[Devel] [PATCH RHEL7 COMMIT] venet: destroy VE IP on venet destruction in NFS is enabled

[Konstantin Khorenko]

The commit is pushed to "branch-rh7-3.10.0-693.1.1.vz7.37.x-ovz" and will 
appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh7-3.10.0-693.1.1.vz7.37.15
-->
commit 4dee6ba7f64e055d5d7ab88a6333c303729db672
Author: Stanislav Kinsburskiy 
Date:   Tue Oct 17 19:30:40 2017 +0300

venet: destroy VE IP on venet destruction in NFS is enabled

We skip VE IP destruction in shutdown hook, if NFS is enabled in CT
(to allow NFS mounts to disappear).
Thus we have to destroy it along with venet device.

https://jira.sw.ru/browse/PSBM-75120
https://jira.sw.ru/browse/PSBM-73614

Signed-off-by: Stanislav Kinsburskiy 
---
 drivers/net/venetdev.c | 7 +--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/venetdev.c b/drivers/net/venetdev.c
index 7a546cc..0adc1fd 100644
--- a/drivers/net/venetdev.c
+++ b/drivers/net/venetdev.c
@@ -759,9 +759,12 @@ static void venet_dellink(struct net_device *dev, struct 
list_head *head)
struct ve_struct *env = dev->nd_net->owner_ve;
 
/* We check ve_netns to avoid races with veip SHUTDOWN hook, called from
-* ve_exit_ns()
+* ve_exit_ns().
+* Also, in veip SHUTDOWN hook we skip veip destruction, if container
+* has VE_FEATURE_NFS enabled. Thus here we have to destroy veip in
+* this case.
 */
-   if (env->ve_netns)
+   if (env->ve_netns || (env->features & VE_FEATURE_NFS))
veip_shutdown(env);
 
env->_venet_dev = NULL;
___
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel

Re: [Devel] [PATCH rh7] netfilter: Allow xt_owner in any user namespace

[Stanislav Kinsburskiy]

17.10.2017 08:53, Andrei Vagin пишет:
> On Mon, Oct 16, 2017 at 05:50:38PM +0200, Stanislav Kinsburskiy wrote:
>> Well, patch looks ok.
>> But shouldn't all the ve_init_user_ns() replaced by the par->net?
> 
> This patch does this.
> 

Yes, but not everywhere.
Say, there are owner_mt_ve0 and owner_mt6_ve0.
Shouldn't there functions also patched?

