mdadm: improve the dlm locking mechanism for clustered raid

[thirdparty/mdadm.git] / mdmon.h

diff --git a/mdmon.h b/mdmon.h
index e4904baa438b389e3837f2498e80ad909248d8bd..818367cbb14c95755d8384702abcc155d41d0e96 100644 (file)
--- a/mdmon.h
+++ b/mdmon.h
@@ -1,8 +1,8 @@
 /*
  * mdmon - monitor external metadata arrays
  *
- * Copyright (C) 2007-2008 Neil Brown <neilb@suse.de>
- * Copyright (C) 2007-2008 Intel Corporation
+ * Copyright (C) 2007-2009 Neil Brown <neilb@suse.de>
+ * Copyright (C) 2007-2009 Intel Corporation
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -18,29 +18,38 @@
  * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
  */
 
+extern const char Name[];
+
 enum array_state { clear, inactive, suspended, readonly, read_auto,
                   clean, active, write_pending, active_idle, bad_word};
 
 enum sync_action { idle, reshape, resync, recover, check, repair, bad_action };
 
-
 struct active_array {
        struct mdinfo info;
        struct supertype *container;
        struct active_array *next, *replaces;
+       int to_remove;
 
        int action_fd;
        int resync_start_fd;
        int metadata_fd; /* for monitoring rw/ro status */
+       int sync_completed_fd; /* for checkpoint notification events */
+       int safe_mode_delay_fd;
+       unsigned long long last_checkpoint; /* sync_completed fires for many
+                                            * reasons this field makes sure the
+                                            * kernel has made progress before
+                                            * moving the checkpoint.  It is
+                                            * cleared by the metadata handler
+                                            * when it determines recovery is
+                                            * terminated.
+                                            */
 
        enum array_state prev_state, curr_state, next_state;
        enum sync_action prev_action, curr_action, next_action;
 
        int check_degraded; /* flag set by mon, read by manage */
-
-       int devnum;
-
-       unsigned long long resync_start;
+       int check_reshape; /* flag set by mon, read by manage */
 };
 
 /*
@@ -63,17 +72,12 @@ extern struct active_array *discard_this;
 extern struct active_array *pending_discard;
 extern struct md_generic_cmd *active_cmd;
 
-
 void remove_pidfile(char *devname);
 void do_monitor(struct supertype *container);
 void do_manager(struct supertype *container);
-int make_control_sock(char *devname);
-int make_pidfile(char *devname, int o_excl);
-extern int socket_hup_requested;
 extern int sigterm;
 
 int read_dev_state(int fd);
-int get_resync_start(struct active_array *a);
 int is_container_member(struct mdstat_ent *mdstat, char *container);
 
 struct mdstat_ent *mdstat_read(int hold, int start);
@@ -85,10 +89,23 @@ extern int monitor_loop_cnt;
 /* helper routine to determine resync completion since MaxSector is a
  * moving target
  */
-static inline int is_resync_complete(struct active_array *a)
+static inline int is_resync_complete(struct mdinfo *array)
 {
-       if (a->resync_start >= a->info.component_size)
-               return 1;
-       return 0;
+       unsigned long long sync_size = 0;
+       int ncopies, l;
+       switch(array->array.level) {
+       case 1:
+       case 4:
+       case 5:
+       case 6:
+               sync_size = array->component_size;
+               break;
+       case 10:
+               l = array->array.layout;
+               ncopies = (l & 0xff) * ((l >> 8) & 0xff);
+               sync_size = array->component_size * array->array.raid_disks;
+               sync_size /= ncopies;
+               break;
+       }
+       return array->resync_start >= sync_size;
 }
-