tools/mic.git
10 years agoAdd mic zsh tab completion support
Sun Lihong [Thu, 12 Sep 2013 13:36:09 +0000 (21:36 +0800)]
Add mic zsh tab completion support

Fixes: #1187

Change-Id: I6a668990556431c5df8304a830a12943cef975a0

10 years agofix typo
Gui Chen [Tue, 24 Sep 2013 08:30:32 +0000 (04:30 -0400)]
fix typo

Change-Id: I4380c6076233079869b4563d93f5619cc54a990c
Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoprefer to use 'pbzip2'/'pgzip' than 'bzip2'/'gzip'
Gui Chen [Mon, 16 Sep 2013 06:29:13 +0000 (02:29 -0400)]
prefer to use 'pbzip2'/'pgzip' than 'bzip2'/'gzip'

Fixes: #1281

Change-Id: I72c996fde9689c7c684848dade4add8b76b00b07
Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agosetup.py: correct project url
Gui Chen [Fri, 13 Sep 2013 08:51:50 +0000 (04:51 -0400)]
setup.py: correct project url

Change-Id: I037b517cb7b6ac89bd07f98150c8bc9bc97337c9
Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobetter handling for installing config file in virtualenv
Gui Chen [Thu, 12 Sep 2013 09:42:23 +0000 (05:42 -0400)]
better handling for installing config file in virtualenv

in this handling, mic will check 'real_prefix' which is
not existed in real python env but in virtualenv to
determine the config file location

Change-Id: Iaa67ba1d78395470bc20dd7682c29ec4c933fd56
Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoupdate Makefile
Gui Chen [Thu, 12 Sep 2013 08:17:22 +0000 (04:17 -0400)]
update Makefile

update handling of TAGVER
drop bz2 compress
add new target 'release' to track taring release tag

Change-Id: Ief896cec98a5b36c96d7a481db88fa720d7c1c1c
Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agodrop VERSION file, use __version__ directly in mic module
Gui Chen [Wed, 28 Aug 2013 09:43:31 +0000 (05:43 -0400)]
drop VERSION file, use __version__ directly in mic module

Change-Id: I0f0ea1a93d8bbc924f5853c0d161c286251eb670
Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoavoid IndexError for loopdev assigning
Gui Chen [Tue, 10 Sep 2013 08:34:27 +0000 (04:34 -0400)]
avoid IndexError for loopdev assigning

when 'losetup -f' returns '' output,
out.split()[0] will raise IndexError

Fixes: #1094
Change-Id: Iad310d70defc3c8a83c221951f0d3f8e6dc5a8a6
Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agofix redundant whitespace in logging info
lihuanhuan [Wed, 17 Apr 2013 05:05:24 +0000 (13:05 +0800)]
fix redundant whitespace in logging info

Change-Id: I7e8103e802dcf7c473ec45681f075108a48597ca

10 years agoAdd mic bash tab completion support
Xiaoning Zhu [Fri, 23 Aug 2013 09:31:54 +0000 (17:31 +0800)]
Add mic bash tab completion support

Change-Id: I6374f93b456ae4789ea79e6eaede8485dd6562dd
Signed-off-by: Xiaoning Zhu <xiaoning.zhu@intel.com>
10 years agoclear output like 'close failed in file object'
lihuanhuan [Sun, 14 Apr 2013 06:01:36 +0000 (14:01 +0800)]
clear output like 'close failed in file object'

Change-Id: I0b07b0dd708fd163ba37863ec550f94d06190903
Fixes: #1192

10 years agoFix native and bootstrap mode switch automatically
lihuanhuan [Sat, 13 Apr 2013 06:22:37 +0000 (14:22 +0800)]
Fix native and bootstrap mode switch automatically

Fixes: #1242
MIC should not fallback to native mode if bootstrap package can not be found

Change-Id: Ia682572aaaf9e028ea8e15f2d8082d0b6f1a781c

10 years agoConfirm user from pwd.getpwuid to avoid compress failure
Sun Lihong [Thu, 5 Sep 2013 10:24:57 +0000 (18:24 +0800)]
Confirm user from pwd.getpwuid to avoid compress failure

In case mic compresses the image by tarfile module,it might fail
by pwd.getpwuid blocking; to avoid this,mic takes a comfirm for
user id from pwd.getpwuid after comfirming root permission.

