diff mbox

[V8,01/17] filter-rewriter: fix memory leak for connection in connection_track_table

Message ID 20180603050546.6827-2-zhangckid@gmail.com (mailing list archive)
State New, archived
Headers show

Commit Message

Zhang Chen June 3, 2018, 5:05 a.m. UTC
After a net connection is closed, we didn't clear its releated resources
in connection_track_table, which will lead to memory leak.

Let't track the state of net connection, if it is closed, its related
resources will be cleared up.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
---
 net/colo.h            |  4 +++
 net/filter-rewriter.c | 69 ++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 66 insertions(+), 7 deletions(-)

Comments

Jason Wang June 4, 2018, 5:51 a.m. UTC | #1
On 2018年06月03日 13:05, Zhang Chen wrote:
> After a net connection is closed, we didn't clear its releated resources
> in connection_track_table, which will lead to memory leak.
>
> Let't track the state of net connection, if it is closed, its related
> resources will be cleared up.
>
> Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
> Signed-off-by: Zhang Chen <zhangckid@gmail.com>
> ---
>   net/colo.h            |  4 +++
>   net/filter-rewriter.c | 69 ++++++++++++++++++++++++++++++++++++++-----
>   2 files changed, 66 insertions(+), 7 deletions(-)
>
> diff --git a/net/colo.h b/net/colo.h
> index da6c36dcf7..cd118510c5 100644
> --- a/net/colo.h
> +++ b/net/colo.h
> @@ -18,6 +18,7 @@
>   #include "slirp/slirp.h"
>   #include "qemu/jhash.h"
>   #include "qemu/timer.h"
> +#include "slirp/tcp.h"
>   
>   #define HASHTABLE_MAX_SIZE 16384
>   
> @@ -86,6 +87,9 @@ typedef struct Connection {
>        * run once in independent tcp connection
>        */
>       int syn_flag;
> +
> +    int tcp_state; /* TCP FSM state */
> +    tcp_seq fin_ack_seq; /* the seq of 'fin=1,ack=1' */

So the question is, the state machine is not complete. I suspect there 
will be corner cases that will be left because of lacking sufficient 
states. LAST_ACK happens only for passive close. How about active close?

So I think we need either maintain a full state machine or not instead 
of a partial one. We don't want endless bugs.

