From: Hans Schillstrom <hans.schillstrom@ericsson•com>
To: Pablo Neira Ayuso <pablo@netfilter•org>
Cc: Jozsef Kadlecsik <kadlec@blackhole•kfki.hu>,
Jan Engelhardt <jengelh@medozas•de>,
Patrick McHardy <kaber@trash•net>,
"netfilter-devel@vger•kernel.org"
<netfilter-devel@vger•kernel.org>,
"netdev@vger•kernel.org" <netdev@vger•kernel.org>
Subject: Re: [PATCH 1/1] netfilter: Add possibility to turn off netfilters defrag per netns
Date: Mon, 9 Jan 2012 09:58:42 +0100 [thread overview]
Message-ID: <201201090958.43017.hans.schillstrom@ericsson.com> (raw)
In-Reply-To: <20120105141859.GA8210@1984>
On Thursday 05 January 2012 15:18:59 Pablo Neira Ayuso wrote:
> On Thu, Jan 05, 2012 at 10:11:28AM +0100, Jozsef Kadlecsik wrote:
> > OK, I see. Conntrack is per net namespace but it's enabled globally.
> >
> > So at the moment I think the best solution is something like your patch
> > variant (but the condition is wrong, it should be "&& !skb->nfct"):
> >
Oops, I'll fix that :-)
> > --- a/net/ipv4/netfilter/nf_defrag_ipv4.c
> > +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c
> > @@ -74,6 +74,14 @@ static unsigned int ipv4_conntrack_defrag(unsigned int
> > hooknum,
> > ...
> > + const struct net_device *dev = (hooknum == NF_INET_LOCAL_OUT ?
> > + out : in);
> > +
> > + /* No defrag and not Previously seen (loopback)? */
> > + if (dev_net(dev)->ct.sysctl_notrac_defrag && skb->nfct) {
> > + /* Attach fake conntrack entry. as in NOTRACK */
> > + skb->nfct = &nf_ct_untracked_get()->ct_general;
> > + skb->nfctinfo = IP_CT_NEW;
> > + nf_conntrack_get(skb->nfct);
> > + return NF_ACCEPT;
> > + }
> > ...
>
> I prefer the sysctl option as well, the new table is too much and it
> remains too specific for this.
>
> I wonder if we can conditionally register the sysctl only if we are
> inside one lxc container.
>
Sure no problem, but the code will not be so nice ...
> I'm telling this because this sysctl does not seem to make any sense
> to me outside of it.
I'm not so sure that we should make it asymetric,
but it's not a big deal.
Anyway here is a sample of the sysctl in a namespace.
It is the "if (!net_eq(net, &init_net)) {..." that does the magic
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index 885f5ab..2a0d530 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -454,6 +454,21 @@ static ctl_table nf_ct_sysctl_table[] = {
},
{ }
};
+#define NFCT_SYSCTL_LAST \
+ ((sizeof(nf_ct_sysctl_table) / sizeof(struct ctl_table)) - 1)
+/*
+ * Not Visible in root name space (init_net)
+ */
+static ctl_table nf_ct_sysctl_ns_table[] = {
+ {
+ .procname = "nf_conntrack_nodefrag",
+ .data = &init_net.ct.sysctl_nodefrag,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = proc_dointvec,
+ },
+ { }
+};
#define NET_NF_CONNTRACK_MAX 2089
@@ -483,9 +498,10 @@ static int nf_conntrack_standalone_init_sysctl(struct net *net)
if (!nf_ct_netfilter_header)
goto out;
}
+ table = kzalloc(sizeof(nf_ct_sysctl_table) +
+ sizeof(nf_ct_sysctl_ns_table), GFP_KERNEL);
+ memcpy(table, nf_ct_sysctl_table, sizeof(nf_ct_sysctl_table));
- table = kmemdup(nf_ct_sysctl_table, sizeof(nf_ct_sysctl_table),
- GFP_KERNEL);
if (!table)
goto out_kmemdup;
@@ -494,6 +510,12 @@ static int nf_conntrack_standalone_init_sysctl(struct net *net)
table[3].data = &net->ct.sysctl_checksum;
table[4].data = &net->ct.sysctl_log_invalid;
+ if (!net_eq(net, &init_net)) {
+ memcpy(&table[NFCT_SYSCTL_LAST], nf_ct_sysctl_ns_table,
+ sizeof(nf_ct_sysctl_ns_table));
+ table[NFCT_SYSCTL_LAST].data = &net->ct.sysctl_nodefrag;
+ }
+
net->ct.sysctl_header = register_net_sysctl_table(net,
nf_net_netfilter_sysctl_path, table);
if (!net->ct.sysctl_header)
--
1.7.2.3
--
Regards
Hans Schillstrom <hans.schillstrom@ericsson•com>
next prev parent reply other threads:[~2012-01-09 8:58 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-04 8:07 [PATCH 1/1] netfilter: Add possibility to turn off netfilters defrag per netns Hans Schillstrom
2012-01-04 8:28 ` Jozsef Kadlecsik
2012-01-04 8:49 ` Hans Schillstrom
2012-01-04 9:03 ` Jozsef Kadlecsik
2012-01-04 9:32 ` Jan Engelhardt
2012-01-04 9:47 ` Hans Schillstrom
2012-01-04 17:23 ` Pablo Neira Ayuso
2012-01-04 9:49 ` Jozsef Kadlecsik
2012-01-04 10:18 ` Hans Schillstrom
2012-01-04 11:17 ` Jan Engelhardt
2012-01-04 11:48 ` Hans Schillstrom
2012-01-04 17:40 ` Pablo Neira Ayuso
2012-01-04 18:05 ` Jozsef Kadlecsik
2012-01-04 20:56 ` Hans Schillstrom
2012-01-04 21:40 ` Jozsef Kadlecsik
2012-01-05 7:19 ` Hans Schillstrom
2012-01-05 9:11 ` Jozsef Kadlecsik
2012-01-05 14:18 ` Pablo Neira Ayuso
2012-01-09 8:58 ` Hans Schillstrom [this message]
2012-01-10 3:17 ` Pablo Neira Ayuso
2012-01-04 20:45 ` Hans Schillstrom
2012-01-04 21:15 ` Hans Schillstrom
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=201201090958.43017.hans.schillstrom@ericsson.com \
--to=hans.schillstrom@ericsson$(echo .)com \
--cc=jengelh@medozas$(echo .)de \
--cc=kaber@trash$(echo .)net \
--cc=kadlec@blackhole$(echo .)kfki.hu \
--cc=netdev@vger$(echo .)kernel.org \
--cc=netfilter-devel@vger$(echo .)kernel.org \
--cc=pablo@netfilter$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox