From 03eb7d494a6b7667a8e4076572edf5dc2ce3c675 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 17 May 2017 11:25:31 +0200 Subject: netfilter: ipt_CLUSTERIP: switch to nf_register_net_hook one of the last remaining users of the old api, hopefully followup commit can remove it soon. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/ipt_CLUSTERIP.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net/ipv4/netfilter') diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 038f293c2376..f30bee8e407b 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -743,14 +743,20 @@ static const struct file_operations clusterip_proc_fops = { static int clusterip_net_init(struct net *net) { struct clusterip_net *cn = net_generic(net, clusterip_net_id); + int ret; INIT_LIST_HEAD(&cn->configs); spin_lock_init(&cn->lock); + ret = nf_register_net_hook(net, &cip_arp_ops); + if (ret < 0) + return ret; + #ifdef CONFIG_PROC_FS cn->procdir = proc_mkdir("ipt_CLUSTERIP", net->proc_net); if (!cn->procdir) { + nf_unregister_net_hook(net, &cip_arp_ops); pr_err("Unable to proc dir entry\n"); return -ENOMEM; } @@ -765,6 +771,7 @@ static void clusterip_net_exit(struct net *net) struct clusterip_net *cn = net_generic(net, clusterip_net_id); proc_remove(cn->procdir); #endif + nf_unregister_net_hook(net, &cip_arp_ops); } static struct pernet_operations clusterip_net_ops = { @@ -786,17 +793,11 @@ static int __init clusterip_tg_init(void) if (ret < 0) goto cleanup_subsys; - ret = nf_register_hook(&cip_arp_ops); - if (ret < 0) - goto cleanup_target; - pr_info("ClusterIP Version %s loaded successfully\n", CLUSTERIP_VERSION); return 0; -cleanup_target: - xt_unregister_target(&clusterip_tg_reg); cleanup_subsys: unregister_pernet_subsys(&clusterip_net_ops); return ret; @@ -806,7 +807,6 @@ static void __exit clusterip_tg_exit(void) { pr_info("ClusterIP Version %s unloading\n", CLUSTERIP_VERSION); - nf_unregister_hook(&cip_arp_ops); xt_unregister_target(&clusterip_tg_reg); unregister_pernet_subsys(&clusterip_net_ops); -- cgit v1.2.3 From 9fd6452d67fb2acda12e5914e2ad371f067f3465 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sun, 21 May 2017 12:52:55 +0200 Subject: netfilter: conntrack: rename nf_ct_iterate_cleanup There are several places where we needlesly call nf_ct_iterate_cleanup, we should instead iterate the full table at module unload time. This is a leftover from back when the conntrack table got duplicated per net namespace. So rename nf_ct_iterate_cleanup to nf_ct_iterate_cleanup_net. A later patch will then add a non-net variant. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/nf_nat_masquerade_ipv4.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/ipv4/netfilter') diff --git a/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c b/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c index dc1dea15c1b4..f39037fca923 100644 --- a/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c +++ b/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c @@ -98,8 +98,8 @@ static int masq_device_event(struct notifier_block *this, */ NF_CT_ASSERT(dev->ifindex != 0); - nf_ct_iterate_cleanup(net, device_cmp, - (void *)(long)dev->ifindex, 0, 0); + nf_ct_iterate_cleanup_net(net, device_cmp, + (void *)(long)dev->ifindex, 0, 0); } return NOTIFY_DONE; -- cgit v1.2.3 From 202f59afd441474cc4c3752d2417cc05dd68ffe5 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 24 May 2017 21:24:37 +0800 Subject: netfilter: ipt_CLUSTERIP: do not hold dev It's a terrible thing to hold dev in iptables target. When the dev is being removed, unregister_netdevice has to wait for the dev to become free. dmesg will keep logging the err: kernel:unregister_netdevice: waiting for veth0_in to become