Thanks

>   } Connection;
>   
>   uint32_t connection_key_hash(const void *opaque);
> diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c
> index 62dad2d773..0909a9a8af 100644
> --- a/net/filter-rewriter.c
> +++ b/net/filter-rewriter.c
> @@ -59,9 +59,9 @@ static int is_tcp_packet(Packet *pkt)
>   }
>   
>   /* handle tcp packet from primary guest */
> -static int handle_primary_tcp_pkt(NetFilterState *nf,
> +static int handle_primary_tcp_pkt(RewriterState *rf,
>                                     Connection *conn,
> -                                  Packet *pkt)
> +                                  Packet *pkt, ConnectionKey *key)
>   {
>       struct tcphdr *tcp_pkt;
>   
> @@ -99,15 +99,44 @@ static int handle_primary_tcp_pkt(NetFilterState *nf,
>               net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_hdr_len,
>                                      pkt->size - pkt->vnet_hdr_len);
>           }
> +        /*
> +         * Case 1:
> +         * The *server* side of this connect is VM, *client* tries to close
> +         * the connection.
> +         *
> +         * We got 'ack=1' packets from client side, it acks 'fin=1, ack=1'
> +         * packet from server side. From this point, we can ensure that there
> +         * will be no packets in the connection, except that, some errors
> +         * happen between the path of 'filter object' and vNIC, if this rare
> +         * case really happen, we can still create a new connection,
> +         * So it is safe to remove the connection from connection_track_table.
> +         *
> +         */
> +        if ((conn->tcp_state == TCPS_LAST_ACK) &&
> +            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
> +            g_hash_table_remove(rf->connection_track_table, key);
> +        }
> +    }
> +    /*
> +     * Case 2:
> +     * The *server* side of this connect is VM, *server* tries to close
> +     * the connection.
> +     *
> +     * We got 'fin=1, ack=1' packet from client side, we need to
> +     * record the seq of 'fin=1, ack=1' packet.
> +     */
> +    if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
> +        conn->fin_ack_seq = htonl(tcp_pkt->th_seq);
> +        conn->tcp_state = TCPS_LAST_ACK;
>       }
>   
>       return 0;
>   }
>   
>   /* handle tcp packet from secondary guest */
> -static int handle_secondary_tcp_pkt(NetFilterState *nf,
> +static int handle_secondary_tcp_pkt(RewriterState *rf,
>                                       Connection *conn,
> -                                    Packet *pkt)
> +                                    Packet *pkt, ConnectionKey *key)
>   {
>       struct tcphdr *tcp_pkt;
>   
> @@ -139,8 +168,34 @@ static int handle_secondary_tcp_pkt(NetFilterState *nf,
>               net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_hdr_len,
>                                      pkt->size - pkt->vnet_hdr_len);
>           }
> +        /*
> +         * Case 2:
> +         * The *server* side of this connect is VM, *server* tries to close
> +         * the connection.
> +         *
> +         * We got 'ack=1' packets from server side, it acks 'fin=1, ack=1'
> +         * packet from client side. Like Case 1, there should be no packets
> +         * in the connection from now know, But the difference here is
> +         * if the packet is lost, We will get the resent 'fin=1,ack=1' packet.
> +         * TODO: Fix above case.
> +         */
> +        if ((conn->tcp_state == TCPS_LAST_ACK) &&
> +            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
> +            g_hash_table_remove(rf->connection_track_table, key);
> +        }
> +    }
> +    /*
> +     * Case 1:
> +     * The *server* side of this connect is VM, *client* tries to close
> +     * the connection.
> +     *
> +     * We got 'fin=1, ack=1' packet from server side, we need to
> +     * record the seq of 'fin=1, ack=1' packet.
> +     */
> +    if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
> +        conn->fin_ack_seq = ntohl(tcp_pkt->th_seq);
> +        conn->tcp_state = TCPS_LAST_ACK;
>       }
> -
>       return 0;
>   }
>   
> @@ -190,7 +245,7 @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
>   
>           if (sender == nf->netdev) {
>               /* NET_FILTER_DIRECTION_TX */
> -            if (!handle_primary_tcp_pkt(nf, conn, pkt)) {
> +            if (!handle_primary_tcp_pkt(s, conn, pkt, &key)) {
>                   qemu_net_queue_send(s->incoming_queue, sender, 0,
>                   (const uint8_t *)pkt->data, pkt->size, NULL);
>                   packet_destroy(pkt, NULL);
> @@ -203,7 +258,7 @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
>               }
>           } else {
>               /* NET_FILTER_DIRECTION_RX */
> -            if (!handle_secondary_tcp_pkt(nf, conn, pkt)) {
> +            if (!handle_secondary_tcp_pkt(s, conn, pkt, &key)) {
>                   qemu_net_queue_send(s->incoming_queue, sender, 0,
>                   (const uint8_t *)pkt->data, pkt->size, NULL);
>                   packet_destroy(pkt, NULL);
Zhang Chen June 10, 2018, 2:08 p.m. UTC | #2
On Mon, Jun 4, 2018 at 1:51 PM, Jason Wang <jasowang@redhat.com> wrote:

>
>
> On 2018年06月03日 13:05, Zhang Chen wrote:
>
>> After a net connection is closed, we didn't clear its releated resources
>> in connection_track_table, which will lead to memory leak.
>>
>> Let't track the state of net connection, if it is closed, its related
>> resources will be cleared up.
>>
>> Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
>> Signed-off-by: Zhang Chen <zhangckid@gmail.com>
>> ---
>>   net/colo.h            |  4 +++
>>   net/filter-rewriter.c | 69 ++++++++++++++++++++++++++++++++++++++-----
>>   2 files changed, 66 insertions(+), 7 deletions(-)
>>
>> diff --git a/net/colo.h b/net/colo.h
>> index da6c36dcf7..cd118510c5 100644
>> --- a/net/colo.h
>> +++ b/net/colo.h
>> @@ -18,6 +18,7 @@
>>   #include "slirp/slirp.h"
>>   #include "qemu/jhash.h"
>>   #include "qemu/timer.h"
>> +#include "slirp/tcp.h"
>>     #define HASHTABLE_MAX_SIZE 16384
>>   @@ -86,6 +87,9 @@ typedef struct Connection {
>>        * run once in independent tcp connection
>>        */
>>       int syn_flag;
>> +
>> +    int tcp_state; /* TCP FSM state */
>> +    tcp_seq fin_ack_seq; /* the seq of 'fin=1,ack=1' */
>>
>
> So the question is, the state machine is not complete. I suspect there
> will be corner cases that will be left because of lacking sufficient
> states. LAST_ACK happens only for passive close. How about active close?
>
> So I think we need either maintain a full state machine or not instead of
> a partial one. We don't want endless bugs.
>

OK, I got it.
I will add a full state machine here in next version.

Thanks
Zhang Chen


>
> Thanks
>
>
>   } Connection;
>>     uint32_t connection_key_hash(const void *opaque);
>> diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c
>> index 62dad2d773..0909a9a8af 100644
>> --- a/net/filter-rewriter.c
>> +++ b/net/filter-rewriter.c
>> @@ -59,9 +59,9 @@ static int is_tcp_packet(Packet *pkt)
>>   }
>>     /* handle tcp packet from primary guest */
>> -static int handle_primary_tcp_pkt(NetFilterState *nf,
>> +static int handle_primary_tcp_pkt(RewriterState *rf,
>>                                     Connection *conn,
>> -                                  Packet *pkt)
>> +                                  Packet *pkt, ConnectionKey *key)
>>   {
>>       struct tcphdr *tcp_pkt;
>>   @@ -99,15 +99,44 @@ static int handle_primary_tcp_pkt(NetFilterState
>> *nf,
>>               net_checksum_calculate((uint8_t *)pkt->data +
>> pkt->vnet_hdr_len,
>>                                      pkt->size - pkt->vnet_hdr_len);
>>           }
>> +        /*
>> +         * Case 1:
>> +         * The *server* side of this connect is VM, *client* tries to
>> close
>> +         * the connection.
>> +         *
>> +         * We got 'ack=1' packets from client side, it acks 'fin=1,
>> ack=1'
>> +         * packet from server side. From this point, we can ensure that
>> there
>> +         * will be no packets in the connection, except that, some errors
>> +         * happen between the path of 'filter object' and vNIC, if this
>> rare
>> +         * case really happen, we can still create a new connection,
>> +         * So it is safe to remove the connection from
>> connection_track_table.
>> +         *
>> +         */
>> +        if ((conn->tcp_state == TCPS_LAST_ACK) &&
>> +            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
>> +            g_hash_table_remove(rf->connection_track_table, key);
>> +        }
>> +    }
>> +    /*
>> +     * Case 2:
>> +     * The *server* side of this connect is VM, *server* tries to close
>> +     * the connection.
>> +     *
>> +     * We got 'fin=1, ack=1' packet from client side, we need to
>> +     * record the seq of 'fin=1, ack=1' packet.
>> +     */
>> +    if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
>> +        conn->fin_ack_seq = htonl(tcp_pkt->th_seq);
>> +        conn->tcp_state = TCPS_LAST_ACK;
>>       }
>>         return 0;
>>   }
>>     /* handle tcp packet from secondary guest */
>> -static int handle_secondary_tcp_pkt(NetFilterState *nf,
>> +static int handle_secondary_tcp_pkt(RewriterState *rf,
>>                                       Connection *conn,
>> -                                    Packet *pkt)
>> +                                    Packet *pkt, ConnectionKey *key)
>>   {
>>       struct tcphdr *tcp_pkt;
>>   @@ -139,8 +168,34 @@ static int handle_secondary_tcp_pkt(NetFilterState
>> *nf,
>>               net_checksum_calculate((uint8_t *)pkt->data +
>> pkt->vnet_hdr_len,
>>                                      pkt->size - pkt->vnet_hdr_len);
>>           }
>> +        /*
>> +         * Case 2:
>> +         * The *server* side of this connect is VM, *server* tries to
>> close
>> +         * the connection.
>> +         *
>> +         * We got 'ack=1' packets from server side, it acks 'fin=1,
>> ack=1'
>> +         * packet from client side. Like Case 1, there should be no
>> packets
>> +         * in the connection from now know, But the difference here is
>> +         * if the packet is lost, We will get the resent 'fin=1,ack=1'
>> packet.
>> +         * TODO: Fix above case.
>> +         */
>> +        if ((conn->tcp_state == TCPS_LAST_ACK) &&
>> +            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
>> +            g_hash_table_remove(rf->connection_track_table, key);
>> +        }
>> +    }
>> +    /*
>> +     * Case 1:
>> +     * The *server* side of this connect is VM, *client* tries to close
>> +     * the connection.
>> +     *
>> +     * We got 'fin=1, ack=1' packet from server side, we need to
>> +     * record the seq of 'fin=1, ack=1' packet.
>> +     */
>> +    if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
>> +        conn->fin_ack_seq = ntohl(tcp_pkt->th_seq);
>> +        conn->tcp_state = TCPS_LAST_ACK;
>>       }
>> -
>>       return 0;
>>   }
>>   @@ -190,7 +245,7 @@ static ssize_t colo_rewriter_receive_iov(NetFilterState
>> *nf,
>>             if (sender == nf->netdev) {
>>               /* NET_FILTER_DIRECTION_TX */
>> -            if (!handle_primary_tcp_pkt(nf, conn, pkt)) {
>> +            if (!handle_primary_tcp_pkt(s, conn, pkt, &key)) {
>>                   qemu_net_queue_send(s->incoming_queue, sender, 0,
>>                   (const uint8_t *)pkt->data, pkt->size, NULL);
>>                   packet_destroy(pkt, NULL);
>> @@ -203,7 +258,7 @@ static ssize_t colo_rewriter_receive_iov(NetFilterState
>> *nf,
>>               }
>>           } else {
>>               /* NET_FILTER_DIRECTION_RX */
>> -            if (!handle_secondary_tcp_pkt(nf, conn, pkt)) {
>> +            if (!handle_secondary_tcp_pkt(s, conn, pkt, &key)) {
>>                   qemu_net_queue_send(s->incoming_queue, sender, 0,
>>                   (const uint8_t *)pkt->data, pkt->size, NULL);
>>                   packet_destroy(pkt, NULL);
>>
>
>
diff mbox

