dlm: fix connection close handling
Closing a connection to a node can create problems if there are outstanding messages for that node. The problems include dlm_send spinning attempting to reconnect, or BUG from tcp_connect_to_sock() attempting to use a partially closed connection. To cleanly close a connection, we now first attempt to send any pending messages, cancel any remaining workqueue work, and flag the connection as closed to avoid reconnect attempts. Signed-off-by: Lars Marowsky-Bree <lmb@suse.de> Signed-off-by: Christine Caulfield <ccaulfie@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
This commit is contained in:
committed by
David Teigland
parent
b5711b8e5a
commit
063c4c9963
@@ -106,6 +106,7 @@ struct connection {
|
|||||||
#define CF_CONNECT_PENDING 3
|
#define CF_CONNECT_PENDING 3
|
||||||
#define CF_INIT_PENDING 4
|
#define CF_INIT_PENDING 4
|
||||||
#define CF_IS_OTHERCON 5
|
#define CF_IS_OTHERCON 5
|
||||||
|
#define CF_CLOSE 6
|
||||||
struct list_head writequeue; /* List of outgoing writequeue_entries */
|
struct list_head writequeue; /* List of outgoing writequeue_entries */
|
||||||
spinlock_t writequeue_lock;
|
spinlock_t writequeue_lock;
|
||||||
int (*rx_action) (struct connection *); /* What to do when active */
|
int (*rx_action) (struct connection *); /* What to do when active */
|
||||||
@@ -299,6 +300,8 @@ static void lowcomms_write_space(struct sock *sk)
|
|||||||
|
|
||||||
static inline void lowcomms_connect_sock(struct connection *con)
|
static inline void lowcomms_connect_sock(struct connection *con)
|
||||||
{
|
{
|
||||||
|
if (test_bit(CF_CLOSE, &con->flags))
|
||||||
|
return;
|
||||||
if (!test_and_set_bit(CF_CONNECT_PENDING, &con->flags))
|
if (!test_and_set_bit(CF_CONNECT_PENDING, &con->flags))
|
||||||
queue_work(send_workqueue, &con->swork);
|
queue_work(send_workqueue, &con->swork);
|
||||||
}
|
}
|
||||||
@@ -1368,6 +1371,13 @@ int dlm_lowcomms_close(int nodeid)
|
|||||||
log_print("closing connection to node %d", nodeid);
|
log_print("closing connection to node %d", nodeid);
|
||||||
con = nodeid2con(nodeid, 0);
|
con = nodeid2con(nodeid, 0);
|
||||||
if (con) {
|
if (con) {
|
||||||
|
clear_bit(CF_CONNECT_PENDING, &con->flags);
|
||||||
|
clear_bit(CF_WRITE_PENDING, &con->flags);
|
||||||
|
set_bit(CF_CLOSE, &con->flags);
|
||||||
|
if (cancel_work_sync(&con->swork))
|
||||||
|
log_print("canceled swork for node %d", nodeid);
|
||||||
|
if (cancel_work_sync(&con->rwork))
|
||||||
|
log_print("canceled rwork for node %d", nodeid);
|
||||||
clean_one_writequeue(con);
|
clean_one_writequeue(con);
|
||||||
close_connection(con, true);
|
close_connection(con, true);
|
||||||
}
|
}
|
||||||
@@ -1393,8 +1403,9 @@ static void process_send_sockets(struct work_struct *work)
|
|||||||
|
|
||||||
if (test_and_clear_bit(CF_CONNECT_PENDING, &con->flags)) {
|
if (test_and_clear_bit(CF_CONNECT_PENDING, &con->flags)) {
|
||||||
con->connect_action(con);
|
con->connect_action(con);
|
||||||
|
set_bit(CF_WRITE_PENDING, &con->flags);
|
||||||
}
|
}
|
||||||
clear_bit(CF_WRITE_PENDING, &con->flags);
|
if (test_and_clear_bit(CF_WRITE_PENDING, &con->flags))
|
||||||
send_to_sock(con);
|
send_to_sock(con);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
Reference in New Issue
Block a user