Most of these were found with valgrind by repeatedly doing lxc_container_new
followed by lxc_container_put. Also free memory when config items are
re-parsed, as happens when lxcapi_set_config_item() is called. Refactored
path type config items to use a common underlying routine.
Signed-off-by: Dwight Engen <dwight.engen@oracle.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

Valgrind showed use of ->next field after item has been free()ed.
Introduce a lxc_list_for_each_safe() which allows traversal of a list
when the body of the loop may remove the currently iterated item.
Signed-off-by: Dwight Engen <dwight.engen@oracle.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

lxc_log_init will leak an fd when it is called by a long running
program that may call lxc_container_new multiple times. Fix by
only opening the log if it is not already open.
Signed-off-by: Dwight Engen <dwight.engen@oracle.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

Use our own len() function for network interfaces as doing
len(container.get_config_item("lxc.network")) will fail when the
list is empty.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

The new version of the pep8 command is detecting more indentation
mistakes than it used to, this fixes them.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

The kernel config option for the memory cgroup was changed in 3.6
from CONFIG_CGROUP_MEM_RES_CTLR to CONFIG_MEMCG with commit c255a458.
Signed-off-by: Dwight Engen <dwight.engen@oracle.com>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

Fix a typo in the previous lxc.conf.sgml.in change.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

Add a container config option to mount and populate /dev in a container.

We might want to add options to specify a max size for /dev other than
the default 100k, and to specify other devices to create.  And maybe
someone can think of a better name than autodev.

Changelog: Don't error out if we couldn't mknod a /dev/ttyN.
Changelog: Describe the option in lxc.conf manpage.
Signed-off-by: Serge Hallyn <serge.hallyn@ubuntu.com>

Add a new lxc-device tool which uses the new add_device() function of
the python API and lets you add a new device node to a running container.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

This introduces a new add_devices() call to the python API.

Parameters:
 - path => Mandatory, path to a character or block device on the host
 - destpath => Optional, alternative path inside the container

The function will allow the node in the container's devices cgroup and
then create the entry in the container.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>

Apparently 'local' is not POSIX. Don't use it.
Signed-off-by: Natanael Copa <ncopa@alpinelinux.org>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

The e60a8164 introduced a bug that caused
first argument passed over to the template script get lost.

This patch fixes it.
Signed-off-by: Natanael Copa <ncopa@alpinelinux.org>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

- use '[ -x /path/prog ]' instead of 'type /path/prog'
- avoid getopt --longoptions
- add \ at after && and || when those are at end of line
- make sure condition expands to empty string if variable is empty
Signed-off-by: Natanael Copa <ncopa@alpinelinux.org>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

As suggested by Serge Hallyn on lxc-devel.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

Add option to create a pidfile for lxc-start. This is helpful for
init scripts and process monitors when running as daemon.
Signed-off-by: Natanael Copa <ncopa@alpinelinux.org>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

Tested on Oracle Linux 6 and Fedora 17
Signed-off-by: Dwight Engen <dwight.engen@oracle.com>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

Signed-off-by: Frederic Crozat <fcrozat@suse.com>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

Signed-off-by: Serge Hallyn <serge.hallyn@ubuntu.com>

Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

One of the system() calls in src/tests/startone.c wasn't checked.
This was causing a build failure now that -Wall -Werror are set by default.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

Signed-off-by: Serge Hallyn <serge.hallyn@ubuntu.com>

This reverts commit 5bf2c5ce.

Signed-off-by: Serge Hallyn <serge.hallyn@ubuntu.com>

Signed-off-by: Serge Hallyn <serge.hallyn@ubuntu.com>

Signed-off-by: Serge Hallyn <serge.hallyn@ubuntu.com>

In some cases it may be useful to pass a full path to an executable
template script directly to lxc-create.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

lxc-ls is currently exiting with return code 1 when called with --help
or when called on a system without containers.

This behaviour isn't documented in the manpage and isn't terribly intuitive.
It's been the source of quite a few weird failures in scripts running with
set -e.

As a user calling --help is a voluntary action, lxc-ls should exit 0.

Also, as lxc-ls's goal is solely to list containers, showing an error and
exiting with return code 1 when there's no container seems counter-intuitive
and error-prone.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

docbook2x-man doesn't have the same name on Debian based systems as
on RedHat based systems, add some magic to configure.ac to detect and
substitute the proper name in Makefile.am
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

The package 'docbook-tools' [1] required to format Docbook 3.0 into man pages
has been obsoleted a long time ago and can no longer be downloaded from its
former homepage. Recent versions of that package -- now called 'docbook2X' --,
cannot deal with that old markup format anymore (and don't support the '-w all'
command line switch either). To remedy these issues, all SGML files have been
updated to Docbook 4.5 so that recent versions of docbook2man can process them.

[1] http://sources.redhat.com/docbook-tools/
[2] http://docbook2x.sourceforge.net/Signed-off-by: Peter Simons <simons@cryp.to>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

Signed-off-by: Frank Scholten <frank@frankscholten.nl>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

The check for conf->rootfs.mount not being equal to LXCROOTFSMOUNT
wasn't done with strcmp which was leading to undefined behaviour
and triggered gcc warnings.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

A previous patch added a clean/distclean target to config/Makefile.am.
This conflicts with automake's own target.

This change replaces those by a clean-local and distclean-local target.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

Set automake's flags to -Wall -Werror as well as the general
CFLAGS to -Wall and -Werror when building using gcc.

This should catch any regression on build warnings now that we are in
a pretty clean state.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>

Change 5fd8314f removed the README file explaining why the
rootfs directory has to exist. Doing so broke the build on
Fedora 17 since this directory will not be created by make
install and thus the spec file cannot find it.
Signed-off-by: Dwight Engen <dwight.engen@oracle.com>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

This fixes lxc-sshd still referring to '${libdir}'.
Signed-off-by: Diego Elio Pettenò <flameeyes@flameeyes.eu>
Acked-by: Stéphane Graber <stgraber@ubuntu.com>

Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>
Signed-off-by: Serge Hallyn <serge.hallyn@ubuntu.com>

This reverts commit 9a84044bc97098821cce2721ea40a1368f17a091.

The changes made the branch to fail to build on Ubuntu/Debian, so
reverting the commits and re-opening the pull request.
Signed-off-by: Stéphane Graber <stgraber@ubuntu.com>