Patch

diff --git a/net/colo.h b/net/colo.h
index da6c36dcf7..cd118510c5 100644
--- a/net/colo.h
+++ b/net/colo.h
@@ -18,6 +18,7 @@ 
 #include "slirp/slirp.h"
 #include "qemu/jhash.h"
 #include "qemu/timer.h"
+#include "slirp/tcp.h"
 
 #define HASHTABLE_MAX_SIZE 16384
 
@@ -86,6 +87,9 @@  typedef struct Connection {
      * run once in independent tcp connection
      */
     int syn_flag;
+
+    int tcp_state; /* TCP FSM state */
+    tcp_seq fin_ack_seq; /* the seq of 'fin=1,ack=1' */
 } Connection;
 
 uint32_t connection_key_hash(const void *opaque);
diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c
index 62dad2d773..0909a9a8af 100644
--- a/net/filter-rewriter.c
+++ b/net/filter-rewriter.c
@@ -59,9 +59,9 @@  static int is_tcp_packet(Packet *pkt)
 }
 
 /* handle tcp packet from primary guest */
-static int handle_primary_tcp_pkt(NetFilterState *nf,
+static int handle_primary_tcp_pkt(RewriterState *rf,
                                   Connection *conn,
-                                  Packet *pkt)
+                                  Packet *pkt, ConnectionKey *key)
 {
     struct tcphdr *tcp_pkt;
 
