diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2011-03-16 16:29:25 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2011-03-16 16:29:25 -0700 |
commit | 7a6362800cb7d1d618a697a650c7aaed3eb39320 (patch) | |
tree | 087f9bc6c13ef1fad4b392c5cf9325cd28fa8523 /net/netfilter/ipvs/ip_vs_sync.c | |
parent | 6445ced8670f37cfc2c5e24a9de9b413dbfc788d (diff) | |
parent | ceda86a108671294052cbf51660097b6534672f5 (diff) | |
download | op-kernel-dev-7a6362800cb7d1d618a697a650c7aaed3eb39320.zip op-kernel-dev-7a6362800cb7d1d618a697a650c7aaed3eb39320.tar.gz |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1480 commits)
bonding: enable netpoll without checking link status
xfrm: Refcount destination entry on xfrm_lookup
net: introduce rx_handler results and logic around that
bonding: get rid of IFF_SLAVE_INACTIVE netdev->priv_flag
bonding: wrap slave state work
net: get rid of multiple bond-related netdevice->priv_flags
bonding: register slave pointer for rx_handler
be2net: Bump up the version number
be2net: Copyright notice change. Update to Emulex instead of ServerEngines
e1000e: fix kconfig for crc32 dependency
netfilter ebtables: fix xt_AUDIT to work with ebtables
xen network backend driver
bonding: Improve syslog message at device creation time
bonding: Call netif_carrier_off after register_netdevice
bonding: Incorrect TX queue offset
net_sched: fix ip_tos2prio
xfrm: fix __xfrm_route_forward()
be2net: Fix UDP packet detected status in RX compl
Phonet: fix aligned-mode pipe socket buffer header reserve
netxen: support for GbE port settings
...
Fix up conflicts in drivers/staging/brcm80211/brcmsmac/wl_mac80211.c
with the staging updates.
Diffstat (limited to 'net/netfilter/ipvs/ip_vs_sync.c')
-rw-r--r-- | net/netfilter/ipvs/ip_vs_sync.c | 1238 |
1 files changed, 981 insertions, 257 deletions
diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index ab85aed..3e7961e 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -5,6 +5,18 @@ * high-performance and highly available server based on a * cluster of servers. * + * Version 1, is capable of handling both version 0 and 1 messages. + * Version 0 is the plain old format. + * Note Version 0 receivers will just drop Ver 1 messages. + * Version 1 is capable of handle IPv6, Persistence data, + * time-outs, and firewall marks. + * In ver.1 "ip_vs_sync_conn_options" will be sent in netw. order. + * Ver. 0 can be turned on by sysctl -w net.ipv4.vs.sync_version=0 + * + * Definitions Message: is a complete datagram + * Sync_conn: is a part of a Message + * Param Data is an option to a Sync_conn. + * * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> * * ip_vs_sync: sync connection info from master load balancer to backups @@ -15,6 +27,8 @@ * Alexandre Cassen : Added SyncID support for incoming sync * messages filtering. * Justin Ossevoort : Fix endian problem on sync message size. + * Hans Schillstrom : Added Version 1: i.e. IPv6, + * Persistence support, fwmark and time-out. */ #define KMSG_COMPONENT "IPVS" @@ -35,6 +49,8 @@ #include <linux/wait.h> #include <linux/kernel.h> +#include <asm/unaligned.h> /* Used for ntoh_seq and hton_seq */ + #include <net/ip.h> #include <net/sock.h> @@ -43,11 +59,13 @@ #define IP_VS_SYNC_GROUP 0xe0000051 /* multicast addr - 224.0.0.81 */ #define IP_VS_SYNC_PORT 8848 /* multicast port */ +#define SYNC_PROTO_VER 1 /* Protocol version in header */ /* * IPVS sync connection entry + * Version 0, i.e. original version. */ -struct ip_vs_sync_conn { +struct ip_vs_sync_conn_v0 { __u8 reserved; /* Protocol, addresses and port numbers */ @@ -71,41 +89,159 @@ struct ip_vs_sync_conn_options { struct ip_vs_seq out_seq; /* outgoing seq. struct */ }; +/* + Sync Connection format (sync_conn) + + 0 1 2 3 + 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Type | Protocol | Ver. | Size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Flags | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | State | cport | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | vport | dport | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | fwmark | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | timeout (in sec.) | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | ... | + | IP-Addresses (v4 or v6) | + | ... | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + Optional Parameters. + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Param. Type | Param. Length | Param. data | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | + | ... | + | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | | Param Type | Param. Length | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Param data | + | Last Param data should be padded for 32 bit alignment | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +*/ + +/* + * Type 0, IPv4 sync connection format + */ +struct ip_vs_sync_v4 { + __u8 type; + __u8 protocol; /* Which protocol (TCP/UDP) */ + __be16 ver_size; /* Version msb 4 bits */ + /* Flags and state transition */ + __be32 flags; /* status flags */ + __be16 state; /* state info */ + /* Protocol, addresses and port numbers */ + __be16 cport; + __be16 vport; + __be16 dport; + __be32 fwmark; /* Firewall mark from skb */ + __be32 timeout; /* cp timeout */ + __be32 caddr; /* client address */ + __be32 vaddr; /* virtual address */ + __be32 daddr; /* destination address */ + /* The sequence options start here */ + /* PE data padded to 32bit alignment after seq. options */ +}; +/* + * Type 2 messages IPv6 + */ +struct ip_vs_sync_v6 { + __u8 type; + __u8 protocol; /* Which protocol (TCP/UDP) */ + __be16 ver_size; /* Version msb 4 bits */ + /* Flags and state transition */ + __be32 flags; /* status flags */ + __be16 state; /* state info */ + /* Protocol, addresses and port numbers */ + __be16 cport; + __be16 vport; + __be16 dport; + __be32 fwmark; /* Firewall mark from skb */ + __be32 timeout; /* cp timeout */ + struct in6_addr caddr; /* client address */ + struct in6_addr vaddr; /* virtual address */ + struct in6_addr daddr; /* destination address */ + /* The sequence options start here */ + /* PE data padded to 32bit alignment after seq. options */ +}; + +union ip_vs_sync_conn { + struct ip_vs_sync_v4 v4; + struct ip_vs_sync_v6 v6; +}; + +/* Bits in Type field in above */ +#define STYPE_INET6 0 +#define STYPE_F_INET6 (1 << STYPE_INET6) + +#define SVER_SHIFT 12 /* Shift to get version */ +#define SVER_MASK 0x0fff /* Mask to strip version */ + +#define IPVS_OPT_SEQ_DATA 1 +#define IPVS_OPT_PE_DATA 2 +#define IPVS_OPT_PE_NAME 3 +#define IPVS_OPT_PARAM 7 + +#define IPVS_OPT_F_SEQ_DATA (1 << (IPVS_OPT_SEQ_DATA-1)) +#define IPVS_OPT_F_PE_DATA (1 << (IPVS_OPT_PE_DATA-1)) +#define IPVS_OPT_F_PE_NAME (1 << (IPVS_OPT_PE_NAME-1)) +#define IPVS_OPT_F_PARAM (1 << (IPVS_OPT_PARAM-1)) + struct ip_vs_sync_thread_data { + struct net *net; struct socket *sock; char *buf; }; -#define SIMPLE_CONN_SIZE (sizeof(struct ip_vs_sync_conn)) +/* Version 0 definition of packet sizes */ +#define SIMPLE_CONN_SIZE (sizeof(struct ip_vs_sync_conn_v0)) #define FULL_CONN_SIZE \ -(sizeof(struct ip_vs_sync_conn) + sizeof(struct ip_vs_sync_conn_options)) +(sizeof(struct ip_vs_sync_conn_v0) + sizeof(struct ip_vs_sync_conn_options)) /* - The master mulitcasts messages to the backup load balancers in the - following format. + The master mulitcasts messages (Datagrams) to the backup load balancers + in the following format. + + Version 1: + Note, first byte should be Zero, so ver 0 receivers will drop the packet. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - | Count Conns | SyncID | Size | + | 0 | SyncID | Size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Count Conns | Version | Reserved, set to Zero | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | IPVS Sync Connection (1) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | . | - | . | + ~ . ~ | . | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | IPVS Sync Connection (n) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + + Version 0 Header + 0 1 2 3 + 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Count Conns | SyncID | Size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | IPVS Sync Connection (1) | */ #define SYNC_MESG_HEADER_LEN 4 #define MAX_CONNS_PER_SYNCBUFF 255 /* nr_conns in ip_vs_sync_mesg is 8 bit */ -struct ip_vs_sync_mesg { +/* Version 0 header */ +struct ip_vs_sync_mesg_v0 { __u8 nr_conns; __u8 syncid; __u16 size; @@ -113,9 +249,16 @@ struct ip_vs_sync_mesg { /* ip_vs_sync_conn entries start here */ }; -/* the maximum length of sync (sending/receiving) message */ -static int sync_send_mesg_maxlen; -static int sync_recv_mesg_maxlen; +/* Version 1 header */ +struct ip_vs_sync_mesg { + __u8 reserved; /* must be zero */ + __u8 syncid; + __u16 size; + __u8 nr_conns; + __s8 version; /* SYNC_PROTO_VER */ + __u16 spare; + /* ip_vs_sync_conn entries start here */ +}; struct ip_vs_sync_buff { struct list_head list; @@ -127,28 +270,6 @@ struct ip_vs_sync_buff { unsigned char *end; }; - -/* the sync_buff list head and the lock */ -static LIST_HEAD(ip_vs_sync_queue); -static DEFINE_SPINLOCK(ip_vs_sync_lock); - -/* current sync_buff for accepting new conn entries */ -static struct ip_vs_sync_buff *curr_sb = NULL; -static DEFINE_SPINLOCK(curr_sb_lock); - -/* ipvs sync daemon state */ -volatile int ip_vs_sync_state = IP_VS_STATE_NONE; -volatile int ip_vs_master_syncid = 0; -volatile int ip_vs_backup_syncid = 0; - -/* multicast interface name */ -char ip_vs_master_mcast_ifn[IP_VS_IFNAME_MAXLEN]; -char ip_vs_backup_mcast_ifn[IP_VS_IFNAME_MAXLEN]; - -/* sync daemon tasks */ -static struct task_struct *sync_master_thread; -static struct task_struct *sync_backup_thread; - /* multicast addr */ static struct sockaddr_in mcast_addr = { .sin_family = AF_INET, @@ -156,41 +277,71 @@ static struct sockaddr_in mcast_addr = { .sin_addr.s_addr = cpu_to_be32(IP_VS_SYNC_GROUP), }; +/* + * Copy of struct ip_vs_seq + * From unaligned network order to aligned host order + */ +static void ntoh_seq(struct ip_vs_seq *no, struct ip_vs_seq *ho) +{ + ho->init_seq = get_unaligned_be32(&no->init_seq); + ho->delta = get_unaligned_be32(&no->delta); + ho->previous_delta = get_unaligned_be32(&no->previous_delta); +} + +/* + * Copy of struct ip_vs_seq + * From Aligned host order to unaligned network order + */ +static void hton_seq(struct ip_vs_seq *ho, struct ip_vs_seq *no) +{ + put_unaligned_be32(ho->init_seq, &no->init_seq); + put_unaligned_be32(ho->delta, &no->delta); + put_unaligned_be32(ho->previous_delta, &no->previous_delta); +} -static inline struct ip_vs_sync_buff *sb_dequeue(void) +static inline struct ip_vs_sync_buff *sb_dequeue(struct netns_ipvs *ipvs) { struct ip_vs_sync_buff *sb; - spin_lock_bh(&ip_vs_sync_lock); - if (list_empty(&ip_vs_sync_queue)) { + spin_lock_bh(&ipvs->sync_lock); + if (list_empty(&ipvs->sync_queue)) { sb = NULL; } else { - sb = list_entry(ip_vs_sync_queue.next, + sb = list_entry(ipvs->sync_queue.next, struct ip_vs_sync_buff, list); list_del(&sb->list); } - spin_unlock_bh(&ip_vs_sync_lock); + spin_unlock_bh(&ipvs->sync_lock); return sb; } -static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create(void) +/* + * Create a new sync buffer for Version 1 proto. + */ +static inline struct ip_vs_sync_buff * +ip_vs_sync_buff_create(struct netns_ipvs *ipvs) { struct ip_vs_sync_buff *sb; if (!(sb=kmalloc(sizeof(struct ip_vs_sync_buff), GFP_ATOMIC))) return NULL; - if (!(sb->mesg=kmalloc(sync_send_mesg_maxlen, GFP_ATOMIC))) { + sb->mesg = kmalloc(ipvs->send_mesg_maxlen, GFP_ATOMIC); + if (!sb->mesg) { kfree(sb); return NULL; } + sb->mesg->reserved = 0; /* old nr_conns i.e. must be zeo now */ + sb->mesg->version = SYNC_PROTO_VER; + sb->mesg->syncid = ipvs->master_syncid; + sb->mesg->size = sizeof(struct ip_vs_sync_mesg); sb->mesg->nr_conns = 0; - sb->mesg->syncid = ip_vs_master_syncid; - sb->mesg->size = 4; - sb->head = (unsigned char *)sb->mesg + 4; - sb->end = (unsigned char *)sb->mesg + sync_send_mesg_maxlen; + sb->mesg->spare = 0; + sb->head = (unsigned char *)sb->mesg + sizeof(struct ip_vs_sync_mesg); + sb->end = (unsigned char *)sb->mesg + ipvs->send_mesg_maxlen; + sb->firstuse = jiffies; return sb; } @@ -201,14 +352,16 @@ static inline void ip_vs_sync_buff_release(struct ip_vs_sync_buff *sb) kfree(sb); } -static inline void sb_queue_tail(struct ip_vs_sync_buff *sb) +static inline void sb_queue_tail(struct netns_ipvs *ipvs) { - spin_lock(&ip_vs_sync_lock); - if (ip_vs_sync_state & IP_VS_STATE_MASTER) - list_add_tail(&sb->list, &ip_vs_sync_queue); + struct