free. \ Usage count = 1 until iptables rules with this target are removed manually. The worse thing is when deleting a netns, a virtual nic will be deleted instead of reset to init_net in default_device_ops exit/exit_batch. As it is earlier than to flush the iptables rules in iptable_filter_net_ops exit, unregister_netdevice will block to wait for the nic to become free. As unregister_netdevice is actually waiting for iptables rules flushing while iptables rules have to be flushed after unregister_netdevice. This 'dead lock' will cause unregister_netdevice to block there forever. As the netns is not available to operate at that moment, iptables rules can not even be flushed manually either. The reproducer can be: # ip netns add test # ip link add veth0_in type veth peer name veth0_out # ip link set veth0_in netns test # ip netns exec test ip link set lo up # ip netns exec test ip link set veth0_in up # ip netns exec test iptables -I INPUT -d 1.2.3.4 -i veth0_in -j \ CLUSTERIP --new --clustermac 89:d4:47:eb:9a:fa --total-nodes 3 \ --local-node 1 --hashmode sourceip-sourceport # ip netns del test This issue can be triggered by all virtual nics with ipt_CLUSTERIP. This patch is to fix it by not holding dev in ipt_CLUSTERIP, but saving the dev->ifindex instead of the dev. As Pablo Neira Ayuso's suggestion, it will refresh c->ifindex and dev's mc by registering a netdevice notifier, just as what xt_TEE does. So it removes the old codes updating dev's mc, and also no need to initialize c->ifindex with dev->ifindex. But as one config can be shared by more than one targets, and the netdev notifier is per config, not per target. It couldn't get e->ip.iniface in the notifier handler. So e->ip.iniface has to be saved into config. Note that for backwards compatibility, this patch doesn't remove the codes checking if the dev exists before creating a config. v1->v2: - As Pablo Neira Ayuso's suggestion, register a netdevice notifier to manage c->ifindex and dev's mc. Reported-by: Jianlin Shi Signed-off-by: Xin Long Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/ipt_CLUSTERIP.c | 101 +++++++++++++++++++++++++++---------- 1 file changed, 73 insertions(+), 28 deletions(-) (limited to 'net/ipv4/netfilter') diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index f30bee8e407b..7d72decb80f9 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -47,7 +47,7 @@ struct clusterip_config { __be32 clusterip; /* the IP address */ u_int8_t clustermac[ETH_ALEN]; /* the MAC address */ - struct net_device *dev; /* device */ + int ifindex; /* device ifindex */ u_int16_t num_total_nodes; /* total number of nodes */ unsigned long local_nodes; /* node number array */ @@ -57,6 +57,9 @@ struct clusterip_config { enum clusterip_hashmode hash_mode; /* which hashing mode */ u_int32_t hash_initval; /* hash initialization */ struct rcu_head rcu; + + char ifname[IFNAMSIZ]; /* device ifname */ + struct notifier_block notifier; /* refresh c->ifindex in it */ }; #ifdef CONFIG_PROC_FS @@ -98,9 +101,8 @@ clusterip_config_put(struct clusterip_config *c) * entry(rule) is removed, remove the config from lists, but don't free it * yet, since proc-files could still be holding references */ static inline void -clusterip_config_entry_put(struct clusterip_config *c) +clusterip_config_entry_put(struct net *net, struct clusterip_config *c) { - struct net *net = dev_net(c->dev); struct clusterip_net *cn = net_generic(net, clusterip_net_id); local_bh_disable(); @@ -109,8 +111,7 @@ clusterip_config_entry_put(struct clusterip_config *c) spin_unlock(&cn->lock); local_bh_enable(); - dev_mc_del(c->dev, c->clustermac); - dev_put(c->dev); + unregister_netdevice_notifier(&c->notifier); /* In case anyone still