Fixes: #1207

Change-Id: Iaa6a886936d5d6d60ba3b21608fa81d9f4c1a238

10 years agoEnsure MIC works well when "/etc/mic/mic.conf" does not exist
Sun Lihong [Tue, 3 Sep 2013 08:56:02 +0000 (16:56 +0800)]
Ensure MIC works well when "/etc/mic/mic.conf" does not exist

when "/etc/mic/mic.conf" does not exist, ensure MIC works in
native mode, and when requring bootstrap mode, MIC switches to
native mode to work,because bootstrap mode needs the existence
of "/etc/mic/mic.conf".

Fixes: #1196

Change-Id: Idf16cbb4da7748665f466eb3f2045a853821915e

10 years agoChange "Warning: Failed to load plugin" to verbose info
Sun Lihong [Tue, 3 Sep 2013 09:50:50 +0000 (17:50 +0800)]
Change "Warning: Failed to load plugin" to verbose info

 Fixes: #1191.

Change-Id: If0fede7c1fec739a29d81ff02f5b6b4816fad78a

10 years agomake customized setting of plugin_dir work in bootstrap mode
Xiaoning Zhu [Thu, 22 Aug 2013 06:29:46 +0000 (14:29 +0800)]
make customized setting of plugin_dir work in bootstrap mode

copy plugin_dir into bootstrap in bootstrap mode

10 years agoRemove several CR chars in log messages
Patrick McCarty [Mon, 26 Aug 2013 17:45:45 +0000 (10:45 -0700)]
Remove several CR chars in log messages

When msger functions are called in an environment where the TTY supports
color, and the log messages begin with a carriage return (CR), then
message headers are formatted differently than they would be without the
CR.

For msger Info messages, it looks like:

  \033[2K\033[32m\rInfo:\033[0m

and is interpreted as:

  1. Clear the entire line
  2. Change color code to 32
  3. Print "\rInfo:"
  4. Reset color code

This logic makes sense for messages that begin with a CR but do not end
with a newline, since TTYs on Linux will expect either a '\n' or a
'\r\n' combination. However, I don't see a situation where MIC needs to
log messages with the leading CR.

To avoid the above logic, remove the CR characters from existing log
messages so that newlines are always added when the messages are
printed.

Change-Id: Ib98ad1a7217033aeb1644098fea5b907d4f4729d

10 years agobump up to release 0.21 0.21
Gui Chen [Wed, 28 Aug 2013 07:15:10 +0000 (03:15 -0400)]
bump up to release 0.21

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoupdate changelog and release notes
Gui Chen [Wed, 28 Aug 2013 07:14:09 +0000 (03:14 -0400)]
update changelog and release notes

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobump up to version 0.21-0.rc4
Gui Chen [Tue, 27 Aug 2013 10:39:38 +0000 (06:39 -0400)]
bump up to version 0.21-0.rc4

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agodon't exit if parted returns non-zero
Gui Chen [Tue, 27 Aug 2013 10:18:59 +0000 (06:18 -0400)]
don't exit if parted returns non-zero

parted actually returns non-zero even if set
'boot' flag succeeded, to be compatible with
all the case, mic is better to give warning
than to exit immediately

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobump up to version 0.21-0.rc3
Gui Chen [Tue, 27 Aug 2013 02:33:30 +0000 (22:33 -0400)]
bump up to version 0.21-0.rc3

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoraise error if load mapper device failed
Gui Chen [Tue, 27 Aug 2013 02:31:19 +0000 (22:31 -0400)]
raise error if load mapper device failed

if calling 'dmsetup mknodes' doesn't take effect, mic
should raise error to avoid user

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix uuid KeyError
Gui Chen [Mon, 26 Aug 2013 07:06:02 +0000 (03:06 -0400)]
fix uuid KeyError

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix kpartx device delay
Gui Chen [Mon, 26 Aug 2013 05:52:27 +0000 (01:52 -0400)]
fix kpartx device delay

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobump up to version 0.21-0.rc2
Gui Chen [Mon, 26 Aug 2013 04:11:10 +0000 (00:11 -0400)]
bump up to version 0.21-0.rc2

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix umount issue
Gui Chen [Mon, 26 Aug 2013 04:08:34 +0000 (00:08 -0400)]
fix umount issue

if mount dest is updated during mount, self.ismounted() won't
check it out, self.mounted flag is valid all the way

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix syslinux requires
Gui Chen [Thu, 22 Aug 2013 09:00:05 +0000 (05:00 -0400)]
fix syslinux requires

syslinux is lower than 4.05 in centos, but it works

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobump up to version 0.21-0.rc1
Gui Chen [Thu, 22 Aug 2013 08:07:08 +0000 (04:07 -0400)]
bump up to version 0.21-0.rc1

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix requires
Gui Chen [Thu, 22 Aug 2013 07:32:00 +0000 (03:32 -0400)]
fix requires

syslinux 4.05 or later fixed the gpt issue
dosfstools has a misused release number

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoExit if partition is marked as active but boot flag can't be set.
Huang Hao [Wed, 14 Aug 2013 06:57:03 +0000 (14:57 +0800)]
Exit if partition is marked as active but boot flag can't be set.