ip_vs_sync_buff *sb = ipvs->sync_buff; + + spin_lock(&ipvs->sync_lock); + if (ipvs->sync_state & IP_VS_STATE_MASTER) + list_add_tail(&sb->list, &ipvs->sync_queue); else ip_vs_sync_buff_release(sb); - spin_unlock(&ip_vs_sync_lock); + spin_unlock(&ipvs->sync_lock); } /* @@ -216,36 +369,101 @@ static inline void sb_queue_tail(struct ip_vs_sync_buff *sb) * than the specified time or the specified time is zero. */ static inline struct ip_vs_sync_buff * -get_curr_sync_buff(unsigned long time) +get_curr_sync_buff(struct netns_ipvs *ipvs, unsigned long time) { struct ip_vs_sync_buff *sb; - spin_lock_bh(&curr_sb_lock); - if (curr_sb && (time == 0 || - time_before(jiffies - curr_sb->firstuse, time))) { - sb = curr_sb; - curr_sb = NULL; + spin_lock_bh(&ipvs->sync_buff_lock); + if (ipvs->sync_buff && + time_after_eq(jiffies - ipvs->sync_buff->firstuse, time)) { + sb = ipvs->sync_buff; + ipvs->sync_buff = NULL; } else sb = NULL; - spin_unlock_bh(&curr_sb_lock); + spin_unlock_bh(&ipvs->sync_buff_lock); return sb; } +/* + * Switch mode from sending version 0 or 1 + * - must handle sync_buf + */ +void ip_vs_sync_switch_mode(struct net *net, int mode) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + + if (!(ipvs->sync_state & IP_VS_STATE_MASTER)) + return; + if (mode == sysctl_sync_ver(ipvs) || !ipvs->sync_buff) + return; + + spin_lock_bh(&ipvs->sync_buff_lock); + /* Buffer empty ? then let buf_create do the job */ + if (ipvs->sync_buff->mesg->size <= sizeof(struct ip_vs_sync_mesg)) { + kfree(ipvs->sync_buff); + ipvs->sync_buff = NULL; + } else { + spin_lock_bh(&ipvs->sync_lock); + if (ipvs->sync_state & IP_VS_STATE_MASTER) + list_add_tail(&ipvs->sync_buff->list, + &ipvs->sync_queue); + else + ip_vs_sync_buff_release(ipvs->sync_buff); + spin_unlock_bh(&ipvs->sync_lock); + } + spin_unlock_bh(&ipvs->sync_buff_lock); +} /* + * Create a new sync buffer for Version 0 proto. + */ +static inline struct ip_vs_sync_buff * +ip_vs_sync_buff_create_v0(struct netns_ipvs *ipvs) +{ + struct ip_vs_sync_buff *sb; + struct ip_vs_sync_mesg_v0 *mesg; + + if (!(sb=kmalloc(sizeof(struct ip_vs_sync_buff), GFP_ATOMIC))) + return NULL; + + sb->mesg = kmalloc(ipvs->send_mesg_maxlen, GFP_ATOMIC); + if (!sb->mesg) { + kfree(sb); + return NULL; + } + mesg = (struct ip_vs_sync_mesg_v0 *)sb->mesg; + mesg->nr_conns = 0; + mesg->syncid = ipvs->master_syncid; + mesg->size = sizeof(struct ip_vs_sync_mesg_v0); + sb->head = (unsigned char *)mesg + sizeof(struct ip_vs_sync_mesg_v0); + sb->end = (unsigned char *)mesg + ipvs->send_mesg_maxlen; + sb->firstuse = jiffies; + return sb; +} + +/* + * Version 0 , could be switched in by sys_ctl. * Add an ip_vs_conn information into the current sync_buff. - * Called by ip_vs_in. */ -void ip_vs_sync_conn(struct ip_vs_conn *cp) +void ip_vs_sync_conn_v0(struct net *net, struct ip_vs_conn *cp) { - struct ip_vs_sync_mesg *m; - struct ip_vs_sync_conn *s; + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_sync_mesg_v0 *m; + struct ip_vs_sync_conn_v0 *s; int len; - spin_lock(&curr_sb_lock); - if (!curr_sb) { - if (!(curr_sb=ip_vs_sync_buff_create())) { - spin_unlock(&curr_sb_lock); + if (unlikely(cp->af != AF_INET)) + return; + /* Do not sync ONE PACKET */ + if (cp->flags & IP_VS_CONN_F_ONE_PACKET) + return; + + spin_lock(&ipvs->sync_buff_lock); + if (!ipvs->sync_buff) { + ipvs->sync_buff = + ip_vs_sync_buff_create_v0(ipvs); + if (!ipvs->sync_buff) { + spin_unlock(&ipvs->sync_buff_lock); pr_err("ip_vs_sync_buff_create failed.\n"); return; } @@ -253,10 +471,11 @@ void ip_vs_sync_conn(struct ip_vs_conn *cp) len = (cp->flags & IP_VS_CONN_F_SEQ_MASK) ? FULL_CONN_SIZE : SIMPLE_CONN_SIZE; - m = curr_sb->mesg; - s = (struct ip_vs_sync_conn *)curr_sb->head; + m = (struct ip_vs_sync_mesg_v0 *)ipvs->sync_buff->mesg; + s = (struct ip_vs_sync_conn_v0 *)ipvs->sync_buff->head; /* copy members */ + s->reserved = 0; s->protocol = cp->protocol; s->cport = cp->cport; s->vport = cp->vport; @@ -274,83 +493,365 @@ void ip_vs_sync_conn(struct ip_vs_conn *cp) m->nr_conns++; m->size += len; - curr_sb->head += len; + ipvs->sync_buff->head += len; /* check if there is a space for next one */ - if (curr_sb->head+FULL_CONN_SIZE > curr_sb->end) { - sb_queue_tail(curr_sb); - curr_sb = NULL; + if (ipvs->sync_buff->head + FULL_CONN_SIZE > ipvs->sync_buff->end) { + sb_queue_tail(ipvs); + ipvs->sync_buff = NULL; } - spin_unlock(&curr_sb_lock); + spin_unlock(&ipvs->sync_buff_lock); /* synchronize its controller if it has */ if (cp->control) - ip_vs_sync_conn(cp->control); + ip_vs_sync_conn(net, cp->control); +} + +/* + * Add an ip_vs_conn information into the current sync_buff. + * Called by ip_vs_in. + * Sending Version 1 messages + */ +void ip_vs_sync_conn(struct net *net, struct ip_vs_conn *cp) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_sync_mesg *m; + union ip_vs_sync_conn *s; + __u8 *p; + unsigned int len, pe_name_len, pad; + + /* Handle old version of the protocol */ + if (sysctl_sync_ver(ipvs) == 0) { + ip_vs_sync_conn_v0(net, cp); + return; + } + /* Do not sync ONE PACKET */ + if (cp->flags & IP_VS_CONN_F_ONE_PACKET) + goto control; +sloop: + /* Sanity checks */ + pe_name_len = 0; + if (cp->pe_data_len) { + if (!cp->pe_data || !cp->dest) { + IP_VS_ERR_RL("SYNC, connection pe_data invalid\n"); + return; + } + pe_name_len = strnlen(cp->pe->name, IP_VS_PENAME_MAXLEN); + } + + spin_lock(&ipvs->sync_buff_lock); + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + len = sizeof(struct ip_vs_sync_v6); + else +#endif + len = sizeof(struct ip_vs_sync_v4); + + if (cp->flags & IP_VS_CONN_F_SEQ_MASK) + len += sizeof(struct ip_vs_sync_conn_options) + 2; + + if (cp->pe_data_len) + len += cp->pe_data_len + 2; /* + Param hdr field */ + if (pe_name_len) + len += pe_name_len + 2; + + /* check if there is a space for this one */ + pad = 0; + if (ipvs->sync_buff) { + pad = (4 - (size_t)ipvs->sync_buff->head) & 3; + if (ipvs->sync_buff->head + len + pad > ipvs->sync_buff->end) { + sb_queue_tail(ipvs); + ipvs->sync_buff = NULL; + pad = 0; + } + } + + if (!ipvs->sync_buff) { + ipvs->sync_buff = ip_vs_sync_buff_create(ipvs); + if (!ipvs->sync_buff) { + spin_unlock(&ipvs->sync_buff_lock); + pr_err("ip_vs_sync_buff_create failed.