diff options
| author | Raghavendra G <raghavendra@gluster.com> | 2012-10-08 14:16:21 +0530 |
|---|---|---|
| committer | Anand Avati <avati@redhat.com> | 2012-10-11 16:35:52 -0700 |
| commit | 92754cd1d12cdda18b27911f3e50eabd907edce6 (patch) | |
| tree | 5b354de1f73658d2fd6e3d865fb0081dc372784a /rpc/rpc-lib/src/rpc-clnt.c | |
| parent | 85866c6a9ef9b7401bb05f93fa9602f844d5a35b (diff) | |
client-handshake: synchronize config.remote_port setting b/w
rpc_clnt_reconnect and client_query_portmap_cbk
problem:
-------
Theoretically there is a possibility that we could complete
querying the remote brick's port number before rpc_transport_connect
can return. If rpc_clnt_reconnect happens to be the caller of
rpc_transport_connect and we've already got the remote brick's port
number by the time rpc_transport_connect returns, without synchronization,
rpc_clnt_connect resets config.remote_port to zero even before we have
attempted a connection with remote brick.
fix:
---
By making only poll thread do setting and resetting of
config.remote_port, we avoid the race-condition.
Change-Id: I51879ba1cac651a80ff5c9c070ec7fe1ceea9e05
BUG: 765051
Signed-off-by: Raghavendra G <raghavendra@gluster.com>
Reviewed-on: http://review.gluster.org/4044
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Anand Avati <avati@redhat.com>
Diffstat (limited to 'rpc/rpc-lib/src/rpc-clnt.c')
| -rw-r--r-- | rpc/rpc-lib/src/rpc-clnt.c | 21 |
1 files changed, 8 insertions, 13 deletions
diff --git a/rpc/rpc-lib/src/rpc-clnt.c b/rpc/rpc-lib/src/rpc-clnt.c index f68c5723071..fef604765c2 100644 --- a/rpc/rpc-lib/src/rpc-clnt.c +++ b/rpc/rpc-lib/src/rpc-clnt.c @@ -426,15 +426,6 @@ rpc_clnt_reconnect (void *trans_ptr) "attempting reconnect"); ret = rpc_transport_connect (trans, conn->config.remote_port); - /* Every time there is a disconnection, processes - should try to connect to 'glusterd' (ie, default - port) or whichever port given as 'option remote-port' - in volume file. */ - /* Below code makes sure the (re-)configured port lasts - for just one successful attempt */ - if (!ret) - conn->config.remote_port = 0; - conn->reconnect = gf_timer_call_after (clnt->ctx, tv, rpc_clnt_reconnect, @@ -923,6 +914,14 @@ rpc_clnt_notify (rpc_transport_t *trans, void *mydata, case RPC_TRANSPORT_CONNECT: { + /* Every time there is a disconnection, processes + should try to connect to 'glusterd' (ie, default + port) or whichever port given as 'option remote-port' + in volume file. */ + /* Below code makes sure the (re-)configured port lasts + for just one successful attempt */ + conn->config.remote_port = 0; + if (clnt->notifyfn) ret = clnt->notifyfn (clnt, clnt->mydata, RPC_CLNT_CONNECT, NULL); @@ -1479,10 +1478,6 @@ rpc_clnt_submit (struct rpc_clnt *rpc, rpc_clnt_prog_t *prog, if (conn->connected == 0) { ret = rpc_transport_connect (conn->trans, conn->config.remote_port); - /* Below code makes sure the (re-)configured port lasts - for just one successful connect attempt */ - if (!ret) - conn->config.remote_port = 0; } ret = rpc_transport_submit_request (rpc->conn.trans, |