If partition is marked with --active in ks file, mic should set
'boot' flag for MBR or 'legacy_boot' flag for GPT.

parted 2.3 doesn't support 'legacy_boot' flag, so this operation
may fail and when it happens the image will be unbootable.

Fixes: #1067

Change-Id: I8e0a5e901522cfed6af8e7dd892372d3649f2b77

10 years agoFix typo in comments and message
Huang Hao [Tue, 20 Aug 2013 06:56:51 +0000 (14:56 +0800)]
Fix typo in comments and message

Change-Id: I7224bb5c500702c032b8a7c9c1dfbe91e7ae73ed

10 years agorefine SimpleLockfile with closing lockf immediately
Gui Chen [Fri, 16 Aug 2013 03:04:38 +0000 (23:04 -0400)]
refine SimpleLockfile with closing lockf immediately

close the lock file immediately after create and open it,
it doesn't block the O_EXCL working because the file already
exists

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agodon't need to clean up extra mounts present
Gui Chen [Fri, 16 Aug 2013 03:22:02 +0000 (23:22 -0400)]
don't need to clean up extra mounts present

cleanup_mount would clean up the chrootdir, which
will block the locking mechanism

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agosetup_qemu_emulator: replace open/close with with_statement
Gui Chen [Thu, 15 Aug 2013 08:02:23 +0000 (04:02 -0400)]
setup_qemu_emulator: replace open/close with with_statement

to avoid unexpected IOError caused by open/close

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobind mount /lib/modules with 'ro' option
Gui Chen [Thu, 15 Aug 2013 07:47:01 +0000 (03:47 -0400)]
bind mount /lib/modules with 'ro' option

to keep the /lib/modules untouch as expected, we must
use 'ro' option when mounting it

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix chroot_lock reference
Gui Chen [Thu, 15 Aug 2013 02:06:01 +0000 (22:06 -0400)]
fix chroot_lock reference

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoinitialize installerfw.features
Gui Chen [Wed, 14 Aug 2013 08:35:13 +0000 (04:35 -0400)]
initialize installerfw.features

this varialbe is not initialized, so mic will throw
traceback: no attribute "features"

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoMerge "remove unused code and check chroot_lock before release" into devel
Gui Chen [Wed, 14 Aug 2013 09:03:05 +0000 (02:03 -0700)]
Merge "remove unused code and check chroot_lock before release" into devel

10 years agoremove unused code and check chroot_lock before release
Gui Chen [Wed, 14 Aug 2013 08:40:19 +0000 (04:40 -0400)]
remove unused code and check chroot_lock before release

the open of chroot lock file is unneccessary, remove it;
if chroot_lock is None, release is invalid, check it.

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoDisable using proxy from /etc/sysconfig/proxy
Zhang Qiang [Thu, 8 Aug 2013 13:20:47 +0000 (21:20 +0800)]
Disable using proxy from /etc/sysconfig/proxy

Always set 'proxy' query paramer to disable using proxy from
/etc/sysconfig/proxy.

Fixes: #1093

Change-Id: I6a75c9e61bb2b2142c8401036ddd21b85886fa28

10 years agotypo fix
Gui Chen [Wed, 14 Aug 2013 07:38:19 +0000 (03:38 -0400)]
typo fix

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix bootloader options omitted
Gui Chen [Wed, 14 Aug 2013 03:47:00 +0000 (23:47 -0400)]
fix bootloader options omitted

when extra option is more than 1, the 1st will be
saved, but the others will be omitted, this patch
is fixing it

Fixes: #607

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agokickstart: add an alias for installerfw
Artem Bityutskiy [Thu, 8 Aug 2013 12:01:59 +0000 (15:01 +0300)]
kickstart: add an alias for installerfw

In commit "d52080c kickstart: add the installerfw KS command" we introduced the
"installerfw" option. However, the later feed-back was that the name is very
confusing and non-intuitive, and we agreed to re-name this option to
"installerfw_plugins". With this name it is more obvious that this command
enables a list of plugins.

The name of the "extlinux" option of this command was also criticized, and we
agreed to rename it to more generic "bootloader" name.

We agreed that we do the following.