\n"); + return; + } + } + + m = ipvs->sync_buff->mesg; + p = ipvs->sync_buff->head; + ipvs->sync_buff->head += pad + len; + m->size += pad + len; + /* Add ev. padding from prev. sync_conn */ + while (pad--) + *(p++) = 0; + + s = (union ip_vs_sync_conn *)p; + + /* Set message type & copy members */ + s->v4.type = (cp->af == AF_INET6 ? STYPE_F_INET6 : 0); + s->v4.ver_size = htons(len & SVER_MASK); /* Version 0 */ + s->v4.flags = htonl(cp->flags & ~IP_VS_CONN_F_HASHED); + s->v4.state = htons(cp->state); + s->v4.protocol = cp->protocol; + s->v4.cport = cp->cport; + s->v4.vport = cp->vport; + s->v4.dport = cp->dport; + s->v4.fwmark = htonl(cp->fwmark); + s->v4.timeout = htonl(cp->timeout / HZ); + m->nr_conns++; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) { + p += sizeof(struct ip_vs_sync_v6); + ipv6_addr_copy(&s->v6.caddr, &cp->caddr.in6); + ipv6_addr_copy(&s->v6.vaddr, &cp->vaddr.in6); + ipv6_addr_copy(&s->v6.daddr, &cp->daddr.in6); + } else +#endif + { + p += sizeof(struct ip_vs_sync_v4); /* options ptr */ + s->v4.caddr = cp->caddr.ip; + s->v4.vaddr = cp->vaddr.ip; + s->v4.daddr = cp->daddr.ip; + } + if (cp->flags & IP_VS_CONN_F_SEQ_MASK) { + *(p++) = IPVS_OPT_SEQ_DATA; + *(p++) = sizeof(struct ip_vs_sync_conn_options); + hton_seq((struct ip_vs_seq *)p, &cp->in_seq); + p += sizeof(struct ip_vs_seq); + hton_seq((struct ip_vs_seq *)p, &cp->out_seq); + p += sizeof(struct ip_vs_seq); + } + /* Handle pe data */ + if (cp->pe_data_len && cp->pe_data) { + *(p++) = IPVS_OPT_PE_DATA; + *(p++) = cp->pe_data_len; + memcpy(p, cp->pe_data, cp->pe_data_len); + p += cp->pe_data_len; + if (pe_name_len) { + /* Add PE_NAME */ + *(p++) = IPVS_OPT_PE_NAME; + *(p++) = pe_name_len; + memcpy(p, cp->pe->name, pe_name_len); + p += pe_name_len; + } + } + + spin_unlock(&ipvs->sync_buff_lock); + +control: + /* synchronize its controller if it has */ + cp = cp->control; + if (!cp) + return; + /* + * Reduce sync rate for templates + * i.e only increment in_pkts for Templates. + */ + if (cp->flags & IP_VS_CONN_F_TEMPLATE) { + int pkts = atomic_add_return(1, &cp->in_pkts); + + if (pkts % sysctl_sync_period(ipvs) != 1) + return; + } + goto sloop; } +/* + * fill_param used by version 1 + */ static inline int -ip_vs_conn_fill_param_sync(int af, int protocol, - const union nf_inet_addr *caddr, __be16 cport, - const union nf_inet_addr *vaddr, __be16 vport, - struct ip_vs_conn_param *p) +ip_vs_conn_fill_param_sync(struct net *net, int af, union ip_vs_sync_conn *sc, + struct ip_vs_conn_param *p, + __u8 *pe_data, unsigned int pe_data_len, + __u8 *pe_name, unsigned int pe_name_len) { - /* XXX: Need to take into account persistence engine */ - ip_vs_conn_fill_param(af, protocol, caddr, cport, vaddr, vport, p); +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + ip_vs_conn_fill_param(net, af, sc->v6.protocol, + (const union nf_inet_addr *)&sc->v6.caddr, + sc->v6.cport, + (const union nf_inet_addr *)&sc->v6.vaddr, + sc->v6.vport, p); + else +#endif + ip_vs_conn_fill_param(net, af, sc->v4.protocol, + (const union nf_inet_addr *)&sc->v4.caddr, + sc->v4.cport, + (const union nf_inet_addr *)&sc->v4.vaddr, + sc->v4.vport, p); + /* Handle pe data */ + if (pe_data_len) { + if (pe_name_len) { + char buff[IP_VS_PENAME_MAXLEN+1]; + + memcpy(buff, pe_name, pe_name_len); + buff[pe_name_len]=0; + p->pe = __ip_vs_pe_getbyname(buff); + if (!p->pe) { + IP_VS_DBG(3, "BACKUP, no %s engine found/loaded\n", + buff); + return 1; + } + } else { + IP_VS_ERR_RL("BACKUP, Invalid PE parameters\n"); + return 1; + } + + p->pe_data = kmemdup(pe_data, pe_data_len, GFP_ATOMIC); + if (!p->pe_data) { + if (p->pe->module) + module_put(p->pe->module); + return -ENOMEM; + } + p->pe_data_len = pe_data_len; + } return 0; } /* - * Process received multicast message and create the corresponding - * ip_vs_conn entries. + * Connection Add / Update. + * Common for version 0 and 1 reception of backup sync_conns. + * Param: ... + * timeout is in sec. */ -static void ip_vs_process_message(const char *buffer, const size_t buflen) +static void ip_vs_proc_conn(struct net *net, struct ip_vs_conn_param *param, + unsigned int flags, unsigned int state, + unsigned int protocol, unsigned int type, + const union nf_inet_addr *daddr, __be16 dport, + unsigned long timeout, __u32 fwmark, + struct ip_vs_sync_conn_options *opt) { - struct ip_vs_sync_mesg *m = (struct ip_vs_sync_mesg *)buffer; - struct ip_vs_sync_conn *s; - struct ip_vs_sync_conn_options *opt; - struct ip_vs_conn *cp; - struct ip_vs_protocol *pp; struct ip_vs_dest *dest; - struct ip_vs_conn_param param; - char *p; - int i; + struct ip_vs_conn *cp; + struct netns_ipvs *ipvs = net_ipvs(net); - if (buflen < sizeof(struct ip_vs_sync_mesg)) { - IP_VS_ERR_RL("sync message header too short\n"); - return; - } + if (!(flags & IP_VS_CONN_F_TEMPLATE)) + cp = ip_vs_conn_in_get(param); + else + cp = ip_vs_ct_in_get(param); - /* Convert size back to host byte order */ - m->size = ntohs(m->size); + if (cp && param->pe_data) /* Free pe_data */ + kfree(param->pe_data); + if (!cp) { + /* + * Find the appropriate destination for the connection. + * If it is not found the connection will remain unbound + * but still handled. + */ + dest = ip_vs_find_dest(net, type, daddr, dport, param->vaddr, + param->vport, protocol, fwmark); - if (buflen != m->size) { - IP_VS_ERR_RL("bogus sync message size\n"); - return; + /* Set the approprite ativity flag */ + if (protocol == IPPROTO_TCP) { + if (state != IP_VS_TCP_S_ESTABLISHED) + flags |= IP_VS_CONN_F_INACTIVE; + else + flags &= ~IP_VS_CONN_F_INACTIVE; + } else if (protocol == IPPROTO_SCTP) { + if (state != IP_VS_SCTP_S_ESTABLISHED) + flags |= IP_VS_CONN_F_INACTIVE; + else + flags &= ~IP_VS_CONN_F_INACTIVE; + } + cp = ip_vs_conn_new(param, daddr, dport, flags, dest, fwmark); + if (dest) + atomic_dec(&dest->refcnt); + if (!cp) { + if (param->pe_data) + kfree(param->pe_data); + IP_VS_DBG(2, "BACKUP, add new conn. failed\n"); + return; + } + } else if (!cp->dest) { + dest = ip_vs_try_bind_dest(cp); + if (dest) + atomic_dec(&dest->refcnt); + } else if ((cp->dest) && (cp->protocol == IPPROTO_TCP) && + (cp->state != state)) { + /* update active/inactive flag for the connection */ + dest = cp->dest; + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (state != IP_VS_TCP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags |= IP_VS_CONN_F_INACTIVE; + } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && + (state == IP_VS_TCP_S_ESTABLISHED)) { + atomic_inc(&dest->activeconns); + atomic_dec(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } + } else if ((cp->dest) && (cp->protocol == IPPROTO_SCTP) && + (cp->state != state)) { + dest = cp->dest; + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (state != IP_VS_SCTP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } } - /* SyncID sanity check */ - if (ip_vs_backup_syncid != 0 && m->syncid != ip_vs_backup_syncid) { - IP_VS_DBG(7, "Ignoring incoming msg with syncid = %d\n", - m->syncid); - return; + if (opt) + memcpy(&cp->in_seq, opt, sizeof(*opt)); + atomic_set(&cp->in_pkts, sysctl_sync_threshold(ipvs)); + cp->state = state; + cp->old_state = cp->state; + /* + * For Ver 0 messages style + * - Not possible to recover the right timeout for templates + * - can not find the right fwmark + * virtual service. If needed, we can do it for + * non-fwmark persistent services. + * Ver 1 messages style. + * - No problem. + */ + if (timeout) { + if (timeout > MAX_SCHEDULE_TIMEOUT / HZ) + timeout = MAX_SCHEDULE_TIMEOUT / HZ; + cp->timeout = timeout*HZ; + } else { + struct ip_vs_proto_data *pd; + + pd = ip_vs_proto_data_get(net, protocol); + if (!(flags & IP_VS_CONN_F_TEMPLATE) && pd && pd->timeout_table) + cp->timeout = pd->timeout_table[state]; + else + cp->timeout = (3*60*HZ); } + ip_vs_conn_put(cp); +} - p = (char *)buffer + sizeof(struct ip_vs_sync_mesg); +/* + * Process received multicast message for Version 0 + */ +static void ip_vs_process_message_v0(struct net *net, const char *buffer, + const size_t buflen) +{ + struct ip_vs_sync_mesg_v0 *m = (struct ip_vs_sync_mesg_v0 *)buffer; + struct ip_vs_sync_conn_v0 *s; + struct ip_vs_sync_conn_options *opt; + struct ip_vs_protocol *pp; + struct ip_vs_conn_param param; + char *p; + int i; + + p = (char *)buffer + sizeof(struct ip_vs_sync_mesg_v0); for (i=0; i<m->nr_conns; i++) { unsigned flags, state; if (p + SIMPLE_CONN_SIZE > buffer+buflen) { - IP_VS_ERR_RL("bogus conn in sync message\n"); + IP_VS_ERR_RL("BACKUP v0, bogus conn\n"); return; } - s = (struct ip_vs_sync_conn *) p; + s = (struct ip_vs_sync_conn_v0 *) p; flags = ntohs(s->flags) | IP_VS_CONN_F_SYNC; flags &= ~IP_VS_CONN_F_HASHED; if (flags & IP_VS_CONN_F_SEQ_MASK) { opt = (struct ip_vs_sync_conn_options *)&s[1]; p += FULL_CONN_SIZE; if (p > buffer+buflen) { - IP_VS_ERR_RL("bogus conn options in sync message\n"); + IP_VS_ERR_RL("BACKUP v0, Dropping buffer bogus conn options\n"); return; } } else { @@ -362,118 +863,286 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) if (!(flags & IP_VS_CONN_F_TEMPLATE)) { pp = ip_vs_proto_get(s->protocol); if (!pp) { - IP_VS_ERR_RL("Unsupported protocol %u in sync msg\n", + IP_VS_DBG(2, "BACKUP v0, Unsupported protocol %u\n", s->protocol); continue; } if (state >= pp->num_states) { - IP_VS_DBG(2, "Invalid %s state %u in sync msg\n", + IP_VS_DBG(2, "BACKUP v0, Invalid %s state %u\n", pp->name, state); continue; } } else { /* protocol in templates is not used for state/timeout */ - pp = NULL; if (state > 0) { - IP_VS_DBG(2, "Invalid template state %u in sync msg\n", + IP_VS_DBG(2, "BACKUP v0, Invalid template state %u\n", state); state = 0; } } - { - if (ip_vs_conn_fill_param_sync(AF_INET, s->protocol, - (union nf_inet_addr *)&s->caddr, - s->cport, - (union nf_inet_addr *)&s->vaddr, - s->vport, ¶m)) { - pr_err("ip_vs_conn_fill_param_sync failed"); - return; + ip_vs_conn_fill_param(net, AF_INET, s->protocol, + (const union nf_inet_addr *)&s->caddr, + s->cport, + (const union nf_inet_addr *)&s->vaddr, + s->vport, ¶m); + + /* Send timeout as Zero */ + ip_vs_proc_conn(net, ¶m, flags, state, s->protocol, AF_INET, + (union nf_inet_addr *)&s->daddr, s->dport, + 0, 0, opt); + } +} + +/* + * Handle options + */ +static inline int ip_vs_proc_seqopt(__u8 *p, unsigned int plen, + __u32 *opt_flags, + struct ip_vs_sync_conn_options *opt) +{ + struct ip_vs_sync_conn_options *topt; + + topt = (struct ip_vs_sync_conn_options *)p; + + if (plen != sizeof(struct ip_vs_sync_conn_options)) { + IP_VS_DBG(2, "BACKUP, bogus conn options length\n"); + return -EINVAL; + } + if (*opt_flags & IPVS_OPT_F_SEQ_DATA) { + IP_VS_DBG(2, "BACKUP, conn options found twice\n"); + return -EINVAL; + } + ntoh_seq(&topt->in_seq, &opt->in_seq); + ntoh_seq(&topt->out_seq, &opt->out_seq); + *opt_flags |= IPVS_OPT_F_SEQ_DATA; + return 0; +} + +static int ip_vs_proc_str(__u8 *p, unsigned int plen, unsigned int *data_len, + __u8 **data, unsigned int maxlen, + __u32 *opt_flags, __u32 flag) +{ + if (plen > maxlen) { + IP_VS_DBG(2, "BACKUP, bogus par.data len > %d\n", maxlen); + return -EINVAL; + } + if (*opt_flags & flag) { + IP_VS_DBG(2, "BACKUP, Par.data found twice 0x%x\n", flag); + return -EINVAL; + } + *data_len = plen; + *data = p; + *opt_flags |= flag; + return 0; +} +/* + * Process a Version 1 sync. connection + */ +static inline int ip_vs_proc_sync_conn(struct net *net, __u8 *p, __u8 *msg_end) +{ + struct ip_vs_sync_conn_options opt; + union ip_vs_sync_conn *s; + struct ip_vs_protocol *pp; + struct ip_vs_conn_param param; + __u32 flags; + unsigned int af, state, pe_data_len=0, pe_name_len=0; + __u8 *pe_data=NULL, *pe_name=NULL; + __u32 opt_flags=0; + int retc=0; + + s = (union ip_vs_sync_conn *) p; + + if (s->v6.type & STYPE_F_INET6) { +#ifdef CONFIG_IP_VS_IPV6 + af = AF_INET6; + p += sizeof(struct ip_vs_sync_v6); +#else + IP_VS_DBG(3,"BACKUP, IPv6 msg received, and IPVS is not compiled for IPv6\n"); + retc = 10; + goto out; +#endif + } else if (!s->v4.type) { + af = AF_INET; + p += sizeof(struct ip_vs_sync_v4); + } else { + return -10; + } + if (p > msg_end) + return -20; + + /* Process optional params check Type & Len. */ + while (p < msg_end) { + int ptype; + int plen; + + if (p+2 > msg_end) + return -30; + ptype = *(p++); + plen = *(p++); + + if (!plen || ((p + plen) > msg_end)) + return -40; + /* Handle seq option p = param data */ + switch (ptype & ~IPVS_OPT_F_PARAM) { + case IPVS_OPT_SEQ_DATA: + if (ip_vs_proc_seqopt(p, plen, &opt_flags, &opt)) + return -50; + break; + + case IPVS_OPT_PE_DATA: + if (ip_vs_proc_str(p, plen, &pe_data_len, &pe_data, + IP_VS_PEDATA_MAXLEN, &opt_flags, + IPVS_OPT_F_PE_DATA)) + return -60; + break; + + case IPVS_OPT_PE_NAME: + if (ip_vs_proc_str(p, plen,&pe_name_len, &pe_name, + IP_VS_PENAME_MAXLEN, &opt_flags, + IPVS_OPT_F_PE_NAME)) + return -70; + break; + + default: + /* Param data mandatory ? */ + if (!