[lxc-devel] Ubuntu container failure on Fedora...

2013-11-12 Thread Michael H. Warfield
Ok all... Seems like this just started happening in the last few weeks and I'm not sure when... But... My newly created Ubuntu containers are failing to start on my Fedora hosts because of a mount problem. So... What's this, what does it do, and is this really required? [root@hydra mhw]#

Re: [lxc-devel] Ubuntu container failure on Fedora...

2013-11-12 Thread Stéphane Graber
On Tue, Nov 12, 2013 at 08:31:03AM -0500, Michael H. Warfield wrote: Ok all... Seems like this just started happening in the last few weeks and I'm not sure when... But... My newly created Ubuntu containers are failing to start on my Fedora hosts because of a mount problem. So...

Re: [lxc-devel] Ubuntu container failure on Fedora...

2013-11-12 Thread Michael H. Warfield
On Tue, 2013-11-12 at 09:59 -0500, Stéphane Graber wrote: On Tue, Nov 12, 2013 at 08:31:03AM -0500, Michael H. Warfield wrote: Ok all... Seems like this just started happening in the last few weeks and I'm not sure when... But... My newly created Ubuntu containers are failing to

Re: [lxc-devel] [PATCH] replace redundant creat() with open()

2013-11-12 Thread S . Çağlar Onur
On Nov 8, 2013 5:20 PM, Serge Hallyn serge.hal...@ubuntu.com wrote: Quoting S.Çağlar Onur (cag...@10ur.org): Hey Serge, On Fri, Nov 8, 2013 at 4:06 PM, Serge Hallyn serge.hal...@ubuntu.com wrote: Quoting S.Çağlar Onur (cag...@10ur.org): creat() is equivalent to open() with flags

Re: [lxc-devel] [PATCH] Protect global variables in log module via mutex

2013-11-12 Thread S . Çağlar Onur
Hi Serge, On Mon, Nov 11, 2013 at 4:04 PM, Serge Hallyn serge.hal...@ubuntu.com wrote: Quoting S.Çağlar Onur (cag...@10ur.org): Log module contains multiple global variables so protect them introducing a new mutex and serialize accessing log functions. Also gather all locking related code

Re: [lxc-devel] [PATCH] Protect global variables in log module via mutex

2013-11-12 Thread Serge Hallyn
Quoting S.Çağlar Onur (cag...@10ur.org): Hi Serge, On Mon, Nov 11, 2013 at 4:04 PM, Serge Hallyn serge.hal...@ubuntu.com wrote: Quoting S.Çağlar Onur (cag...@10ur.org): Log module contains multiple global variables so protect them introducing a new mutex and serialize accessing log

[lxc-devel] [PATCH] lxc-ubuntu*: Mark non-essential mounts optional

2013-11-12 Thread Stéphane Graber
Signed-off-by: Stéphane Graber stgra...@ubuntu.com --- templates/lxc-ubuntu-cloud.in | 6 +++--- templates/lxc-ubuntu.in | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/templates/lxc-ubuntu-cloud.in b/templates/lxc-ubuntu-cloud.in index 41f1c70..aeadc2d 100644 ---

Re: [lxc-devel] [PATCH] fix multithreaded create()

2013-11-12 Thread S . Çağlar Onur
On Tue, Nov 12, 2013 at 2:04 PM, Dwight Engen dwight.en...@oracle.com wrote: We were calling save_config() twice within the create() flow, each from a different process. Depending on order of scheduling, sometimes the data from the first save_config() (which was just the stuff from

Re: [lxc-devel] [PATCH 1/2] Setup devtmpfs and /dev for autodev bind mounts.

2013-11-12 Thread Michael H. Warfield
Crap... Playing around with git and I totally forgot to sign off on these things... Signed-off-by: Michael H. Warfield m...@wittsend.com On Mon, 2013-11-11 at 20:08 -0500, Michael H. Warfield wrote: If autodev is not specifically set to 0 or 1, attempts to determine if systemd is being

Re: [lxc-devel] [PATCH 2/2] Support files for systemd on Fedora.