1. Add "installerfw_plugins" alias for "installerfw"
2. Add "bootloader" alias for "extlinux"
3. When users use the old "installerfw" and "extlinux" names, accept this,
   but print a warning which says that these obsolete names will be removed in
   future releases
4. Remove the old keywords later, after a couple releases.

This patch implements just that - introduces the aliases, prints warnings, makes
sure that internally we only use the new names, amends commentaries.

Base on suggestions and the initial patch from Alexander Kanevskiy.

Change-Id: I4646d437833f5252061cb963385269dc240f5792
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoraw.py: print a message when compressing the image
Artem Bityutskiy [Thu, 8 Aug 2013 12:48:01 +0000 (15:48 +0300)]
raw.py: print a message when compressing the image

Image compression may take very long time. The last message on the screen
before compression starts is "Generating map file(s)", and then a very long
delay, so that it looks like it takes so long to generate map files, but it is
not, it is actually the compression which is slow.

Let's inform users that we start compressing the image to make it obvious what
causes the delay.

Change-Id: Ie547451519aae7be5173c05bcfab1aca6c165148
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoremove myfuser to detect the instance
Gui Chen [Fri, 9 Aug 2013 08:42:53 +0000 (04:42 -0400)]
remove myfuser to detect the instance

the detection of .chroot.lock is not so correct in logic,
if the .chroot.lock is used by another process, it works;
if it is used by this process, it faults. so fuser is not
working as expected here. also removing fuser will break
a dependency of 'fuser' command

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agofix pylint
Gui Chen [Fri, 9 Aug 2013 06:45:28 +0000 (02:45 -0400)]
fix pylint

W: 50,0:elf_arch: Dangerous default value ['/bin/bash', '/sbin/init'] as argument
C: 76,4:global_mounts.totuple: Missing docstring
C:126,8:bind_mount: Invalid name "b" for type variable (should match [a-z_][a-z0-9_]{2,30}$)
W:134,4:setup_resolv: No exception type(s) specified
C:168,8:bind_unmount: Invalid name "b" for type variable (should match [a-z_][a-z0-9_]{2,30}$)
W:178,4:cleanup_resolv: No exception type(s) specified
C:175,8:cleanup_resolv: Invalid name "fd" for type variable (should match [a-z_][a-z0-9_]{2,30}$)
C:184,8:kill_processes: Invalid name "fp" for type variable (should match [a-z_][a-z0-9_]{2,30}$)
W:189,8:kill_processes: No exception type(s) specified
C:283,4:chroot.mychroot: Missing docstring

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoadd docstring for functions
Gui Chen [Fri, 9 Aug 2013 06:29:35 +0000 (02:29 -0400)]
add docstring for functions

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoadd lockfile support for mic/chroot.py module
Gui Chen [Fri, 9 Aug 2013 03:58:40 +0000 (23:58 -0400)]
add lockfile support for mic/chroot.py module

a simple lockfile implementation is added,
and it is supposed to provide an exclusive
locking in the chroot dir

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agokill processes inside instroot chroot
Gui Chen [Thu, 8 Aug 2013 05:50:58 +0000 (01:50 -0400)]
kill processes inside instroot chroot

when processes forked inside chroot is alive,
mic should kill it first before umount

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agosplit save fs code to savefs_before_chroot
Gui Chen [Wed, 7 Aug 2013 08:50:51 +0000 (04:50 -0400)]
split save fs code to savefs_before_chroot

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agorewrite chroot arch detecting
Gui Chen [Wed, 7 Aug 2013 08:00:44 +0000 (04:00 -0400)]
rewrite chroot arch detecting

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agorefactor mic/chroot.py:global_mounts
Gui Chen [Wed, 7 Aug 2013 03:18:13 +0000 (23:18 -0400)]
refactor mic/chroot.py:global_mounts

describe more clear logic in this function
refine the /lib/modules handling

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agorefactor mic/chroot.py
Gui Chen [Wed, 7 Aug 2013 02:58:56 +0000 (22:58 -0400)]
refactor mic/chroot.py

put inner function outside to be api of
the chroot module
correct logic in setup_chrootenv and
cleanup_chrootenv
remove unused dev_null code

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoSupport user/password in proxy
Zhang Qiang [Wed, 7 Aug 2013 08:27:58 +0000 (16:27 +0800)]
Support user/password in proxy

If password contains special char ':' which should be replaced with
its' quoted char '%3A'.