>>
>> 14.10.2017 01:20, Andrei Vagin пишет:
>>> From: "Eric W. Biederman" 
>>>
>>> ML: 9847371a84b0be330f4bc4aaa98904101ee8573d
>>> https://jira.sw.ru/browse/PSBM-69409?
>>>
>>> Making this work is a little tricky as it really isn't kosher to
>>> change the xt_owner_match_info in a check function.
>>>
>>> Without changing xt_owner_match_info we need to know the user
>>> namespace the uids and gids are specified in.  In the common case
>>> net->user_ns == current_user_ns().  Verify net->user_ns ==
>>> current_user_ns() in owner_check so we can later assume it in
>>> owner_mt.
>>>
>>> In owner_check also verify that all of the uids and gids specified are
>>> in net->user_ns and that the expected min/max relationship exists
>>> between the uids and gids in xt_owner_match_info.
>>>
>>> In owner_mt get the network namespace from the outgoing socket, as this
>>> must be the same network namespace as the netfilter rules, and use that
>>> network namespace to find the user namespace the uids and gids in
>>> xt_match_owner_info are encoded in.  Then convert from their encoded
>>> from into the kernel internal format for uids and gids and perform the
>>> owner match.
>>>
>>> Similar to ping_group_range, this code does not try to detect
>>> noncontiguous UID/GID ranges.
>>>
>>> Signed-off-by: "Eric W. Biederman" 
>>> Signed-off-by: Kevin Cernekee 
>>> Signed-off-by: Pablo Neira Ayuso 
>>> Signed-off-by: Andrei Vagin 
>>> ---
>>>  net/netfilter/xt_owner.c | 41 +++--
>>>  1 file changed, 35 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/net/netfilter/xt_owner.c b/net/netfilter/xt_owner.c
>>> index 31dec4a..1744f78 100644
>>> --- a/net/netfilter/xt_owner.c
>>> +++ b/net/netfilter/xt_owner.c
>>> @@ -80,11 +80,39 @@ owner_mt6_v0(const struct sk_buff *skb, struct 
>>> xt_action_param *par)
>>>  static int owner_check(const struct xt_mtchk_param *par)
>>>  {
>>> struct xt_owner_match_info *info = par->matchinfo;
>>> +   struct net *net = par->net;
>>>  
>>> -   /* For now only allow adding matches from the initial user namespace */
>>> +   /* Only allow the common case where the userns of the writer
>>> +* matches the userns of the network namespace.
>>> +*/
>>> if ((info->match & (XT_OWNER_UID|XT_OWNER_GID)) &&
>>> -   !current_user_ns_initial())
>>> +   (current_user_ns() != net->user_ns))
>>> return -EINVAL;
>>> +
>>> +   /* Ensure the uids are valid */
>>> +   if (info->match & XT_OWNER_UID) {
>>> +   kuid_t uid_min = make_kuid(net->user_ns, info->uid_min);
>>> +   kuid_t uid_max = make_kuid(net->user_ns, info->uid_max);
>>> +
>>> +   if (!uid_valid(uid_min) || !uid_valid(uid_max) ||
>>> +   (info->uid_max < info->uid_min) ||
>>> +   uid_lt(uid_max, uid_min)) {
>>> +   return -EINVAL;
>>> +   }
>>> +   }
>>> +
>>> +   /* Ensure the gids are valid */
>>> +   if (info->match & XT_OWNER_GID) {
>>> +   kgid_t gid_min = make_kgid(net->user_ns, info->gid_min);
>>> +   kgid_t gid_max = make_kgid(net->user_ns, info->gid_max);
>>> +
>>> +   if (!gid_valid(gid_min) || !gid_valid(gid_max) ||
>>> +   (info->gid_max < info->gid_min) ||
>>> +   gid_lt(gid_max, gid_min)) {
>>> +   return -EINVAL;
>>> +   }
>>> +   }
>>> +
>>> return 0;
>>>  }
>>>  
>>> @@ -93,6 +121,7 @@ owner_mt(const struct sk_buff *skb, struct 
>>> xt_action_param *par)
>>>  {
>>> const struct xt_owner_match_info *info = par->matchinfo;
>>> const struct file *filp;
>>> +   struct net *net = dev_net(par->in ? par->in : par->out);
>>>  
>>> if (skb->sk == NULL || skb->sk->sk_socket == NULL)
>>> return (info->match ^ info->invert) == 0;
>>> @@ -109,8 +138,8 @@ owner_mt(const struct sk_buff *skb, struct 
>>> xt_action_param *par)
>>>(XT_OWNER_UID | XT_OWNER_GID)) == 0;
>>>  
>>> if (info->match & XT_OWNER_UID) {
>>> -   kuid_t uid_min = make_kuid(ve_init_user_ns(), info->uid_min);
>>> -   kuid_t uid_max = make_kuid(ve_init_user_ns(), info->uid_max);
>>> +   kuid_t uid_min = make_kuid(net->user_ns, info->uid_min);
>>> +   kuid_t uid_max = make_kuid(net->user_ns, info->uid_max);
>>> if ((uid_gte(filp->f_cred->fsuid, uid_min) &&
>>>  uid_lte(filp->f_cred->fsuid, uid_max)) ^
>>> !(info->invert & XT_OWNER_UID))
>>> @@ -118,8 +147,8 @@ owner_mt(const struct sk_buff *skb, struct 

Re: [Devel] [PATCH rh7] netfilter: Allow xt_owner in any user namespace

[Andrei Vagin]

On Mon, Oct 16, 2017 at 01:33:12PM +0300, Konstantin Khorenko wrote:
> Stas, please review the patch.
> 
> Andrey, why do we need to support deeper user namespaces at all?
> Someone app tries to create a new userns inside a vz7 CT and use ipt_owner 
> inside it?

The kernel grabs userns when we send START to the "state" file of a
container ve cgroup. But vzctl does this after restoring a container,
so we don't know a ve userns, when we are retoriung iptable rules.

CRIU can't dump nested userns, so if any app will create a new userns,
criu dump will return an error.