(ptype & IPVS_OPT_F_PARAM)) { + IP_VS_DBG(3, "BACKUP, Unknown mandatory param %d found\n", + ptype & ~IPVS_OPT_F_PARAM); + retc = 20; + goto out; } - if (!(flags & IP_VS_CONN_F_TEMPLATE)) - cp = ip_vs_conn_in_get(¶m); - else - cp = ip_vs_ct_in_get(¶m); } - if (!cp) { - /* - * Find the appropriate destination for the connection. - * If it is not found the connection will remain unbound - * but still handled. - */ - dest = ip_vs_find_dest(AF_INET, - (union nf_inet_addr *)&s->daddr, - s->dport, - (union nf_inet_addr *)&s->vaddr, - s->vport, - s->protocol); - /* Set the approprite ativity flag */ - if (s->protocol == IPPROTO_TCP) { - if (state != IP_VS_TCP_S_ESTABLISHED) - flags |= IP_VS_CONN_F_INACTIVE; - else - flags &= ~IP_VS_CONN_F_INACTIVE; - } else if (s->protocol == IPPROTO_SCTP) { - if (state != IP_VS_SCTP_S_ESTABLISHED) - flags |= IP_VS_CONN_F_INACTIVE; - else - flags &= ~IP_VS_CONN_F_INACTIVE; + p += plen; /* Next option */ + } + + /* Get flags and Mask off unsupported */ + flags = ntohl(s->v4.flags) & IP_VS_CONN_F_BACKUP_MASK; + flags |= IP_VS_CONN_F_SYNC; + state = ntohs(s->v4.state); + + if (!(flags & IP_VS_CONN_F_TEMPLATE)) { + pp = ip_vs_proto_get(s->v4.protocol); + if (!pp) { + IP_VS_DBG(3,"BACKUP, Unsupported protocol %u\n", + s->v4.protocol); + retc = 30; + goto out; + } + if (state >= pp->num_states) { + IP_VS_DBG(3, "BACKUP, Invalid %s state %u\n", + pp->name, state); + retc = 40; + goto out; + } + } else { + /* protocol in templates is not used for state/timeout */ + if (state > 0) { + IP_VS_DBG(3, "BACKUP, Invalid template state %u\n", + state); + state = 0; + } + } + if (ip_vs_conn_fill_param_sync(net, af, s, ¶m, pe_data, + pe_data_len, pe_name, pe_name_len)) { + retc = 50; + goto out; + } + /* If only IPv4, just silent skip IPv6 */ + if (af == AF_INET) + ip_vs_proc_conn(net, ¶m, flags, state, s->v4.protocol, af, + (union nf_inet_addr *)&s->v4.daddr, s->v4.dport, + ntohl(s->v4.timeout), ntohl(s->v4.fwmark), + (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL) + ); +#ifdef CONFIG_IP_VS_IPV6 + else + ip_vs_proc_conn(net, ¶m, flags, state, s->v6.protocol, af, + (union nf_inet_addr *)&s->v6.daddr, s->v6.dport, + ntohl(s->v6.timeout), ntohl(s->v6.fwmark), + (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL) + ); +#endif + return 0; + /* Error exit */ +out: + IP_VS_DBG(2, "BACKUP, Single msg dropped err:%d\n", retc); + return retc; + +} +/* + * Process received multicast message and create the corresponding + * ip_vs_conn entries. + * Handles Version 0 & 1 + */ +static void ip_vs_process_message(struct net *net, __u8 *buffer, + const size_t buflen) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_sync_mesg *m2 = (struct ip_vs_sync_mesg *)buffer; + __u8 *p, *msg_end; + int i, nr_conns; + + if (buflen < sizeof(struct ip_vs_sync_mesg_v0)) { + IP_VS_DBG(2, "BACKUP, message header too short\n"); + return; + } + /* Convert size back to host byte order */ + m2->size = ntohs(m2->size); + + if (buflen != m2->size) { + IP_VS_DBG(2, "BACKUP, bogus message size\n"); + return; + } + /* SyncID sanity check */ + if (ipvs->backup_syncid != 0 && m2->syncid != ipvs->backup_syncid) { + IP_VS_DBG(7, "BACKUP, Ignoring syncid = %d\n", m2->syncid); + return; + } + /* Handle version 1 message */ + if ((m2->version == SYNC_PROTO_VER) && (m2->reserved == 0) + && (m2->spare == 0)) { + + msg_end = buffer + sizeof(struct ip_vs_sync_mesg); + nr_conns = m2->nr_conns; + + for (i=0; i<nr_conns; i++) { + union ip_vs_sync_conn *s; + unsigned size; + int retc; + + p = msg_end; + if (p + sizeof(s->v4) > buffer+buflen) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, to small\n"); + return; } - cp = ip_vs_conn_new(¶m, - (union nf_inet_addr *)&s->daddr, - s->dport, flags, dest); - if (dest) - atomic_dec(&dest->refcnt); - if (!cp) { - pr_err("ip_vs_conn_new failed\n"); + s = (union ip_vs_sync_conn *)p; + size = ntohs(s->v4.ver_size) & SVER_MASK; + msg_end = p + size; + /* Basic sanity checks */ + if (msg_end > buffer+buflen) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, msg > buffer\n"); return; } - } else if (!cp->dest) { - dest = ip_vs_try_bind_dest(cp); - if (dest) - atomic_dec(&dest->refcnt); - } else if ((cp->dest) && (cp->protocol == IPPROTO_TCP) && - (cp->state != state)) { - /* update active/inactive flag for the connection */ - dest = cp->dest; - if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && - (state != IP_VS_TCP_S_ESTABLISHED)) { - atomic_dec(&dest->activeconns); - atomic_inc(&dest->inactconns); - cp->flags |= IP_VS_CONN_F_INACTIVE; - } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && - (state == IP_VS_TCP_S_ESTABLISHED)) { - atomic_inc(&dest->activeconns); - atomic_dec(&dest->inactconns); - cp->flags &= ~IP_VS_CONN_F_INACTIVE; + if (ntohs(s->v4.ver_size) >> SVER_SHIFT) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, Unknown version %d\n", + ntohs(s->v4.ver_size) >> SVER_SHIFT); + return; } - } else if ((cp->dest) && (cp->protocol == IPPROTO_SCTP) && - (cp->state != state)) { - dest = cp->dest; - if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && - (state != IP_VS_SCTP_S_ESTABLISHED)) { - atomic_dec(&dest->activeconns); - atomic_inc(&dest->inactconns); - cp->flags &= ~IP_VS_CONN_F_INACTIVE; + /* Process a single sync_conn */ + retc = ip_vs_proc_sync_conn(net, p, msg_end); + if (retc < 0) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, Err: %d in decoding\n", + retc); + return; } + /* Make sure we have 32 bit alignment */ + msg_end = p + ((size + 3) & ~3); } - - if (opt) - memcpy(&cp->in_seq, opt, sizeof(*opt)); - atomic_set(&cp->in_pkts, sysctl_ip_vs_sync_threshold[0]); - cp->state = state; - cp->old_state = cp->state; - /* - * We can not recover the right timeout for templates - * in all cases, we can not find the right fwmark - * virtual service. If needed, we can do it for - * non-fwmark persistent services. - */ - if (!