Fixes: #541

Change-Id: I12575208c7c337f0bad3a40bac0988e17b160819

10 years agoadjust some requires for suse and tizen
Gui Chen [Tue, 23 Jul 2013 02:56:05 +0000 (22:56 -0400)]
adjust some requires for suse and tizen

rpm-python is called as python-rpm in tizen
isomd5sum is not provided in tizen, livecd will be not supported in tizen
change qemu requires for suse and tizen to file requires

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoremove record file to fix duplicated manifest
Gui Chen [Tue, 23 Jul 2013 02:56:30 +0000 (22:56 -0400)]
remove record file to fix duplicated manifest

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoAdded psmisc runtime dependency
Ed Bartosh [Fri, 19 Jul 2013 15:42:57 +0000 (18:42 +0300)]
Added psmisc runtime dependency

Mic users fuser utility from psmisc package, but doesn't depend on
psmic. This causes crash on the system without psmisc installed:

Error <creator>: Command 'fuser' is not available.
Exception mic.utils.errors.CreatorError: CreatorError() in <bound method
LiveUSBImageCreator.__del__ of <mic.imager.liveusb.LiveUSBImageCreator
object at 0x15a9fd0>> ignored
Warning: Can't cleanup loop device /dev/loop1
Info: mic instance shutdown
Warning: No image and other files in output dir, mic failed?

Change-Id: Id6026d0cc4a805397dc6c206e93f645ff669f7ce
Signed-off-by: Ed Bartosh <eduard.bartosh@intel.com>
10 years agoupdate release notes release-0.20
Gui Chen [Mon, 8 Jul 2013 06:06:32 +0000 (02:06 -0400)]
update release notes

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobump up to 0.20 0.20
Gui Chen [Fri, 5 Jul 2013 10:51:36 +0000 (06:51 -0400)]
bump up to 0.20

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agobump up to 0.20-0.rc1
Gui Chen [Thu, 4 Jul 2013 02:15:12 +0000 (22:15 -0400)]
bump up to 0.20-0.rc1

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoliveusb: fix vfat UUID
Artem Bityutskiy [Mon, 1 Jul 2013 12:25:41 +0000 (15:25 +0300)]
liveusb: fix vfat UUID

Commit '3577b518715c29175c5f517c6f3cf7a70b0e3d39' fixed the UUID string for
raw images, but broke it for liveusb images. And this patch fixes the issue by
making the liveusb code stop adding a dash to the UUID, because now the dash is
there, just like in UUID of ext4.

Change-Id: Ibfccdf204b301808ef63589dbc3934e517a0212d
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agobump up to 0.20-0.rc0
Gui Chen [Mon, 1 Jul 2013 09:50:52 +0000 (05:50 -0400)]
bump up to 0.20-0.rc0

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoimplement '--check-pkgs' to check if packages included in image
Gui Chen [Thu, 27 Jun 2013 07:41:59 +0000 (03:41 -0400)]
implement '--check-pkgs' to check if packages included in image

check the given package list, if it will be installed, remove it
from the checking list. when check done, it proves some given
packages won't be installed to the image, mic should abort

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agointroduce '--check-pkgs' to check packages included
Gui Chen [Thu, 27 Jun 2013 07:04:27 +0000 (03:04 -0400)]
introduce '--check-pkgs' to check packages included

in some situation like pre-release image, they expect mic
not going on if there are some packages not going to include
in the image

Example: --check-pkgs=eglibc,bash
if either of eglibc and bash will not be present in the image,
mic should not go on the image creation

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoraw: support the 'fstab' installerfw attribute
Artem Bityutskiy [Thu, 27 Jun 2013 04:35:55 +0000 (07:35 +0300)]
raw: support the 'fstab' installerfw attribute

Similarly to the 'extlinux' attribute of the 'installerfw' KS command, start
supporitng the 'fstab' attribute. This means, for example, if the KS file has
this command:

installerfw "extlinux,fstab"

then MIC will not install/configure extlinux, and it will not generate
/etc/fstab. Instead, installer framework scripts will do that.

I've added the support only to raw images, since installerfw "extlinux" is
allso supported only by raw images. The liveusb support can be added later. I
do not do this because I cannot test liveusb images at this point, so I am
afraid of breaking it.

Change-Id: Ic6bb7241783aeff571956762a42665fcd8881e3f
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agobaseimager: export partition device node
Artem Bityutskiy [Thu, 27 Jun 2013 03:56:29 +0000 (06:56 +0300)]
baseimager: export partition device node

Also export the device node name for each partition. This will be helpful when
we start generating fstab entries in the installer framework scripts.

In MIC environment, the device node name is easily constructed from the disk
name.  The disk name compes from the KS file from the --ondisk option of the
'part' command. And --ondisk is a mandatory argument.

