ctdb-tests: Common code to wait for synchronization across cluster

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>

diff --git a/ctdb/tests/src/cluster_wait.c b/ctdb/tests/src/cluster_wait.c
new file mode 100644 (file)
index 0000000..ddc3e02
--- /dev/null
+++ b/ctdb/tests/src/cluster_wait.c
@@ -0,0 +1,331 @@
+/*
+   Cluster wide synchronization
+
+   Copyright (C) Amitay Isaacs  2015
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, see <http://www.gnu.org/licenses/>.
+*/
+
+#include "replace.h"
+#include "system/network.h"
+
+#include "lib/util/tevent_unix.h"
+
+#include "client/client.h"
+
+#include "tests/src/cluster_wait.h"
+
+#define MSG_ID_JOIN    (CTDB_SRVID_TEST_RANGE | 0x1)
+#define MSG_ID_SYNC    (CTDB_SRVID_TEST_RANGE | 0x2)
+
+/* Wait for all the clients to initialize */
+
+struct cluster_wait_state {
+       struct tevent_context *ev;
+       struct ctdb_client_context *client;
+       int num_nodes;
+       bool *ready;
+};
+
+static void cluster_wait_join_registered(struct tevent_req *subreq);
+static void cluster_wait_sync_registered(struct tevent_req *subreq);
+static void cluster_wait_join(struct tevent_req *subreq);
+static void cluster_wait_join_sent(struct tevent_req *subreq);
+static void cluster_wait_join_handler(uint64_t srvid, TDB_DATA data,
+                                     void *private_data);
+static void cluster_wait_sync_sent(struct tevent_req *subreq);
+static void cluster_wait_join_unregistered(struct tevent_req *subreq);
+static void cluster_wait_sync_handler(uint64_t srvid, TDB_DATA data,
+                                     void *private_data);
+static void cluster_wait_sync_unregistered(struct tevent_req *subreq);
+
+struct tevent_req *cluster_wait_send(TALLOC_CTX *mem_ctx,
+                                    struct tevent_context *ev,
+                                    struct ctdb_client_context *client,
+                                    int num_nodes)
+{
+       struct tevent_req *req, *subreq;
+       struct cluster_wait_state *state;
+
+       req = tevent_req_create(mem_ctx, &state, struct cluster_wait_state);
+       if (req == NULL) {
+               return NULL;
+       }
+
+       state->ev = ev;
+       state->client = client;
+       state->num_nodes = num_nodes;
+
+       if (ctdb_client_pnn(client) == 0) {
+               state->ready = talloc_zero_array(state, bool, num_nodes);
+               if (tevent_req_nomem(state->ready, req)) {
+                       return tevent_req_post(req, ev);
+               }
+
+               subreq = ctdb_client_set_message_handler_send(
+                                       state, ev, client, MSG_ID_JOIN,
+                                       cluster_wait_join_handler, req);
+               if (tevent_req_nomem(subreq, req)) {
+                       return tevent_req_post(req, ev);
+               }
+               tevent_req_set_callback(subreq, cluster_wait_join_registered,
+                                       req);
+       }
+
+       subreq = ctdb_client_set_message_handler_send(
+                                       state, ev, client, MSG_ID_SYNC,
+                                       cluster_wait_sync_handler, req);
+       if (tevent_req_nomem(subreq, req)) {
+               return tevent_req_post(req, ev);
+       }
+       tevent_req_set_callback(subreq, cluster_wait_sync_registered, req);
+
+       /* If cluster is not synchronized within 30 seconds, time out */
+       tevent_req_set_endtime(req, ev, tevent_timeval_current_ofs(30, 0));
+
+       return req;
+}
+
+static void cluster_wait_join_registered(struct tevent_req *subreq)
+{
+       struct tevent_req *req = tevent_req_callback_data(
+               subreq, struct tevent_req);
+       bool status;
+       int ret;
+
+       status = ctdb_client_set_message_handler_recv(subreq, &ret);
+       TALLOC_FREE(subreq);
+       if (! status) {
+               tevent_req_error(req, ret);
+               return;
+       }
+
+       printf("Waiting for cluster\n");
+       fflush(stdout);
+}
+
+static void cluster_wait_sync_registered(struct tevent_req *subreq)
+{
+       struct tevent_req *req = tevent_req_callback_data(
+               subreq, struct tevent_req);
+       struct cluster_wait_state *state = tevent_req_data(
+               req, struct cluster_wait_state);
+       bool status;
+       int ret;
+
+       status = ctdb_client_set_message_handler_recv(subreq, &ret);
+       TALLOC_FREE(subreq);
+       if (! status) {
+               tevent_req_error(req, ret);
+               return;
+       }
+
+       subreq = tevent_wakeup_send(state, state->ev,
+                                   tevent_timeval_current_ofs(1, 0));
+       if (tevent_req_nomem(subreq, req)) {
+               return;
+       }
+       tevent_req_set_callback(subreq, cluster_wait_join, req);
+}
+
+static void cluster_wait_join(struct tevent_req *subreq)
+{
+       struct tevent_req *req = tevent_req_callback_data(
+               subreq, struct tevent_req);
+       struct cluster_wait_state *state = tevent_req_data(
+               req, struct cluster_wait_state);
+       struct ctdb_req_message msg;
+       uint32_t pnn;
+       bool status;
+
+       status = tevent_wakeup_recv(subreq);
+       TALLOC_FREE(subreq);
+       if (! status) {
+               tevent_req_error(req, EIO);