2013-11-12 Thread Michael H. Warfield
Crap... Playing around with git and I totally forgot to sign off on these things... Signed-off-by: Michael H. Warfield m...@wittsend.com On Mon, 2013-11-11 at 20:10 -0500, Michael H. Warfield wrote: Added a file lxc.service for a systemd service file. Added a file lxc-devsetup to setup

[lxc-devel] lxc and network namespace

2013-11-12 Thread Arjun Sambamoorthy
[Trying dev list as there was no response from user list] Hi, I am using lxc with a couple of veths on ubuntu. I'd like to know its network namespace name so that i can dynamically add/remove veths to the netns after the lxc is started. Could this be done in lxc? Also, executing ip netns list

Re: [lxc-devel] lxc and network namespace

2013-11-12 Thread Stéphane Graber
On Tue, Nov 12, 2013 at 12:11:19PM -0800, Arjun Sambamoorthy wrote: [Trying dev list as there was no response from user list] Hi, I am using lxc with a couple of veths on ubuntu. I'd like to know its network namespace name so that i can dynamically add/remove veths to the netns after the

Re: [lxc-devel] [PATCH] fix multithreaded create()

2013-11-12 Thread Dwight Engen
On Tue, 12 Nov 2013 14:35:58 -0500 S.Çağlar Onur cag...@10ur.org wrote: On Tue, Nov 12, 2013 at 2:04 PM, Dwight Engen dwight.en...@oracle.com wrote: We were calling save_config() twice within the create() flow, each from a different process. Depending on order of scheduling, sometimes the

Re: [lxc-devel] [PATCH] fix multithreaded create()

2013-11-12 Thread S . Çağlar Onur
Hi Dwight, On Tue, Nov 12, 2013 at 3:31 PM, Dwight Engen dwight.en...@oracle.com wrote: On Tue, 12 Nov 2013 14:35:58 -0500 S.Çağlar Onur cag...@10ur.org wrote: On Tue, Nov 12, 2013 at 2:04 PM, Dwight Engen dwight.en...@oracle.com wrote: We were calling save_config() twice within the

Re: [lxc-devel] [PATCH] fix multithreaded create()

2013-11-12 Thread Serge Hallyn
Quoting Dwight Engen (dwight.en...@oracle.com): We were calling save_config() twice within the create() flow, each from a different process. Depending on order of scheduling, sometimes the data from the first save_config() (which was just the stuff from LXC_DEFAULT_CONFIG) would overwrite the

[lxc-devel] [lxc/lxc] 6c6892: fix multithreaded create()

2013-11-12 Thread GitHub
Branch: refs/heads/master Home: https://github.com/lxc/lxc Commit: 6c6892b5c5c6281d1e74c979168d1cbbe6a50ffc https://github.com/lxc/lxc/commit/6c6892b5c5c6281d1e74c979168d1cbbe6a50ffc Author: Dwight Engen dwight.en...@oracle.com Date: 2013-11-12 (Tue, 12 Nov 2013) Changed

Re: [lxc-devel] [PATCH] lxc-ubuntu*: Mark non-essential mounts optional

2013-11-12 Thread Serge Hallyn
Quoting Stéphane Graber (stgra...@ubuntu.com): Signed-off-by: Stéphane Graber stgra...@ubuntu.com Acked-by: Serge E. Hallyn serge.hal...@ubuntu.com --- templates/lxc-ubuntu-cloud.in | 6 +++--- templates/lxc-ubuntu.in | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-)

[lxc-devel] [lxc/lxc] 41ace3: Update lxc_create.c corrected argument usage examp...

