[SCM] Samba Shared Repository - branch master updated

[Martin Schwenke]

The branch, master has been updated
       via  0f92de8 ctdb-daemon: Fix the usage for lock helper
       via  57310f8 ctdb-recoverd: If obtaining recovery lock fails, try again
      from  f1e281c ctdb-scripts: Fix the regular expresssion for parsing /proc/locks

http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master


- Log -----------------------------------------------------------------
commit 0f92de8463b71a2d7e9acdd27454be7859713436
Author: Amitay Isaacs <amitay at gmail.com>
Date:   Thu Sep 25 17:55:15 2014 +1000

    ctdb-daemon: Fix the usage for lock helper
    
    Signed-off-by: Amitay Isaacs <amitay at gmail.com>
    Reviewed-by: Martin Schwenke <martin at meltin.net>
    
    Autobuild-User(master): Martin Schwenke <martins at samba.org>
    Autobuild-Date(master): Thu Sep 25 17:16:31 CEST 2014 on sn-devel-104

commit 57310f80c9b8146a0978d912f73b0a64fde7697e
Author: Amitay Isaacs <amitay at gmail.com>
Date:   Thu Sep 25 17:17:04 2014 +1000

    ctdb-recoverd: If obtaining recovery lock fails, try again
    
    When ctdb daemon starts up, it considers itself the recovery master
    and tries to do first recovery.  However, it's possible that there is
    already a recovery master and the current node has not yet heard from it.
    So do not ban ourselves immediately if ctdb_recovery_lock() fails when
    doing first recovery.
    
    Signed-off-by: Amitay Isaacs <amitay at gmail.com>
    Reviewed-by: Martin Schwenke <martin at meltin.net>

-----------------------------------------------------------------------

Summary of changes:
 ctdb/server/ctdb_lock_helper.c |    4 ++--
 ctdb/server/ctdb_recoverd.c    |   18 ++++++++++++++++++
 2 files changed, 20 insertions(+), 2 deletions(-)


Changeset truncated at 500 lines:

diff --git a/ctdb/server/ctdb_lock_helper.c b/ctdb/server/ctdb_lock_helper.c
index 49c317e..2161a9a 100644
--- a/ctdb/server/ctdb_lock_helper.c
+++ b/ctdb/server/ctdb_lock_helper.c
@@ -36,9 +36,9 @@ static void send_result(int fd, char result)
 static void usage(void)
 {
 	fprintf(stderr, "\n");
-	fprintf(stderr, "Usage: %s <ctdbd-pid> <output-fd> RECORD <db-path> <db-key>\n",
+	fprintf(stderr, "Usage: %s <log-fd> <ctdbd-pid> <output-fd> RECORD <db-path> <db-key>\n",
 		progname);
-	fprintf(stderr, "       %s <ctdbd-pid> <output-fd> DB <db1-path> [<db2-path> ...]\n",
+	fprintf(stderr, "       %s <log-fd> <ctdbd-pid> <output-fd> DB <db1-path> [<db2-path> ...]\n",
 		progname);
 }
 
diff --git a/ctdb/server/ctdb_recoverd.c b/ctdb/server/ctdb_recoverd.c
index 14e6ea8..945b01c 100644
--- a/ctdb/server/ctdb_recoverd.c
+++ b/ctdb/server/ctdb_recoverd.c
@@ -1815,6 +1815,16 @@ static int do_recovery(struct ctdb_recoverd *rec,
 		DEBUG(DEBUG_ERR,("Taking out recovery lock from recovery daemon\n"));
 		start_time = timeval_current();
 		if (!ctdb_recovery_lock(ctdb, true)) {
+			if (ctdb->runstate == CTDB_RUNSTATE_FIRST_RECOVERY) {
+				/* If ctdb is trying first recovery, it's
+				 * possible that current node does not know yet
+				 * who the recmaster is.
+				 */
+				DEBUG(DEBUG_ERR, ("Unable to get recovery lock"
+						" - retrying recovery\n"));
+				return -1;
+			}
+
 			DEBUG(DEBUG_ERR,("Unable to get recovery lock - aborting recovery "
 					 "and ban ourself for %u seconds\n",
 					 ctdb->tunable.recovery_ban_period));
@@ -3593,6 +3603,14 @@ static void main_loop(struct ctdb_context *ctdb, struct ctdb_recoverd *rec,
 		return;
 	}
 
+	/* get runstate */
+	ret = ctdb_ctrl_get_runstate(ctdb, CONTROL_TIMEOUT(),
+				     CTDB_CURRENT_NODE, &ctdb->runstate);
+	if (ret != 0) {
+		DEBUG(DEBUG_ERR, ("Failed to get runstate - retrying\n"));
+		return;
+	}
+
 	/* get the current recovery lock file from the server */
 	if (update_recovery_lock_file(ctdb) != 0) {
 		DEBUG(DEBUG_ERR,("Failed to update the recovery lock file\n"));


-- 
Samba Shared Repository