Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 8:07 PM, Serge Hallyn wrote: > Quoting Andy Lutomirski (l...@amacapital.net): >> On Mon, Sep 29, 2014 at 4:36 PM, Eric W. Biederman >> wrote: >> > Andy Lutomirski writes: >> > >> >> On Mon, Sep 29, 2014 at 4:22 PM, Eric W. Biederman >> >> wrote: >> >>> Andy Lutomirski w

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Quoting Andy Lutomirski (l...@amacapital.net): > On Mon, Sep 29, 2014 at 4:36 PM, Eric W. Biederman > wrote: > > Andy Lutomirski writes: > > > >> On Mon, Sep 29, 2014 at 4:22 PM, Eric W. Biederman > >> wrote: > >>> Andy Lutomirski writes: > >>> > To me, this smells like MNT_DETACH does som

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 4:36 PM, Eric W. Biederman wrote: > Andy Lutomirski writes: > >> On Mon, Sep 29, 2014 at 4:22 PM, Eric W. Biederman >> wrote: >>> Andy Lutomirski writes: >>> To me, this smells like MNT_DETACH does something awful when there are mounts under the detached mount.

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Andy Lutomirski writes: > On Mon, Sep 29, 2014 at 4:22 PM, Eric W. Biederman > wrote: >> Andy Lutomirski writes: >> >>> To me, this smells like MNT_DETACH does something awful when there are >>> mounts under the detached mount. >>> >>> For example: >>> >>> mount --rbind / /mnt >>> umount -l /mn

Re: [lxc-devel] [PATCH] pivot_root: umount ., not /

On Mon, 2014-09-29 at 18:56 -0400, Stéphane Graber wrote: > On Mon, Sep 29, 2014 at 10:51:35PM +, Serge Hallyn wrote: > > This fixes pivot_root on 3.11 and older kernels. > > > > Signed-off-by: Serge Hallyn > Got to love one char bugfixes :) +1 > Acked-by: Stéphane Graber > > > --- > >

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 4:22 PM, Eric W. Biederman wrote: > Andy Lutomirski writes: > >> To me, this smells like MNT_DETACH does something awful when there are >> mounts under the detached mount. >> >> For example: >> >> mount --rbind / /mnt >> umount -l /mnt >> >> does *not* end well on my syste

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Andy Lutomirski writes: > To me, this smells like MNT_DETACH does something awful when there are > mounts under the detached mount. > > For example: > > mount --rbind / /mnt > umount -l /mnt > > does *not* end well on my system. I find it hard to believe that this > behavior is intentional. Hmm

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 4:13 PM, Andy Lutomirski wrote: > On Mon, Sep 29, 2014 at 4:07 PM, Eric W. Biederman > wrote: >> Andy Lutomirski writes: >> >>> On Mon, Sep 29, 2014 at 3:46 PM, Serge Hallyn >>> wrote: Quoting Andy Lutomirski (l...@amacapital.net): > On Mon, Sep 29, 2014 at 2:4

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 4:07 PM, Eric W. Biederman wrote: > Andy Lutomirski writes: > >> On Mon, Sep 29, 2014 at 3:46 PM, Serge Hallyn >> wrote: >>> Quoting Andy Lutomirski (l...@amacapital.net): On Mon, Sep 29, 2014 at 2:46 PM, Serge Hallyn wrote: I'm not sure that "/" is well

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Andy Lutomirski writes: > On Mon, Sep 29, 2014 at 3:46 PM, Serge Hallyn wrote: >> Quoting Andy Lutomirski (l...@amacapital.net): >>> On Mon, Sep 29, 2014 at 2:46 PM, Serge Hallyn >>> wrote: >>> I'm not sure that "/" is well-defined. You have oldroot mounted on >> >> Whoa. Seems you're right.

[lxc-devel] [lxc/lxc] 7981ea: pivot_root: umount ., not /

Branch: refs/heads/master Home: https://github.com/lxc/lxc Commit: 7981ea4684c878fe433d34743073c43bfadb870a https://github.com/lxc/lxc/commit/7981ea4684c878fe433d34743073c43bfadb870a Author: Serge Hallyn Date: 2014-09-29 (Mon, 29 Sep 2014) Changed paths: M src/lxc/conf.c

Re: [lxc-devel] [PATCH] pivot_root: umount ., not /

On Mon, Sep 29, 2014 at 10:51:35PM +, Serge Hallyn wrote: > This fixes pivot_root on 3.11 and older kernels. > > Signed-off-by: Serge Hallyn Got to love one char bugfixes :) Acked-by: Stéphane Graber > --- > src/lxc/conf.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff

[lxc-devel] [PATCH] pivot_root: umount ., not /

This fixes pivot_root on 3.11 and older kernels. Signed-off-by: Serge Hallyn --- src/lxc/conf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/lxc/conf.c b/src/lxc/conf.c index 31673d5..e8979c9 100644 --- a/src/lxc/conf.c +++ b/src/lxc/conf.c @@ -1062,7 +1062,7 @@ static

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 3:46 PM, Serge Hallyn wrote: > Quoting Andy Lutomirski (l...@amacapital.net): >> On Mon, Sep 29, 2014 at 2:46 PM, Serge Hallyn >> wrote: >> I'm not sure that "/" is well-defined. You have oldroot mounted on > > Whoa. Seems you're right. I would have expected it to mean

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Quoting Andy Lutomirski (l...@amacapital.net): > On Mon, Sep 29, 2014 at 2:46 PM, Serge Hallyn wrote: > I'm not sure that "/" is well-defined. You have oldroot mounted on Whoa. Seems you're right. I would have expected it to mean precisely the dentry+vfsmount which I pivot-rooted to. Which ha

[lxc-devel] [lxc/lxc] 888cf0: sysconfig/lxc: Reverse sourcing logic

Branch: refs/heads/master Home: https://github.com/lxc/lxc Commit: 888cf064dd0f10ec834ce7c95d9269653cf8dd2c https://github.com/lxc/lxc/commit/888cf064dd0f10ec834ce7c95d9269653cf8dd2c Author: Stéphane Graber Date: 2014-09-29 (Mon, 29 Sep 2014) Changed paths: M config/sysc

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 2:46 PM, Serge Hallyn wrote: > Quoting Andy Lutomirski (l...@amacapital.net): >> On Mon, Sep 29, 2014 at 1:55 PM, Serge Hallyn >> wrote: >> > Quoting Dwight Engen (dwight.en...@oracle.com): >> >> On Sat, 20 Sep 2014 03:15:44 + >> >> Serge Hallyn wrote: >> >> >> >> >

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Quoting Andy Lutomirski (l...@amacapital.net): > On Mon, Sep 29, 2014 at 1:55 PM, Serge Hallyn wrote: > > Quoting Dwight Engen (dwight.en...@oracle.com): > >> On Sat, 20 Sep 2014 03:15:44 + > >> Serge Hallyn wrote: > >> > >> > This idea came from Andy Lutomirski. Instead of using a > >> > te

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Quoting Andy Lutomirski (l...@amacapital.net): > My previous attachment works on 3.2, and I didn't even fix the umount2 call. Are you sure? It fails for me. That is, the program works, but then doing 'mount --bind /mnt /mnt' fails. ___ lxc-devel mailin

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 2:06 PM, Andy Lutomirski wrote: > On Mon, Sep 29, 2014 at 1:55 PM, Serge Hallyn wrote: >> Quoting Dwight Engen (dwight.en...@oracle.com): >>> On Sat, 20 Sep 2014 03:15:44 + >>> Serge Hallyn wrote: >>> >>> > This idea came from Andy Lutomirski. Instead of using a >>>

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

On Mon, Sep 29, 2014 at 1:55 PM, Serge Hallyn wrote: > Quoting Dwight Engen (dwight.en...@oracle.com): >> On Sat, 20 Sep 2014 03:15:44 + >> Serge Hallyn wrote: >> >> > This idea came from Andy Lutomirski. Instead of using a >> > temporary directory for the pivot_root put-old, use "." both >>

Re: [lxc-devel] [PATCH 1/1] pivot_root: switch to a new mechanism (v2)

Quoting Dwight Engen (dwight.en...@oracle.com): > On Sat, 20 Sep 2014 03:15:44 + > Serge Hallyn wrote: > > > This idea came from Andy Lutomirski. Instead of using a > > temporary directory for the pivot_root put-old, use "." both > > for new-root and old-root. Then fchdir into the old root

Re: [lxc-devel] [PATCH] add file/func/line to debug info

Quoting Stéphane Graber (stgra...@ubuntu.com): > On Mon, Sep 29, 2014 at 04:14:28PM +, Serge Hallyn wrote: > > > > Signed-off-by: Serge Hallyn > > I wonder if we should have some kind of way to prevent this information > from showing up for some cases, possibly adding a new log level called

[lxc-devel] [lxc/lxc] 15bc51: add file/func/line to debug info

Branch: refs/heads/master Home: https://github.com/lxc/lxc Commit: 15bc516ea987be2fa3c085010d2e77a32820769c https://github.com/lxc/lxc/commit/15bc516ea987be2fa3c085010d2e77a32820769c Author: Serge Hallyn Date: 2014-09-29 (Mon, 29 Sep 2014) Changed paths: M src/lxc/log.c

Re: [lxc-devel] Haskell LXC API bindings

On Sun, Sep 28, 2014 at 11:38:37AM +0400, Nickolay Kudasov wrote: > Hello, list! > > I am pleased to announce Haskell LXC API bindings. > Bindings are represented by 2 cabal packages: lxc > and bindings-lxc >

Re: [lxc-devel] [PATCH] add file/func/line to debug info

On Mon, Sep 29, 2014 at 04:14:28PM +, Serge Hallyn wrote: > > Signed-off-by: Serge Hallyn I wonder if we should have some kind of way to prevent this information from showing up for some cases, possibly adding a new log level called "HELP" which when used won't log the timestamp or anything

[lxc-devel] [lxc/lxc] 0af993: Rework init scripts

Branch: refs/heads/master Home: https://github.com/lxc/lxc Commit: 0af993195cbd57cff700c7c90ee8f95c2a1ef8aa https://github.com/lxc/lxc/commit/0af993195cbd57cff700c7c90ee8f95c2a1ef8aa Author: Michael H. Warfield Date: 2014-09-26 (Fri, 26 Sep 2014) Changed paths: M config/

Re: [lxc-devel] [PATCH v2 on top of Rework init scripts] fixups to init script rework

On Fri, Sep 26, 2014 at 03:21:51PM -0400, Dwight Engen wrote: > - move action() from common to sysvinit wrapper since its only really > applicable for sysvinit and not the other init systems > > - fix bug in action() fallback, need to shift away msg before executing action > > - make lxc-net 98

[lxc-devel] [PATCH] add file/func/line to debug info

Signed-off-by: Serge Hallyn --- src/lxc/log.c |7 +-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/src/lxc/log.c b/src/lxc/log.c index e3c64ee..0fe376c 100644 --- a/src/lxc/log.c +++ b/src/lxc/log.c @@ -72,6 +72,7 @@ static int log_append_stderr(const struct lxc_log_appe

Re: [lxc-devel] Nested namespaces

Cgroups and Namespaces are two completely different mechanism of the Linux kernel. Cgroups is for resource isolation while Namespaces are for kernel datastructure isolation. In other words, unsharing a namespace will have no impact on cgroups: all child processes are added to current cgroup

[lxc-devel] [PATCH 9/9] lxc-device: rewrite lxc-device.

As there is a function named attach_interface to pass a interface to container now, we do not need to relay on python impolementation for lxc-device any more. Signed-off-by: Dongsheng Yang --- src/lxc/Makefile.am | 4 +- src/lxc/lxc-device | 97 - src/lxc/lxc_de

[lxc-devel] [PATCH 6/9] utils: move useful helper functions from lxccontainer to utils.

Function of enter_to_ns() is useful but currently is static for lxccontainer.c. This patch split it into two parts named as switch_to_newuser() and switch_to_newnet() into utils.c. Signed-off-by: Dongsheng Yang --- src/lxc/lxccontainer.c | 54 +++---

Re: [lxc-devel] [PATCH 6/9] utils: move useful helper functions from lxccontainer to utils.

oops, forgot --in-reply-on . Please ignore this one. On 09/29/2014 03:08 PM, Dongsheng Yang wrote: Function of enter_to_ns() is useful but currently is static for lxccontainer.c. This patch split it into two parts named as switch_to_newuser() and switch_to_newnet() into utils.c. Signed-off-by:

[lxc-devel] [PATCH 7/9] container: introduce two functions named as {at/de}tach_interface().

Currently, we depends on ip command to attach interface to container. It means we only implemented it by python. This patch implement adding and removing interface by c and added them in struct container. Signed-off-by: Dongsheng Yang --- src/lxc/lxccontainer.c | 88

[lxc-devel] [PATCH 7/9] container: introduce two functions named as {at/de}tach_interface().

Currently, we depends on ip command to attach interface to container. It means we only implemented it by python. This patch implement adding and removing interface by c and added them in struct container. Signed-off-by: Dongsheng Yang --- src/lxc/lxccontainer.c | 88

[lxc-devel] [PATCH 6/9] utils: move useful helper functions from lxccontainer to utils.

Function of enter_to_ns() is useful but currently is static for lxccontainer.c. This patch split it into two parts named as switch_to_newuser() and switch_to_newnet() into utils.c. Signed-off-by: Dongsheng Yang --- src/lxc/lxccontainer.c | 54 +++---

[lxc-devel] [PATCH 5/9] network: introduce a interface named lxc_netdev_isup().

When we need to know some info about a netdev, such as is_up or not, we need to read the flag for the netdev. This patch introduce a interface function named lxc_netdev_isup() to check is a netdev up or down. And introduce a network private function named netdev_get_flag() to get flag for netdev