In other installer environments, the disk names may be very different.

Change-Id: Idafce950e511f4a7693e17be10e69fbb1e7d00d8
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoraw: export file-system UUIDs
Artem Bityutskiy [Thu, 27 Jun 2013 03:49:41 +0000 (06:49 +0300)]
raw: export file-system UUIDs

This will be useful when we want to create fstab entries ourselfs by using the
installer framework.

Change-Id: Ibc425517888d45a86b73e8c73abba5639eb7e093
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoVfatDiskMount: fix vfat UUID string
Artem Bityutskiy [Wed, 26 Jun 2013 14:28:13 +0000 (17:28 +0300)]
VfatDiskMount: fix vfat UUID string

The right format for the UUID string for vfat in Linux is 'XXXX-XXXX'. This is
how it can be used in fstab, for example (UUID='XXXX-XXXX'). Also, Linux udev
creates /dev/disk/by-uuid/XXXX-XXXX entries for vfat volumes.

The problem in VfatDiskMount is that it provides it in the 'XXXXXXXX' form
(without dash). However, mkfs.vfat does not like the dash, and this is why we
did not have it, I guess.

This patch fixes the situation and makes VfatDiskMount class expose UUID in a
fstab-friendly format. This makes --fstab-entry=uuid option also work for VFAT
volumes (I tested this).

Change-Id: I1d676e3707ef1777df910273381fe4437b415f41
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoRevert "VfatDiskMount: do not initialize uuid"
Artem Bityutskiy [Wed, 26 Jun 2013 14:21:54 +0000 (17:21 +0300)]
Revert "VfatDiskMount: do not initialize uuid"

This reverts commit 83410bc120b1f214401517f00da86c53e9f3d1b2.

This patch is actually wrong. I discovered that FATFS actually _does_ support
UUID, it is just shorted - it is a 32-bit integer, but has to have the
'XXXX-XXXX' format for fstab.

Let's revert this patch and fix the problem properly.

Change-Id: I1d2019f2f93094d8e222b926ac2519e8e9b9c0b5
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agobaseimager: export INSTALLERFW_MOUNT_PREFIX
Artem Bityutskiy [Wed, 26 Jun 2013 09:15:28 +0000 (12:15 +0300)]
baseimager: export INSTALLERFW_MOUNT_PREFIX

The installer framework scripts sometimes need to know the real current path
where the partitions are mounted. In MIC environment, the real path is the same
as the mountpoint, except when we are not in chroot, in which case the real
path has a prefix. Export this information in the 'INSTALLERFW_MOUNT_PREFIX'
variable.

In other environments, like pc-installer, the current path may be different,
and pc-installer will export it in 'INSTALLERFW_MOUNT_PREFIX'.

This will allow us writing configuration scripts which will work in both MIC
and pc-installer environments.

Change-Id: Ia867685acb43836c7086b062e2f34908f378f355
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agobaseimager: move environment variable definitions
Artem Bityutskiy [Wed, 26 Jun 2013 09:08:49 +0000 (12:08 +0300)]
baseimager: move environment variable definitions

It is cleaner to have a single place where we define all the environement
variables. Move a couple of historical variables to the
'_get_post_scripts_env()' function.

Change-Id: I58e5d2dbe4477effccf254c48a1f74ccad484c8c
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoRevert "baseimager: add a couple of more installer FW variables"
Artem Bityutskiy [Wed, 26 Jun 2013 08:36:27 +0000 (11:36 +0300)]
Revert "baseimager: add a couple of more installer FW variables"

A better solution is invented.

This reverts commit be0a53342bec78025413dbbd0bf4de5d9c8b454f.

10 years agobaseimager: add a couple of more installer FW variables
Artem Bityutskiy [Wed, 26 Jun 2013 05:25:38 +0000 (08:25 +0300)]
baseimager: add a couple of more installer FW variables

Add two more installer framework variables:

INSTALLERFW_ROOT_PART_NUM and INSTALLERFW_BOOT_PART_NUM

which define the boot and root parition numbers.

Change-Id: I6b717d3051faa23edb58b44265b3eae3aab2a63e
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agofix rpm header not including 'VCS' tag
Gui Chen [Tue, 25 Jun 2013 09:19:54 +0000 (05:19 -0400)]
fix rpm header not including 'VCS' tag

Signed-off-by: Gui Chen <gui.chen@intel.com>
10 years agoExtDiskMount: fix the UUID feature for extX
Artem Bityutskiy [Mon, 24 Jun 2013 10:12:42 +0000 (13:12 +0300)]
ExtDiskMount: fix the UUID feature for extX

The current implementation of ExtDiskMount is strange. Instead of generating a
random UUID and then ask mkfs.extX to use that UUID, it run mkfs.extX without
-U, let's mkfs.extX generate a random UUID, and then uses e2fsdump and parses
its output.