@@ -99,15 +99,44 @@  static int handle_primary_tcp_pkt(NetFilterState *nf,
             net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_hdr_len,
                                    pkt->size - pkt->vnet_hdr_len);
         }
+        /*
+         * Case 1:
+         * The *server* side of this connect is VM, *client* tries to close
+         * the connection.
+         *
+         * We got 'ack=1' packets from client side, it acks 'fin=1, ack=1'
+         * packet from server side. From this point, we can ensure that there
+         * will be no packets in the connection, except that, some errors
+         * happen between the path of 'filter object' and vNIC, if this rare
+         * case really happen, we can still create a new connection,
+         * So it is safe to remove the connection from connection_track_table.
+         *
+         */
+        if ((conn->tcp_state == TCPS_LAST_ACK) &&
+            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
+            g_hash_table_remove(rf->connection_track_table, key);
+        }
+    }
+    /*
+     * Case 2:
+     * The *server* side of this connect is VM, *server* tries to close
+     * the connection.
+     *
+     * We got 'fin=1, ack=1' packet from client side, we need to
+     * record the seq of 'fin=1, ack=1' packet.
+     */
+    if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
+        conn->fin_ack_seq = htonl(tcp_pkt->th_seq);
+        conn->tcp_state = TCPS_LAST_ACK;
     }
 
     return 0;
 }
 
 /* handle tcp packet from secondary guest */
-static int handle_secondary_tcp_pkt(NetFilterState *nf,
+static int handle_secondary_tcp_pkt(RewriterState *rf,
                                     Connection *conn,
-                                    Packet *pkt)
+                                    Packet *pkt, ConnectionKey *key)
 {
     struct tcphdr *tcp_pkt;
 
@@ -139,8 +168,34 @@  static int handle_secondary_tcp_pkt(NetFilterState *nf,
             net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_hdr_len,
                                    pkt->size - pkt->vnet_hdr_len);
         }
+        /*
+         * Case 2:
+         * The *server* side of this connect is VM, *server* tries to close
+         * the connection.
+         *
+         * We got 'ack=1' packets from server side, it acks 'fin=1, ack=1'
+         * packet from client side. Like Case 1, there should be no packets
+         * in the connection from now know, But the difference here is
+         * if the packet is lost, We will get the resent 'fin=1,ack=1' packet.
+         * TODO: Fix above case.
+         */
+        if ((conn->tcp_state == TCPS_LAST_ACK) &&
+            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
+            g_hash_table_remove(rf->connection_track_table, key);
+        }
+    }
+    /*
+     * Case 1:
+     * The *server* side of this connect is VM, *client* tries to close
+     * the connection.
+     *
+     * We got 'fin=1, ack=1' packet from server side, we need to
+     * record the seq of 'fin=1, ack=1' packet.
+     */
+    if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
+        conn->fin_ack_seq = ntohl(tcp_pkt->th_seq);
+        conn->tcp_state = TCPS_LAST_ACK;
     }
-
     return 0;
 }
 
@@ -190,7 +245,7 @@  static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
 
         if (sender == nf->netdev) {
             /* NET_FILTER_DIRECTION_TX */
-            if (!handle_primary_tcp_pkt(nf, conn, pkt)) {
+            if (!handle_primary_tcp_pkt(s, conn, pkt, &key)) {
                 qemu_net_queue_send(s->incoming_queue, sender, 0,
                 (const uint8_t *)pkt->data, pkt->size, NULL);
                 packet_destroy(pkt, NULL);
@@ -203,7 +258,7 @@  static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
             }
         } else {
             /* NET_FILTER_DIRECTION_RX */
-            if (!handle_secondary_tcp_pkt(nf, conn, pkt)) {
+            if (!handle_secondary_tcp_pkt(s, conn, pkt, &key)) {
                 qemu_net_queue_send(s->incoming_queue, sender, 0,
                 (const uint8_t *)pkt->data, pkt->size, NULL);
                 packet_destroy(pkt, NULL);