github-mirror/linux
1
0
Fork 0
mirror of https://github.com/torvalds/linux.git synced 2026-09-14 16:10:02 +02:00
Code Issues Actions Packages Projects Releases Wiki Activity Graph

net/rds: No shortcut out of RDS_CONN_ERROR

Browse Source 
RDS connections carry a state "rds_conn_path::cp_state"
and transitions from one state to another and are conditional
upon an expected state: "rds_conn_path_transition."

There is one exception to this conditionality, which is
"RDS_CONN_ERROR" that can be enforced by "rds_conn_path_drop"
regardless of what state the condition is currently in.

But as soon as a connection enters state "RDS_CONN_ERROR",
the connection handling code expects it to go through the
shutdown-path.

The RDS/TCP multipath changes added a shortcut out of
"RDS_CONN_ERROR" straight back to "RDS_CONN_CONNECTING"
via "rds_tcp_accept_one_path" (e.g. after "rds_tcp_state_change").

A subsequent "rds_tcp_reset_callbacks" can then transition
the state to "RDS_CONN_RESETTING" with a shutdown-worker queued.

That'll trip up "rds_conn_init_shutdown", which was
never adjusted to handle "RDS_CONN_RESETTING" and subsequently
drops the connection with the dreaded "DR_INV_CONN_STATE",
which leaves "RDS_SHUTDOWN_WORK_QUEUED" on forever.

So we do two things here:

a) Don't shortcut "RDS_CONN_ERROR", but take the longer
   path through the shutdown code.

b) Add "RDS_CONN_RESETTING" to the expected states in
  "rds_conn_init_shutdown" so that we won't error out
  and get stuck, if we ever hit weird state transitions
  like this again."

Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com>
Signed-off-by: Allison Henderson <allison.henderson@oracle.com>
Link: https://patch.msgid.link/20260122055213.83608-2-achender@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This commit is contained in:
Gerd Rausch 2026-01-21 22:52:12 -07:00 committed by Jakub Kicinski
parent 6a9e8b60f0
commit ad22d24be6
2 changed files with 2 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
net/rds/connection.c
View File

@ -395,6 +395,8 @@ void rds_conn_shutdown(struct rds_conn_path *cp)
if (!rds_conn_path_transition(cp, RDS_CONN_UP, if (!rds_conn_path_transition(cp, RDS_CONN_UP,
RDS_CONN_DISCONNECTING) && RDS_CONN_DISCONNECTING) &&
!rds_conn_path_transition(cp, RDS_CONN_ERROR, !rds_conn_path_transition(cp, RDS_CONN_ERROR,
RDS_CONN_DISCONNECTING) &&
!rds_conn_path_transition(cp, RDS_CONN_RESETTING,
RDS_CONN_DISCONNECTING)) { RDS_CONN_DISCONNECTING)) {
rds_conn_path_error(cp, rds_conn_path_error(cp,
"shutdown called in state %d\n", "shutdown called in state %d\n",

5
net/rds/tcp_listen.c
View File

@ -59,9 +59,6 @@ void rds_tcp_keepalive(struct socket *sock)
* socket and force a reconneect from smaller -> larger ip addr. The reason * socket and force a reconneect from smaller -> larger ip addr. The reason
* we special case cp_index 0 is to allow the rds probe ping itself to itself * we special case cp_index 0 is to allow the rds probe ping itself to itself
* get through efficiently. * get through efficiently.
* Since reconnects are only initiated from the node with the numerically
* smaller ip address, we recycle conns in RDS_CONN_ERROR on the passive side
* by moving them to CONNECTING in this function.
*/ */
static static
struct rds_tcp_connection *rds_tcp_accept_one_path(struct rds_connection *conn) struct rds_tcp_connection *rds_tcp_accept_one_path(struct rds_connection *conn)
@ -86,8 +83,6 @@ struct rds_tcp_connection *rds_tcp_accept_one_path(struct rds_connection *conn)
struct rds_conn_path *cp = &conn->c_path[i]; struct rds_conn_path *cp = &conn->c_path[i];
if (rds_conn_path_transition(cp, RDS_CONN_DOWN, if (rds_conn_path_transition(cp, RDS_CONN_DOWN,
RDS_CONN_CONNECTING) ||
rds_conn_path_transition(cp, RDS_CONN_ERROR,
RDS_CONN_CONNECTING)) { RDS_CONN_CONNECTING)) {
return cp->cp_transport_data; return cp->cp_transport_data;
} }