This is not very logical, and this also does not work with the version of
mkfs that we use at the moment.

Change the logic and simply use mkfs.extX -U.

Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoVfatDiskMount: do not initialize uuid
Artem Bityutskiy [Mon, 24 Jun 2013 08:30:31 +0000 (11:30 +0300)]
VfatDiskMount: do not initialize uuid

Do not try to be too smart and initialize UUID to a time-stamp. This UUID is
later on used form mounting in /etc/fstab (UUID=...), and of course this fails
for VFAT because VfatDiskMount provides a fake UUID.

Instead, make VfatDiskMount provide no UUID at all.

Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agobaseimager: export a number of helful environment variables
Artem Bityutskiy [Fri, 21 Jun 2013 06:02:04 +0000 (09:02 +0300)]
baseimager: export a number of helful environment variables

Make the MIC export a number of environment variables for the %post section of
the KickStart file.

Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agopartitionedfs: preserve the KS partition number
Artem Bityutskiy [Mon, 24 Jun 2013 06:09:29 +0000 (09:09 +0300)]
partitionedfs: preserve the KS partition number

The 'add_partition()' function takes partition from the KS file and processes
them. A commentary in 'add_partition()' says: "partitions have to be added in
the first-to-last order", so the order is important.

The order number is basically the partition entry number in the KS file.
Unfortunately, it is lost once the partition is added, and later on it is not
easy to match a partition from the KS file to the processed partition.

This patch teaches 'add_partition()' to preserve the KS file partition number
for later use. The new key 'ks_pnum' will be useful in the next patch, when we
export information about partitions and we want it to be in the KS file order.

Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoraw: add an option to avoid installing extlinux
Artem Bityutskiy [Thu, 20 Jun 2013 13:58:27 +0000 (16:58 +0300)]
raw: add an option to avoid installing extlinux

Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agokickstart: add the installerfw KS command
Gui Chen [Tue, 25 Jun 2013 07:44:18 +0000 (03:44 -0400)]
kickstart: add the installerfw KS command

This command will be used in the KS file and it will specify the list legacy
MIC features which have to be disabled. Instead of these feature, the general
installer framework infrastructure mechanisms will be used.

The first option will be "extlinux" which will disable the legacy extlinux
installation feature. At some point, installerfw=all will be supported,
it will disable all the legacy features (for keyboard, fstab, users, etc).

Along with the option, introduce a helper function which checks if a the
installer framework has to be used for a feature. Usage example:

if use_installerfw(ks, "extlinux"):
    # do not execute MIC's built-in extlinux installation code
else:
    # the compatibility legacy mode

original written by Artem
Artem: added the use_installerfw() function.

Signed-off-by: Gui Chen <gui.chen@intel.com>
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
10 years agoShow stdout and stderr log of post script using msger
Yeongil Jang [Thu, 10 Jan 2013 12:42:21 +0000 (21:42 +0900)]
Show stdout and stderr log of post script using msger

 - Msger can't show and recode stdout and stderr log of post script
 because subprocess excuted post script with sys.stdout and sys.stderr.
 Then mic get the log using subprocess pipe and show log using msger.

Signed-off-by: Yeongil Jang <yg0577.jang@samsung.com>
10 years agoReverted c619010d3b5efa9dc0ebec7839b73cbcd59edcb2
Ed Bartosh [Sun, 16 Jun 2013 13:39:40 +0000 (16:39 +0300)]
Reverted c619010d3b5efa9dc0ebec7839b73cbcd59edcb2

Reverted wrong implementation of getting default architecture. It was
caused inconsitent determination of arch. Different arch was determined
for different orders of repos in .ks file. As a result of wrong
determination of architecture there are image creation failures for the
configurations, where arch was determined incorrectly.