+               return;
+       }
+
+       pnn = ctdb_client_pnn(state->client);
+
+       msg.srvid = MSG_ID_JOIN;
+       msg.data.data.dsize = sizeof(pnn);
+       msg.data.data.dptr = (uint8_t *)&pnn;
+
+       subreq = ctdb_client_message_send(state, state->ev, state->client,
+                                         0, &msg);
+       if (tevent_req_nomem(subreq, req)) {
+               return;
+       }
+       tevent_req_set_callback(subreq, cluster_wait_join_sent, req);
+}
+
+static void cluster_wait_join_sent(struct tevent_req *subreq)
+{
+       struct tevent_req *req = tevent_req_callback_data(
+               subreq, struct tevent_req);
+       struct cluster_wait_state *state = tevent_req_data(
+               req, struct cluster_wait_state);
+       bool status;
+       int ret;
+
+       status = ctdb_client_message_recv(subreq, &ret);
+       TALLOC_FREE(subreq);
+       if (! status) {
+               tevent_req_error(req, ret);
+               return;
+       }
+
+       subreq = tevent_wakeup_send(state, state->ev,
+                                   tevent_timeval_current_ofs(1, 0));
+       if (tevent_req_nomem(subreq, req)) {
+               return;
+       }
+       tevent_req_set_callback(subreq, cluster_wait_join, req);
+}
+
+static void cluster_wait_join_handler(uint64_t srvid, TDB_DATA data,
+                                     void *private_data)
+{
+       struct tevent_req *req = talloc_get_type_abort(
+               private_data, struct tevent_req);
+       struct cluster_wait_state *state = tevent_req_data(
+               req, struct cluster_wait_state);
+       struct ctdb_req_message msg;
+       struct tevent_req *subreq;
+       uint32_t pnn;
+       int i;
+
+       if (srvid != MSG_ID_JOIN) {
+               return;
+       }
+
+       if (data.dsize != sizeof(uint32_t)) {
+               return;
+       }
+
+       pnn = *(uint32_t *)data.dptr;
+
+       if (pnn > state->num_nodes) {
+               return;
+       }
+
+       state->ready[pnn] = true;
+
+       for (i=0; i<state->num_nodes; i++) {
+               if (! state->ready[i]) {
+                       return;
+               }
+       }
+
+       msg.srvid = MSG_ID_SYNC;
+       msg.data.data = tdb_null;
+
+       subreq = ctdb_client_message_send(state, state->ev, state->client,
+                                         CTDB_BROADCAST_ALL, &msg);
+       if (tevent_req_nomem(subreq, req)) {
+               return;
+       }
+       tevent_req_set_callback(subreq, cluster_wait_sync_sent, req);
+}
+
+static void cluster_wait_sync_sent(struct tevent_req *subreq)
+{
+       struct tevent_req *req = tevent_req_callback_data(
+               subreq, struct tevent_req);
+       struct cluster_wait_state *state = tevent_req_data(
+               req, struct cluster_wait_state);
+       bool status;
+       int ret;
+
+       status = ctdb_client_message_recv(subreq, &ret);
+       TALLOC_FREE(subreq);
+       if (! status) {
+               tevent_req_error(req, ret);
+               return;
+       }
+
+       subreq = ctdb_client_remove_message_handler_send(
+                                       state, state->ev, state->client,
+                                       MSG_ID_JOIN, req);
+       if (tevent_req_nomem(subreq, req)) {
+               return;
+       }
+       tevent_req_set_callback(subreq, cluster_wait_join_unregistered, req);
+}
+
+static void cluster_wait_join_unregistered(struct tevent_req *subreq)
+{
+       struct tevent_req *req = tevent_req_callback_data(
+               subreq, struct tevent_req);
+       bool status;
+       int ret;
+
+       status = ctdb_client_remove_message_handler_recv(subreq, &ret);
+       if (! status) {
+               tevent_req_error(req, ret);
+               return;
+       }
+}
+
+static void cluster_wait_sync_handler(uint64_t srvid, TDB_DATA data,
+                                     void *private_data)
+{
+       struct tevent_req *req = talloc_get_type_abort(
+               private_data, struct tevent_req);
+       struct cluster_wait_state *state = tevent_req_data(
+               req, struct cluster_wait_state);
+       struct tevent_req *subreq;
+
+       if (srvid != MSG_ID_SYNC) {
+               return;
+       }
+
+       subreq = ctdb_client_remove_message_handler_send(
+                                       state, state->ev, state->client,
+                                       MSG_ID_SYNC, req);
+       if (tevent_req_nomem(subreq, req)) {
+               return;
+       }
+       tevent_req_set_callback(subreq, cluster_wait_sync_unregistered, req);
+}
+
+static void cluster_wait_sync_unregistered(struct tevent_req *subreq)
+{
+       struct tevent_req *req = tevent_req_callback_data(
+               subreq, struct tevent_req);
+       bool status;
+       int ret;
+
+       status = ctdb_client_remove_message_handler_recv(subreq, &ret);
+       TALLOC_FREE(subreq);
+       if (! status) {
+               tevent_req_error(req, ret);
+               return;
+       }
+
+       tevent_req_done(req);
+}
+
+bool cluster_wait_recv(struct tevent_req *req, int *perr)
+{
+       int err;
+
+       if (tevent_req_is_unix_error(req, &err)) {
+               if (perr != NULL) {
+                       *perr = err;
+               }
+               return false;
+       }
+       return true;
+}