(flags & IP_VS_CONN_F_TEMPLATE) && pp->timeout_table) - cp->timeout = pp->timeout_table[state]; - else - cp->timeout = (3*60*HZ); - ip_vs_conn_put(cp); + } else { + /* Old type of message */ + ip_vs_process_message_v0(net, buffer, buflen); + return; } } @@ -511,8 +1180,10 @@ static int set_mcast_if(struct sock *sk, char *ifname) { struct net_device *dev; struct inet_sock *inet = inet_sk(sk); + struct net *net = sock_net(sk); - if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + dev = __dev_get_by_name(net, ifname); + if (!dev) return -ENODEV; if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if) @@ -531,30 +1202,33 @@ static int set_mcast_if(struct sock *sk, char *ifname) * Set the maximum length of sync message according to the * specified interface's MTU. */ -static int set_sync_mesg_maxlen(int sync_state) +static int set_sync_mesg_maxlen(struct net *net, int sync_state) { + struct netns_ipvs *ipvs = net_ipvs(net); struct net_device *dev; int num; if (sync_state == IP_VS_STATE_MASTER) { - if ((dev = __dev_get_by_name(&init_net, ip_vs_master_mcast_ifn)) == NULL) + dev = __dev_get_by_name(net, ipvs->master_mcast_ifn); + if (!dev) return -ENODEV; num = (dev->mtu - sizeof(struct iphdr) - sizeof(struct udphdr) - SYNC_MESG_HEADER_LEN - 20) / SIMPLE_CONN_SIZE; - sync_send_mesg_maxlen = SYNC_MESG_HEADER_LEN + + ipvs->send_mesg_maxlen = SYNC_MESG_HEADER_LEN + SIMPLE_CONN_SIZE * min(num, MAX_CONNS_PER_SYNCBUFF); IP_VS_DBG(7, "setting the maximum length of sync sending " - "message %d.\n", sync_send_mesg_maxlen); + "message %d.\n", ipvs->send_mesg_maxlen); } else if (sync_state == IP_VS_STATE_BACKUP) { - if ((dev = __dev_get_by_name(&init_net, ip_vs_backup_mcast_ifn)) == NULL) + dev = __dev_get_by_name(net, ipvs->backup_mcast_ifn); + if (!dev) return -ENODEV; - sync_recv_mesg_maxlen = dev->mtu - + ipvs->recv_mesg_maxlen = dev->mtu - sizeof(struct iphdr) - sizeof(struct udphdr); IP_VS_DBG(7, "setting the maximum length of sync receiving " - "message %d.\n", sync_recv_mesg_maxlen); + "message %d.\n", ipvs->recv_mesg_maxlen); } return 0; @@ -569,6 +1243,7 @@ static int set_sync_mesg_maxlen(int sync_state) static int join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname) { + struct net *net = sock_net(sk); struct ip_mreqn mreq; struct net_device *dev; int ret; @@ -576,7 +1251,8 @@ join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname) memset(&mreq, 0, sizeof(mreq)); memcpy(&mreq.imr_multiaddr, addr, sizeof(struct in_addr)); - if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + dev = __dev_get_by_name(net, ifname); + if (!dev) return -ENODEV; if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if) return -EINVAL; @@ -593,11 +1269,13 @@ join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname) static int bind_mcastif_addr(struct socket *sock, char *ifname) { + struct net *net = sock_net(sock->sk); struct net_device *dev; __be32 addr; struct sockaddr_in sin; - if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + dev = __dev_get_by_name(net, ifname); + if (!dev) return -ENODEV; addr = inet_select_addr(dev, 0, RT_SCOPE_UNIVERSE); @@ -619,19 +1297,20 @@ static int bind_mcastif_addr(struct socket *sock, char *ifname) /* * Set up sending multicast socket over UDP */ -static struct socket * make_send_sock(void) +static struct socket *make_send_sock(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); struct socket *sock; int result; /* First create a socket */ - result = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock); + result = __sock_create(net, PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock, 1); if (result < 0) { pr_err("Error during creation of socket; terminating\n"); return ERR_PTR(result); } - result = set_mcast_if(sock->sk, ip_vs_master_mcast_ifn); + result = set_mcast_if(sock->sk, ipvs->master_mcast_ifn); if (result < 0) { pr_err("Error setting outbound mcast interface\n"); goto error; @@ -640,7 +1319,7 @@ static struct socket * make_send_sock(void) set_mcast_loop(sock->sk, 0); set_mcast_ttl(sock->sk, 1); - result = bind_mcastif_addr(sock, ip_vs_master_mcast_ifn); + result = bind_mcastif_addr(sock, ipvs->master_mcast_ifn); if (result < 0) { pr_err("Error binding address of the mcast interface\n"); goto error; @@ -664,13 +1343,14 @@ static struct socket * make_send_sock(void) /* * Set up receiving multicast socket over UDP */ -static struct socket * make_receive_sock(void) +static struct socket *make_receive_sock(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); struct socket *sock; int result; /* First create a socket */ - result = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock); + result = __sock_create(net, PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock, 1); if (result < 0) { pr_err("Error during creation of socket; terminating\n"); return ERR_PTR(result); @@ -689,7 +1369,7 @@ static struct socket * make_receive_sock(void) /* join the multicast group */ result = join_mcast_group(sock->sk, (struct in_addr *) &mcast_addr.sin_addr, - ip_vs_backup_mcast_ifn); + ipvs->backup_mcast_ifn); if (result < 0) { pr_err("Error joining to the multicast group\n"); goto error; @@ -760,20 +1440,21 @@ ip_vs_receive(struct socket *sock, char *buffer, const size_t buflen) static int sync_thread_master(void *data) { struct ip_vs_sync_thread_data *tinfo = data; + struct netns_ipvs *ipvs = net_ipvs(tinfo->net); struct ip_vs_sync_buff *sb; pr_info("sync thread started: state = MASTER, mcast_ifn = %s, " "syncid = %d\n", - ip_vs_master_mcast_ifn, ip_vs_master_syncid); + ipvs->master_mcast_ifn, ipvs->master_syncid); while (!kthread_should_stop()) { - while ((sb = sb_dequeue())) { + while ((sb = sb_dequeue(ipvs))) { ip_vs_send_sync_msg(tinfo->sock, sb->mesg); ip_vs_sync_buff_release(sb); } - /* check if entries stay in curr_sb for 2 seconds */ - sb = get_curr_sync_buff(2 * HZ); + /* check if entries stay in ipvs->sync_buff for 2 seconds */ + sb = get_curr_sync_buff(ipvs, 2 * HZ); if (sb) { ip_vs_send_sync_msg(tinfo->sock, sb->mesg); ip_vs_sync_buff_release(sb); @@ -783,14 +1464,13 @@ static int sync_thread_master(void *data) } /* clean up the sync_buff queue */ - while ((sb=sb_dequeue())) { + while ((sb = sb_dequeue(ipvs))) ip_vs_sync_buff_release(sb); - } /* clean up the current sync_buff */ - if ((sb = get_curr_sync_buff(0))) { + sb = get_curr_sync_buff(ipvs, 0); + if (sb) ip_vs_sync_buff_release(sb); - } /* release the sending multicast socket */ sock_release(tinfo->sock); @@ -803,11 +1483,12 @@ static int sync_thread_master(void *data) static int sync_thread_backup(void *data) { struct ip_vs_sync_thread_data *tinfo = data; + struct netns_ipvs *ipvs = net_ipvs(tinfo->net); int len; pr_info("sync thread started: state = BACKUP, mcast_ifn = %s, " "syncid = %d\n", - ip_vs_backup_mcast_ifn, ip_vs_backup_syncid); + ipvs->backup_mcast_ifn, ipvs->backup_syncid); while (!kthread_should_stop()) { wait_event_interruptible(*sk_sleep(tinfo->sock->sk), @@ -817,7 +1498,7 @@ static int sync_thread_backup(void *data) /* do we have data now? */ while (!skb_queue_empty(&(tinfo->sock->sk->sk_receive_queue))) { len = ip_vs_receive(tinfo->sock, tinfo->buf, - sync_recv_mesg_maxlen); + ipvs->recv_mesg_maxlen); if (len <= 0) { pr_err("receiving message error\n"); break; @@ -826,7 +1507,7 @@ static int sync_thread_backup(void *data) /* disable bottom half, because it accesses the data shared by softirq while getting/creating conns */ local_bh_disable(); - ip_vs_process_message(tinfo->buf, len); + ip_vs_process_message(tinfo->net, tinfo->buf, len); local_bh_enable(); } } @@ -840,41 +1521,42 @@ static int sync_thread_backup(void *data) } -int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) +int start_sync_thread(struct net *net, int state, char *mcast_ifn, __u8 syncid) { struct ip_vs_sync_thread_data *tinfo; struct task_struct **realtask, *task; struct socket *sock; + struct netns_ipvs *ipvs = net_ipvs(net); char *name, *buf = NULL; int (*threadfn)(void *data); int result = -ENOMEM; IP_VS_DBG(7, "%s(): pid %d\n", __func__, task_pid_nr(current)); IP_VS_DBG(7, "Each ip_vs_sync_conn entry needs %Zd bytes\n", - sizeof(struct ip_vs_sync_conn)); + sizeof(struct ip_vs_sync_conn_v0)); if (state == IP_VS_STATE_MASTER) { - if (sync_master_thread) + if (ipvs->master_thread) return -EEXIST; - strlcpy(ip_vs_master_mcast_ifn, mcast_ifn, - sizeof(ip_vs_master_mcast_ifn)); - ip_vs_master_syncid = syncid; - realtask = &sync_master_thread; - name = "ipvs_syncmaster"; + strlcpy(ipvs->master_mcast_ifn, mcast_ifn, + sizeof(ipvs->master_mcast_ifn)); + ipvs->master_syncid = syncid; + realtask = &ipvs->master_thread; + name = "ipvs_master:%d"; threadfn = sync_thread_master; - sock = make_send_sock(); + sock = make_send_sock(net); } else if (state == IP_VS_STATE_BACKUP) { - if (sync_backup_thread) + if (ipvs->backup_thread) return -EEXIST; - strlcpy(ip_vs_backup_mcast_ifn, mcast_ifn, - sizeof(ip_vs_backup_mcast_ifn)); - ip_vs_backup_syncid = syncid; - realtask = &sync_backup_thread; - name = "ipvs_syncbackup"; + strlcpy(ipvs->backup_mcast_ifn, mcast_ifn, + sizeof(ipvs->backup_mcast_ifn)); + ipvs->backup_syncid = syncid; + realtask = &ipvs->backup_thread; + name = "ipvs_backup:%d"; threadfn = sync_thread_backup; - sock = make_receive_sock(); + sock = make_receive_sock(net); } else { return -EINVAL; } @@ -884,9 +1566,9 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) goto out; } - set_sync_mesg_maxlen(state); + set_sync_mesg_maxlen(net, state); if (state == IP_VS_STATE_BACKUP) { - buf = kmalloc(sync_recv_mesg_maxlen, GFP_KERNEL); + buf = kmalloc(ipvs->recv_mesg_maxlen, GFP_KERNEL); if (!buf) goto outsocket; } @@ -895,10 +1577,11 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) if (!tinfo) goto outbuf; + tinfo->net = net; tinfo->sock = sock; tinfo->buf = buf; - task = kthread_run(threadfn, tinfo, name); + task = kthread_run(threadfn, tinfo, name, ipvs->gen); if (IS_ERR(task)) { result = PTR_ERR(task); goto outtinfo; @@ -906,7 +1589,7 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) /* mark as active */ *realtask = task; - ip_vs_sync_state |= state; + ipvs->sync_state |= state; /* increase the module use count */ ip_vs_use_count_inc(); @@ -924,16 +1607,18 @@ out: } -int stop_sync_thread(int state) +int stop_sync_thread(struct net *net, int state) { + struct netns_ipvs *ipvs = net_ipvs(net); + IP_VS_DBG(7, "%s(): pid %d\n", __func__, task_pid_nr(current)); if (state == IP_VS_STATE_MASTER) { - if (!sync_master_thread) + if (!ipvs->master_thread) return -ESRCH; pr_info("stopping master sync thread %d ...\n", - task_pid_nr(sync_master_thread)); + task_pid_nr(ipvs->master_thread)); /* * The lock synchronizes with sb_queue_tail(), so that we don't @@ -941,21 +1626,21 @@ int stop_sync_thread(int state) * progress of stopping the master sync daemon. */ - spin_lock_bh(&ip_vs_sync_lock); - ip_vs_sync_state &= ~IP_VS_STATE_MASTER; - spin_unlock_bh(&ip_vs_sync_lock); - kthread_stop(sync_master_thread); - sync_master_thread = NULL; + spin_lock_bh(&ipvs->sync_lock); + ipvs->sync_state &= ~IP_VS_STATE_MASTER; + spin_unlock_bh(&ipvs->sync_lock); + kthread_stop(ipvs->master_thread); + ipvs->master_thread = NULL; } else if (state == IP_VS_STATE_BACKUP) { - if (!sync_backup_thread) + if (!ipvs->backup_thread) return -ESRCH; pr_info("stopping backup sync thread %d ...\n", - task_pid_nr(sync_backup_thread)); + task_pid_nr(ipvs->backup_thread)); - ip_vs_sync_state &= ~IP_VS_STATE_BACKUP; - kthread_stop(sync_backup_thread); - sync_backup_thread = NULL; + ipvs->sync_state &= ~IP_VS_STATE_BACKUP; + kthread_stop(ipvs->backup_thread); + ipvs->backup_thread = NULL; } else { return -EINVAL; } @@ -965,3 +1650,42 @@ int stop_sync_thread(int state) return 0; } + +/* + * Initialize data struct for each netns + */ +static int __net_init __ip_vs_sync_init(struct net *net) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + + INIT_LIST_HEAD(&ipvs->sync_queue); + spin_lock_init(&ipvs->sync_lock); + spin_lock_init(&ipvs->sync_buff_lock); + + ipvs->sync_mcast_addr.sin_family = AF_INET; + ipvs->sync_mcast_addr.sin_port = cpu_to_be16(IP_VS_SYNC_PORT); + ipvs->sync_mcast_addr.sin_addr.s_addr = cpu_to_be32(IP_VS_SYNC_GROUP); + return 0; +} + +static void __ip_vs_sync_cleanup(struct net *net) +{ + stop_sync_thread(net, IP_VS_STATE_MASTER); + stop_sync_thread(net, IP_VS_STATE_BACKUP); +} + +static struct pernet_operations ipvs_sync_ops = { + .init = __ip_vs_sync_init, + .exit = __ip_vs_sync_cleanup, +}; + + +int __init ip_vs_sync_init(void) +{ + return register_pernet_subsys(&ipvs_sync_ops); +} + +void ip_vs_sync_cleanup(void) +{ + unregister_pernet_subsys(&ipvs_sync_ops); +} |