onnode: add "any" nodespec to select any node with running CTDB.

In testing and other situations (e.g. eventscripts) it is necessary to
select a node where a ctdb command can be run.  The whole idea here is
to avoid nodes where ctdbd is not running and where most ctdb commands
would fail.  This implements a standard way of doing this involving a
recursive onnode command.

There is still a small window for a race, where the selected node is
suddenly shutdown, but this is unavoidable.

Signed-off-by: Martin Schwenke <martin@meltin.net>

diff --git a/tools/onnode b/tools/onnode
index e00cd2b1d917feedc4ca2f46214d90df3a0e4349..79623a4a3cbfbf3c4d70c50bcd5d01d840543182 100755 (executable)
--- a/tools/onnode
+++ b/tools/onnode
@@ -35,7 +35,7 @@ Usage: onnode [OPTION] ... <NODES> <COMMAND> ...
     -p          Run command in parallel on specified nodes.
     -q          Do not print node addresses (overrides -v).
     -v          Print node address even for a single node.
-  <NODES>       "all", "ok" (or "healthy"), "con" (or "connected"),
+  <NODES>       "all", "any", "ok" (or "healthy"), "con" (or "connected"),
                 "rm" (or "recmaster"), "lvs" (or "lvsmaster"),
                 "natgw" (or "natgwlist");
                 or a node number (0 base); or
@@ -112,7 +112,7 @@ parse_nodespec ()
            case "$i" in
                *-*) seq "${i%-*}" "${i#*-}" 2>/dev/null || invalid_nodespec ;;
                # Separate lines for readability.
-               all|ok|healthy|con|connected) echo "$i" ;;
+               all|any|ok|healthy|con|connected) echo "$i" ;;
                rm|recmaster|lvs|lvsmaster|natgw|natgwlist) echo "$i" ;;
                *)
                    [ $i -gt -1 ] 2>/dev/null || invalid_nodespec
@@ -198,6 +198,24 @@ get_node_with_property ()
     fi
 }
 
+get_any_available_node ()
+{
+    local all_nodes="$1"
+
+    # We do a recursive onnode to find which nodes are up and running.
+    local out=$($0 -pq all ctdb pnn 2>&1)
+    local line
+    while read line ; do 
+       local pnn="${line#PNN:}"
+       if [ "$pnn" != "$line" ] ; then
+           echo_nth "$pnn" $all_nodes
+           return 0
+       fi
+       # Else must be an error message from a down node.
+    done <<<"$out"
+    return 1
+}
+
 get_nodes ()
 {
     local all_nodes
@@ -217,6 +235,9 @@ get_nodes ()
            all)
                echo "${all_nodes//#DEAD/}"
                ;;
+           any)
+               get_any_available_node "$all_nodes" || exit 1
+               ;;
            ok|healthy) 
                get_nodes_with_status "$all_nodes" "healthy" || exit 1
                ;;