diff --git a/ctdb/tests/src/cluster_wait.h b/ctdb/tests/src/cluster_wait.h
new file mode 100644 (file)
index 0000000..20aa6e9
--- /dev/null
+++ b/ctdb/tests/src/cluster_wait.h
@@ -0,0 +1,30 @@
+/*
+   Cluster wide synchronization
+
+   Copyright (C) Amitay Isaacs  2015
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, see <http://www.gnu.org/licenses/>.
+*/
+
+#ifndef __CLUSTER_WAIT_H__
+#define __CLUSTER_WAIT_H__
+
+struct tevent_req *cluster_wait_send(TALLOC_CTX *mem_ctx,
+                                    struct tevent_context *ev,
+                                    struct ctdb_client_context *client,
+                                    int num_nodes);
+
+bool cluster_wait_recv(struct tevent_req *req, int *perr);
+
+#endif /* __CLUSTER_WAIT_H__ */

diff --git a/ctdb/wscript b/ctdb/wscript
index 96158c0f7d36f6908c03bc895a42b065446b8a1c..bdce3623194cad25769f85cc828a5caf5e212517 100755 (executable)
--- a/ctdb/wscript
+++ b/ctdb/wscript
@@ -703,6 +703,12 @@ def build(bld):
                          deps='ctdb-client ctdb-common ctdb-util',
                          install_path='${CTDB_TEST_LIBDIR}')
 
+    bld.SAMBA_SUBSYSTEM('ctdb-tests-common',
+                        source=bld.SUBDIR('tests/src',
+                                          'cluster_wait.c'),
+                        includes='include',
+                        deps='replace popt talloc tevent tdb')
+
     bld.SAMBA_BINARY('ctdb_takeover_tests',
                      source='tests/src/ctdb_takeover_tests.c',
                      deps='''replace popt tdb tevent talloc ctdb-system