2013-11-12 Thread GitHub
Branch: refs/heads/master Home: https://github.com/lxc/lxc Commit: 41ace3de7a17d21438dfeaac771b929e81ea4815 https://github.com/lxc/lxc/commit/41ace3de7a17d21438dfeaac771b929e81ea4815 Author: pc-wurm yorda...@pc-wurm.de Date: 2013-11-12 (Tue, 12 Nov 2013) Changed paths: M

[lxc-devel] [lxc/lxc] 56390c: doc: Update Japanese lxc-top(1) for kernel memory ...

2013-11-12 Thread GitHub
Branch: refs/heads/master Home: https://github.com/lxc/lxc Commit: 56390ccb5764945d7dbf8bca9efc541ef178d688 https://github.com/lxc/lxc/commit/56390ccb5764945d7dbf8bca9efc541ef178d688 Author: KATOH Yasufumi ka...@jazz.email.ne.jp Date: 2013-11-12 (Tue, 12 Nov 2013) Changed

Re: [lxc-devel] [PATCH] add process_lock()/unlock() in save_config()

2013-11-12 Thread Serge Hallyn
Quoting Dwight Engen (dwight.en...@oracle.com): On Tue, 12 Nov 2013 14:42:54 -0600 Serge Hallyn serge.hal...@ubuntu.com wrote: Quoting Dwight Engen (dwight.en...@oracle.com): This is likely unnecessary, but is consistent with other uses of fopen()/fclose() in lxc. Signed-off-by:

Re: [lxc-devel] [PATCH] add process_lock()/unlock() in save_config()

2013-11-12 Thread Serge Hallyn
Quoting Dwight Engen (dwight.en...@oracle.com): On Tue, 12 Nov 2013 14:42:54 -0600 Serge Hallyn serge.hal...@ubuntu.com wrote: Quoting Dwight Engen (dwight.en...@oracle.com): This is likely unnecessary, but is consistent with other uses of fopen()/fclose() in lxc. Signed-off-by:

[lxc-devel] [PATCH] introduce lxcapi_add_device_node and lxcapi_remove_device_node to API (v2)

2013-11-12 Thread S . Çağlar Onur
Adding block/char devices to running container is a common operation so provide a common implementation for users to consume. changes since v2; * removed duplicated code Signed-off-by: S.Çağlar Onur cag...@10ur.org --- src/lxc/lxccontainer.c | 84

Re: [lxc-devel] [PATCH] introduce lxcapi_add_device_node and lxcapi_remove_device_node to API (v2)

2013-11-12 Thread Stéphane Graber
On Tue, Nov 12, 2013 at 10:47:06PM -0500, S.Çağlar Onur wrote: Adding block/char devices to running container is a common operation so provide a common implementation for users to consume. changes since v2; * removed duplicated code lxc-device which currently implements something like

Re: [lxc-devel] [PATCH] introduce lxcapi_add_device_node and lxcapi_remove_device_node to API (v2)

2013-11-12 Thread S . Çağlar Onur
Hi Stéphane, On Tue, Nov 12, 2013 at 10:51 PM, Stéphane Graber stgra...@ubuntu.com wrote: On Tue, Nov 12, 2013 at 10:47:06PM -0500, S.Çağlar Onur wrote: Adding block/char devices to running container is a common operation so provide a common implementation for users to consume. changes since

Re: [lxc-devel] [PATCH] fix multithreaded create()

2013-11-12 Thread S . Çağlar Onur
Hey Dwight, On Tue, Nov 12, 2013 at 3:35 PM, S.Çağlar Onur cag...@10ur.org wrote: Hmm, you are testing through the go binding correct? I ran these two without error: ./src/tests/lxc-test-concurrent -i 200 -j 8 create,destroy ./src/tests/lxc-test-concurrent -i 100 -j 8 I wonder if

[lxc-devel] [PATCH] introduce lxcapi_add_device_node and lxcapi_remove_device_node to API (v3)

2013-11-12 Thread S . Çağlar Onur
Adding block/char devices to running container is a common operation so provide a common implementation for users to consume. changes since v2; * lets the user set an alternate path inside the container as Stéphane suggested changes since v1; * removed duplicated code Signed-off-by: S.Çağlar