netfilter: ctnetlink: deliver events for conntracks changed from userspace
As for now, the creation and update of conntracks via ctnetlink do not
propagate an event to userspace. This can result in inconsistent situations
if several userspace processes modify the connection tracking table by means
of ctnetlink at the same time. Specifically, using the conntrack command
line tool and conntrackd at the same time can trigger unconsistencies.
This patch also modifies the event cache infrastructure to pass the
process PID and the ECHO flag to nfnetlink_send() to report back
to userspace if the process that triggered the change needs so.
Based on a suggestion from Patrick McHardy.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 4f6486c..ccc5ef1 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -410,7 +410,8 @@
struct nlmsghdr *nlh;
struct nfgenmsg *nfmsg;
struct nlattr *nest_parms;
- struct nf_conn *ct = (struct nf_conn *)ptr;
+ struct nf_ct_event *item = (struct nf_ct_event *)ptr;
+ struct nf_conn *ct = item->ct;
struct sk_buff *skb;
unsigned int type;
sk_buff_data_t b;
@@ -443,7 +444,7 @@
b = skb->tail;
type |= NFNL_SUBSYS_CTNETLINK << 8;
- nlh = NLMSG_PUT(skb, 0, 0, type, sizeof(struct nfgenmsg));
+ nlh = NLMSG_PUT(skb, item->pid, 0, type, sizeof(struct nfgenmsg));
nfmsg = NLMSG_DATA(nlh);
nlh->nlmsg_flags = flags;
@@ -511,7 +512,7 @@
rcu_read_unlock();
nlh->nlmsg_len = skb->tail - b;
- nfnetlink_send(skb, 0, group, 0);
+ nfnetlink_send(skb, item->pid, group, item->report);
return NOTIFY_DONE;
nla_put_failure:
@@ -722,7 +723,9 @@
err = ctnetlink_parse_tuple(cda, &tuple, CTA_TUPLE_REPLY, u3);
else {
/* Flush the whole table */
- nf_conntrack_flush(&init_net);
+ nf_conntrack_flush(&init_net,
+ NETLINK_CB(skb).pid,
+ nlmsg_report(nlh));
return 0;
}
@@ -743,6 +746,14 @@
}
}
+ nf_conntrack_event_report(IPCT_DESTROY,
+ ct,
+ NETLINK_CB(skb).pid,
+ nlmsg_report(nlh));
+
+ /* death_by_timeout would report the event again */
+ set_bit(IPS_DYING_BIT, &ct->status);
+
nf_ct_kill(ct);
nf_ct_put(ct);
@@ -1088,11 +1099,35 @@
return 0;
}
+static inline void
+ctnetlink_event_report(struct nf_conn *ct, u32 pid, int report)
+{
+ unsigned int events = 0;
+
+ if (test_bit(IPS_EXPECTED_BIT, &ct->status))
+ events |= IPCT_RELATED;
+ else
+ events |= IPCT_NEW;
+
+ nf_conntrack_event_report(IPCT_STATUS |
+ IPCT_HELPER |
+ IPCT_REFRESH |
+ IPCT_PROTOINFO |
+ IPCT_NATSEQADJ |
+ IPCT_MARK |
+ events,
+ ct,
+ pid,
+ report);
+}
+
static int
ctnetlink_create_conntrack(struct nlattr *cda[],
struct nf_conntrack_tuple *otuple,
struct nf_conntrack_tuple *rtuple,
- struct nf_conn *master_ct)
+ struct nf_conn *master_ct,
+ u32 pid,
+ int report)
{
struct nf_conn *ct;
int err = -EINVAL;
@@ -1198,9 +1233,12 @@
ct->master = master_ct;
}