> 
> --
> Best regards,
> 
> Konstantin Khorenko,
> Virtuozzo Linux Kernel Team
> 
> On 10/14/2017 02:20 AM, Andrei Vagin wrote:
> > From: "Eric W. Biederman" 
> > 
> > ML: 9847371a84b0be330f4bc4aaa98904101ee8573d
> > https://jira.sw.ru/browse/PSBM-69409?
> > 
> > Making this work is a little tricky as it really isn't kosher to
> > change the xt_owner_match_info in a check function.
> > 
> > Without changing xt_owner_match_info we need to know the user
> > namespace the uids and gids are specified in.  In the common case
> > net->user_ns == current_user_ns().  Verify net->user_ns ==
> > current_user_ns() in owner_check so we can later assume it in
> > owner_mt.
> > 
> > In owner_check also verify that all of the uids and gids specified are
> > in net->user_ns and that the expected min/max relationship exists
> > between the uids and gids in xt_owner_match_info.
> > 
> > In owner_mt get the network namespace from the outgoing socket, as this
> > must be the same network namespace as the netfilter rules, and use that
> > network namespace to find the user namespace the uids and gids in
> > xt_match_owner_info are encoded in.  Then convert from their encoded
> > from into the kernel internal format for uids and gids and perform the
> > owner match.
> > 
> > Similar to ping_group_range, this code does not try to detect
> > noncontiguous UID/GID ranges.
> > 
> > Signed-off-by: "Eric W. Biederman" 
> > Signed-off-by: Kevin Cernekee 
> > Signed-off-by: Pablo Neira Ayuso 
> > Signed-off-by: Andrei Vagin 
> > ---
> >  net/netfilter/xt_owner.c | 41 +++--
> >  1 file changed, 35 insertions(+), 6 deletions(-)
> > 
> > diff --git a/net/netfilter/xt_owner.c b/net/netfilter/xt_owner.c
> > index 31dec4a..1744f78 100644
> > --- a/net/netfilter/xt_owner.c
> > +++ b/net/netfilter/xt_owner.c
> > @@ -80,11 +80,39 @@ owner_mt6_v0(const struct sk_buff *skb, struct 
> > xt_action_param *par)
> >  static int owner_check(const struct xt_mtchk_param *par)
> >  {
> > struct xt_owner_match_info *info = par->matchinfo;
> > +   struct net *net = par->net;
> > 
> > -   /* For now only allow adding matches from the initial user namespace */
> > +   /* Only allow the common case where the userns of the writer
> > +* matches the userns of the network namespace.
> > +*/
> > if ((info->match & (XT_OWNER_UID|XT_OWNER_GID)) &&
> > -   !current_user_ns_initial())
> > +   (current_user_ns() != net->user_ns))
> > return -EINVAL;
> > +
> > +   /* Ensure the uids are valid */
> > +   if (info->match & XT_OWNER_UID) {
> > +   kuid_t uid_min = make_kuid(net->user_ns, info->uid_min);
> > +   kuid_t uid_max = make_kuid(net->user_ns, info->uid_max);
> > +
> > +   if (!uid_valid(uid_min) || !uid_valid(uid_max) ||
> > +   (info->uid_max < info->uid_min) ||
> > +   uid_lt(uid_max, uid_min)) {
> > +   return -EINVAL;
> > +   }
> > +   }
> > +
> > +   /* Ensure the gids are valid */
> > +   if (info->match & XT_OWNER_GID) {
> > +   kgid_t gid_min = make_kgid(net->user_ns, info->gid_min);
> > +   kgid_t gid_max = make_kgid(net->user_ns, info->gid_max);
> > +
> > +   if (!gid_valid(gid_min) || !gid_valid(gid_max) ||
> > +   (info->gid_max < info->gid_min) ||
> > +   gid_lt(gid_max, gid_min)) {
> > +   return -EINVAL;
> > +   }
> > +   }
> > +
> > return 0;
> >  }
> > 
> > @@ -93,6 +121,7 @@ owner_mt(const struct sk_buff *skb, struct 
> > xt_action_param *par)
> >  {
> > const struct xt_owner_match_info *info = par->matchinfo;
> > const struct file *filp;
> > +   struct net *net = dev_net(par->in ? par->in : par->out);
> > 
> > if (skb->sk == NULL || skb->sk->sk_socket == NULL)
> > return (info->match ^ info->invert) == 0;
> > @@ -109,8 +138,8 @@ owner_mt(const struct sk_buff *skb, struct 
> > xt_action_param *par)
> >(XT_OWNER_UID | XT_OWNER_GID)) == 0;
> > 
> > if (info->match & XT_OWNER_UID) {
> > -   kuid_t uid_min = make_kuid(ve_init_user_ns(), info->uid_min);
> > -   kuid_t uid_max = make_kuid(ve_init_user_ns(), info->uid_max);
> > +   kuid_t uid_min = make_kuid(net->user_ns, info->uid_min);
> >