IPVS: Allow configuration of persistence engines

Allow the persistence engine of a virtual service to be set, edited
and unset.

This feature only works with the netlink user-space interface.

Signed-off-by: Simon Horman <horms@verge.net.au>
Acked-by: Julian Anastasov <ja@ssi.bg>

3 files changed, 58 insertions, 3 deletions

diff --git a/include/linux/ip_vs.h b/include/linux/ip_vs.h
index 0a9c44d64292..5f43a3b2e3ad 100644
--- a/include/linux/ip_vs.h
+++ b/include/linux/ip_vs.h
@@ -336,6 +336,9 @@ enum {
         IPVS_SVC_ATTR_NETMASK,          /* persistent netmask */
 
         IPVS_SVC_ATTR_STATS,            /* nested attribute for service stats */
+
+        IPVS_SVC_ATTR_PE_NAME,          /* name of ct retriever */
+
         __IPVS_SVC_ATTR_MAX,
 };
 
diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index a6b93a26d4e2..52fbe2308c38 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -444,6 +444,7 @@ struct ip_vs_service_user_kern {
 
         /* virtual service options */
         char                    *sched_name;
+        char                    *pe_name;
         unsigned                flags;          /* virtual service flags */
         unsigned                timeout;        /* persistent timeout in sec */
         u32                     netmask;        /* persistent netmask */
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 5f20caf47a1d..a697591d0e23 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -1134,6 +1134,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
 {
         int ret = 0;
         struct ip_vs_scheduler *sched = NULL;
+        struct ip_vs_pe *pe = NULL;
         struct ip_vs_service *svc = NULL;
 
         /* increase the module use count */
@@ -1147,6 +1148,16 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
                 goto out_err;
         }
 
+        if (u->pe_name && *u->pe_name) {
+                pe = ip_vs_pe_get(u->pe_name);
+                if (pe == NULL) {
+                        pr_info("persistence engine module ip_vs_pe_%s "
+                                "not found\n", u->pe_name);
+                        ret = -ENOENT;
+                        goto out_err;
+                }
+        }
+
 #ifdef CONFIG_IP_VS_IPV6
         if (u->af == AF_INET6 && (u->netmask < 1 || u->netmask > 128)) {
                 ret = -EINVAL;
@@ -1184,6 +1195,10 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
                 goto out_err;
         sched = NULL;
 
+        /* Bind the ct retriever */
+        ip_vs_bind_pe(svc, pe);
+        pe = NULL;
+
         /* Update the virtual service counters */
         if (svc->port == FTPPORT)
                 atomic_inc(&ip_vs_ftpsvc_counter);
@@ -1215,6 +1230,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
                 kfree(svc);
         }
         ip_vs_scheduler_put(sched);
+        ip_vs_pe_put(pe);
 
         /* decrease the module use count */
         ip_vs_use_count_dec();
@@ -1230,6 +1246,7 @@ static int
 ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u)
 {
         struct ip_vs_scheduler *sched, *old_sched;
+        struct ip_vs_pe *pe = NULL, *old_pe = NULL;
         int ret = 0;
 
         /*
@@ -1242,6 +1259,17 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u)
         }
         old_sched = sched;
 
+        if (u->pe_name && *u->pe_name) {
+                pe = ip_vs_pe_get(u->pe_name);
+                if (pe == NULL) {
+                        pr_info("persistence engine module ip_vs_pe_%s "
+                                "not found\n", u->pe_name);
+                        ret = -ENOENT;
+                        goto out;
+                }
+                old_pe = pe;
+        }
+
 #ifdef CONFIG_IP_VS_IPV6
         if (u->af == AF_INET6 && (u->netmask < 1 || u->netmask > 128)) {
                 ret = -EINVAL;
@@ -1293,12 +1321,17 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u)
                 }
         }
 
+        old_pe = svc->pe;
+        if (pe != old_pe) {
+                ip_vs_unbind_pe(svc);
+                ip_vs_bind_pe(svc, pe);
+        }
+
   out_unlock:
         write_unlock_bh(&__ip_vs_svc_lock);
-#ifdef CONFIG_IP_VS_IPV6
   out:
-#endif
         ip_vs_scheduler_put(old_sched);
+        ip_vs_pe_put(old_pe);
         return ret;
 }
 
@@ -1312,6 +1345,9 @@ static void __ip_vs_del_service(struct ip_vs_service *svc)
 {
         struct ip_vs_dest *dest, *nxt;
         struct ip_vs_scheduler *old_sched;
+        struct ip_vs_pe *old_pe;
+
+        pr_info("%s: enter\n", __func__);
 
         /* Count only IPv4 services for old get/setsockopt interface */
         if (svc->af == AF_INET)
@@ -1324,6 +1360,11 @@ static void __ip_vs_del_service(struct ip_vs_service *svc)
         ip_vs_unbind_scheduler(svc);
         ip_vs_scheduler_put(old_sched);
 
+        /* Unbind persistence engine */
+        old_pe = svc->pe;
+        ip_vs_unbind_pe(svc);
+        ip_vs_pe_put(old_pe);
+
         /* Unbind app inc */
         if (svc->inc) {
                 ip_vs_app_inc_put(svc->inc);
@@ -2026,6 +2067,8 @@ static const unsigned char set_arglen[SET_CMDID(IP_VS_SO_SET_MAX)+1] = {
 static void ip_vs_copy_usvc_compat(struct ip_vs_service_user_kern *usvc,
                                   struct ip_vs_service_user *usvc_compat)
 {
+        memset(usvc, 0, sizeof(*usvc));
+
         usvc->af                = AF_INET;
         usvc->protocol          = usvc_compat->protocol;
         usvc->addr.ip           = usvc_compat->addr;
@@ -2043,6 +2086,8 @@ static void ip_vs_copy_usvc_compat(struct ip_vs_service_user_kern *usvc,
 static void ip_vs_copy_udest_compat(struct ip_vs_dest_user_kern *udest,
                                    struct ip_vs_dest_user *udest_compat)
 {
+        memset(udest, 0, sizeof(*udest));
+
         udest->addr.ip          = udest_compat->addr;
         udest->port             = udest_compat->port;