Signed-off-by: Ed Bartosh <eduard.bartosh@intel.com>
11 years agoMerge branch 'release-0.19' into devel
Gui Chen [Mon, 20 May 2013 05:55:04 +0000 (01:55 -0400)]
Merge branch 'release-0.19' into devel

11 years agobump up to release 0.19 0.19
Gui Chen [Mon, 20 May 2013 05:44:17 +0000 (01:44 -0400)]
bump up to release 0.19

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agobump up to 0.19-0.rc2
Gui Chen [Fri, 17 May 2013 05:44:34 +0000 (01:44 -0400)]
bump up to 0.19-0.rc2

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agofix empty value in max built-in
Gui Chen [Fri, 17 May 2013 06:00:07 +0000 (02:00 -0400)]
fix empty value in max built-in

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agotry to wait for mapping device mapper
Gui Chen [Fri, 17 May 2013 05:40:23 +0000 (01:40 -0400)]
try to wait for mapping device mapper

a delay for setup multipath device will throw error:
/dev/loop0X: No such file or directory

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agogpt_parser: bugfix: correctly update alternate GPT header
Artem Bityutskiy [Thu, 16 May 2013 10:44:02 +0000 (13:44 +0300)]
gpt_parser: bugfix: correctly update alternate GPT header

This patch is a bug-fix which fixes the --part-type option.

The sysmptom of the issue was the following warning from kpartx:

GPT: partition_entry_array_crc32 values don't match: 0xa24d9e34 != 0xc4e77ef0
GPT: Use GNU Parted to correct GPT errors.

The reason of the issue was that we did not update the alternate partition
array. The root-cause is that we cannot rely on the 'offs' element of the
partition entry dictionary because it points to the primary prition array
offset, while we needed the alternate. Instead, we have to calculate the offset
ourselves.

Change-Id: Iec35ebefa28ba8f7a65c414177c909747b6512ed
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
11 years agogpt_parser: bugfix: correctly update alternate GPT header
Artem Bityutskiy [Thu, 16 May 2013 10:44:02 +0000 (13:44 +0300)]
gpt_parser: bugfix: correctly update alternate GPT header

This patch is a bug-fix which fixes the --part-type option.

The sysmptom of the issue was the following warning from kpartx:

GPT: partition_entry_array_crc32 values don't match: 0xa24d9e34 != 0xc4e77ef0
GPT: Use GNU Parted to correct GPT errors.

The reason of the issue was that we did not update the alternate partition
array. The root-cause is that we cannot rely on the 'offs' element of the
partition entry dictionary because it points to the primary prition array
offset, while we needed the alternate. Instead, we have to calculate the offset
ourselves.

Change-Id: Iec35ebefa28ba8f7a65c414177c909747b6512ed
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
11 years agobump up to 0.19-0.rc1
Gui Chen [Thu, 16 May 2013 06:59:52 +0000 (02:59 -0400)]
bump up to 0.19-0.rc1

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agoupdate release date
Gui Chen [Thu, 16 May 2013 06:55:37 +0000 (02:55 -0400)]
update release date

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agofix OSError caused by symlink removing
Gui Chen [Thu, 16 May 2013 06:46:03 +0000 (02:46 -0400)]
fix OSError caused by symlink removing

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agobump up to 0.19-0.rc
Gui Chen [Thu, 16 May 2013 02:44:11 +0000 (22:44 -0400)]
bump up to 0.19-0.rc

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agofix symlink bind mount dirname
Gui Chen [Thu, 16 May 2013 01:58:41 +0000 (21:58 -0400)]
fix symlink bind mount dirname

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agofix relative symbolic link in bind mount
Gui Chen [Wed, 15 May 2013 12:11:36 +0000 (08:11 -0400)]
fix relative symbolic link in bind mount

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agotypo fix
Gui Chen [Wed, 15 May 2013 11:07:37 +0000 (07:07 -0400)]
typo fix

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agofix '/var/lock' non-existent throw traceback
Gui Chen [Wed, 15 May 2013 07:18:22 +0000 (03:18 -0400)]
fix '/var/lock' non-existent throw traceback

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agofix symlink bind mount left issue
Gui Chen [Wed, 15 May 2013 07:08:09 +0000 (03:08 -0400)]
fix symlink bind mount left issue

when bind mount point is symbolic link,
the umount will make it left when clean up

Signed-off-by: Gui Chen <gui.chen@intel.com>
11 years agoraw: use bmap creation code from bmap-tools
Artem Bityutskiy [Tue, 7 May 2013 06:42:12 +0000 (09:42 +0300)]
raw: use bmap creation code from bmap-tools

This patch copies the bmap creation code from the bmap-tools project to mic.
Hopefully, at some point we can just make mic package depend on bmap-tools and
directly invoke bmap creation functions from bmap-tools. However, mid
maintainers have issues with this at this point, so we just copy the code.

Unlike the old mic implementation which uses FIBMAP ioctl, bmap-tools use the
FIEMAP ioctl, which is better and also works on btrfs.

Change-Id: I8807ac32d3fcb31e9d410250348318fd04a63275
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>