accesses the file, the open/close * functions are also incrementing the refcount on their own, @@ -170,19 +171,55 @@ clusterip_config_init_nodelist(struct clusterip_config *c, set_bit(i->local_nodes[n] - 1, &c->local_nodes); } -static struct clusterip_config * -clusterip_config_init(const struct ipt_clusterip_tgt_info *i, __be32 ip, - struct net_device *dev) +static int +clusterip_netdev_event(struct notifier_block *this, unsigned long event, + void *ptr) { - struct net *net = dev_net(dev); + struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct clusterip_config *c; + + c = container_of(this, struct clusterip_config, notifier); + switch (event) { + case NETDEV_REGISTER: + if (!strcmp(dev->name, c->ifname)) { + c->ifindex = dev->ifindex; + dev_mc_add(dev, c->clustermac); + } + break; + case NETDEV_UNREGISTER: + if (dev->ifindex == c->ifindex) { + dev_mc_del(dev, c->clustermac); + c->ifindex = -1; + } + break; + case NETDEV_CHANGENAME: + if (!strcmp(dev->name, c->ifname)) { + c->ifindex = dev->ifindex; + dev_mc_add(dev, c->clustermac); + } else if (dev->ifindex == c->ifindex) { + dev_mc_del(dev, c->clustermac); + c->ifindex = -1; + } + break; + } + + return NOTIFY_DONE; +} + +static struct clusterip_config * +clusterip_config_init(struct net *net, const struct ipt_clusterip_tgt_info *i, + __be32 ip, const char *iniface) +{ struct clusterip_net *cn = net_generic(net, clusterip_net_id); + struct clusterip_config *c; + int err; c = kzalloc(sizeof(*c), GFP_ATOMIC); if (!c) return ERR_PTR(-ENOMEM); - c->dev = dev; + strcpy(c->ifname, iniface); + c->ifindex = -1; c->clusterip = ip; memcpy(&c->clustermac, &i->clustermac, ETH_ALEN); c->num_total_nodes = i->num_total_nodes; @@ -213,17 +250,27 @@ clusterip_config_init(const struct ipt_clusterip_tgt_info *i, __be32 ip, cn->procdir, &clusterip_proc_fops, c); if (!c->pde) { - spin_lock_bh(&cn->lock); - list_del_rcu(&c->list); - spin_unlock_bh(&cn->lock); - kfree(c); - - return ERR_PTR(-ENOMEM); + err = -ENOMEM; + goto err; } } #endif - return c; + c->notifier.notifier_call = clusterip_netdev_event; + err = register_netdevice_notifier(&c->notifier); + if (!err) + return c; + +#ifdef CONFIG_PROC_FS + proc_remove(c->pde); +err: +#endif + spin_lock_bh(&cn->lock); + list_del_rcu(&c->list); + spin_unlock_bh(&cn->lock); + kfree(c); + + return ERR_PTR(err); } #ifdef CONFIG_PROC_FS @@ -425,14 +472,13 @@ static int clusterip_tg_check(const struct xt_tgchk_param *par) e->ip.iniface); return -ENOENT; } + dev_put(dev); - config = clusterip_config_init(cipinfo, - e->ip.dst.s_addr, dev); - if (IS_ERR(config)) { - dev_put(dev); + config = clusterip_config_init(par->net, cipinfo, + e->ip.dst.s_addr, + e->ip.iniface); + if (IS_ERR(config)) return PTR_ERR(config); - } - dev_mc_add(config->dev, config->clustermac); } } cipinfo->config = config; @@ -458,7 +504,7 @@ static void clusterip_tg_destroy(const struct xt_tgdtor_param *par) /* if no more entries are referencing the config, remove it * from the list and destroy the proc entry */ - clusterip_config_entry_put(cipinfo->config); + clusterip_config_entry_put(par->net, cipinfo->config); clusterip_config_put(cipinfo->config); @@ -558,10 +604,9 @@ arp_mangle(void *priv, * addresses on different interfacs. However, in the CLUSTERIP case * this wouldn't work, since we didn't subscribe the mcast group on * other interfaces */ - if (c->dev != state->out) { - pr_debug("not mangling arp reply on different " - "interface: cip'%s'-skb'%s'\n", - c->dev->name, state->out->name); + if (c->ifindex != state->out->ifindex) { + pr_debug("not mangling arp reply on different interface: cip'%d'-skb'%d'\n", + c->ifindex, state->out->ifindex); clusterip_config_put(c); return NF_ACCEPT; } -- cgit v1.2.3