaned up some cruft142 - if vacuum-packing partition (i.e. size=0MB --> max) then say, "(maximum)"143 not, "(0 MB)"144 145 11/22/2001146 - preliminary code review147 - created on Nov 22nd, 2001148 */149 13 150 14 /**
Note:
See TracChangeset
for help on using the changeset viewer.