+ nf_conntrack_get(&ct->ct_general);
add_timer(&ct->timeout);
nf_conntrack_hash_insert(ct);
rcu_read_unlock();
+ ctnetlink_event_report(ct, pid, report);
+ nf_ct_put(ct);
return 0;
@@ -1265,7 +1303,9 @@
err = ctnetlink_create_conntrack(cda,
&otuple,
&rtuple,
- master_ct);
+ master_ct,
+ NETLINK_CB(skb).pid,
+ nlmsg_report(nlh));
if (err < 0 && master_ct)
nf_ct_put(master_ct);
@@ -1277,6 +1317,8 @@
* so there's no need to increase the refcount */
err = -EEXIST;
if (!(nlh->nlmsg_flags & NLM_F_EXCL)) {
+ struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
+
/* we only allow nat config for new conntracks */
if (cda[CTA_NAT_SRC] || cda[CTA_NAT_DST]) {
err = -EOPNOTSUPP;
@@ -1287,8 +1329,19 @@
err = -EOPNOTSUPP;
goto out_unlock;
}
- err = ctnetlink_change_conntrack(nf_ct_tuplehash_to_ctrack(h),
- cda);
+
+ err = ctnetlink_change_conntrack(ct, cda);
+ if (err == 0) {
+ nf_conntrack_get(&ct->ct_general);
+ spin_unlock_bh(&nf_conntrack_lock);
+ ctnetlink_event_report(ct,
+ NETLINK_CB(skb).pid,
+ nlmsg_report(nlh));
+ nf_ct_put(ct);
+ } else
+ spin_unlock_bh(&nf_conntrack_lock);
+
+ return err;
}
out_unlock:
@@ -1423,7 +1476,8 @@
{
struct nlmsghdr *nlh;
struct nfgenmsg *nfmsg;
- struct nf_conntrack_expect *exp = (struct nf_conntrack_expect *)ptr;
+ struct nf_exp_event *item = (struct nf_exp_event *)ptr;
+ struct nf_conntrack_expect *exp = item->exp;
struct sk_buff *skb;
unsigned int type;
sk_buff_data_t b;
@@ -1445,7 +1499,7 @@
b = skb->tail;
type |= NFNL_SUBSYS_CTNETLINK_EXP << 8;
- nlh = NLMSG_PUT(skb, 0, 0, type, sizeof(struct nfgenmsg));
+ nlh = NLMSG_PUT(skb, item->pid, 0, type, sizeof(struct nfgenmsg));
nfmsg = NLMSG_DATA(nlh);
nlh->nlmsg_flags = flags;
@@ -1459,7 +1513,7 @@
rcu_read_unlock();
nlh->nlmsg_len = skb->tail - b;
- nfnetlink_send(skb, 0, NFNLGRP_CONNTRACK_EXP_NEW, 0);
+ nfnetlink_send(skb, item->pid, NFNLGRP_CONNTRACK_EXP_NEW, item->report);
return NOTIFY_DONE;
nla_put_failure:
@@ -1673,7 +1727,7 @@
}
static int
-ctnetlink_create_expect(struct nlattr *cda[], u_int8_t u3)
+ctnetlink_create_expect(struct nlattr *cda[], u_int8_t u3, u32 pid, int report)
{
struct nf_conntrack_tuple tuple, mask, master_tuple;
struct nf_conntrack_tuple_hash *h = NULL;
@@ -1720,7 +1774,7 @@
memcpy(&exp->mask.src.u3, &mask.src.u3, sizeof(exp->mask.src.u3));
exp->mask.src.u.all = mask.src.u.all;
- err = nf_ct_expect_related(exp);
+ err = nf_ct_expect_related_report(exp, pid, report);
nf_ct_expect_put(exp);
out:
@@ -1753,8 +1807,12 @@
if (!exp) {
spin_unlock_bh(&nf_conntrack_lock);
err = -ENOENT;
- if (nlh->nlmsg_flags & NLM_F_CREATE)
- err = ctnetlink_create_expect(cda, u3);
+ if (nlh->nlmsg_flags & NLM_F_CREATE) {
+ err = ctnetlink_create_expect(cda,
+ u3,
+ NETLINK_CB(skb).pid,
+ nlmsg_report(nlh));
+ }
return err;
}