#include <unistd.h>
#include <sys/stat.h>
#include <libdevmapper.h>
-#include <wait.h>
+#include <sys/wait.h>
#include <sys/mman.h>
#include <sys/types.h>
#include <fcntl.h>
#include <limits.h>
#include <linux/oom.h>
#include <libudev.h>
+#include <urcu.h>
#ifdef USE_SYSTEMD
#include <systemd/sd-daemon.h>
#endif
#include <semaphore.h>
-#include <mpath_cmd.h>
-#include <mpath_persist.h>
#include <time.h>
+#include <stdbool.h>
+
+/*
+ * libmultipath
+ */
+#include "time-util.h"
/*
* libcheckers
*/
-#include <checkers.h>
+#include "checkers.h"
#ifdef USE_SYSTEMD
static int use_watchdog;
#endif
+int uxsock_timeout;
+
/*
* libmultipath
*/
-#include <parser.h>
-#include <vector.h>
-#include <memory.h>
-#include <config.h>
-#include <util.h>
-#include <hwtable.h>
-#include <defaults.h>
-#include <structs.h>
-#include <blacklist.h>
-#include <structs_vec.h>
-#include <dmparser.h>
-#include <devmapper.h>
-#include <sysfs.h>
-#include <dict.h>
-#include <discovery.h>
-#include <debug.h>
-#include <propsel.h>
-#include <uevent.h>
-#include <switchgroup.h>
-#include <print.h>
-#include <configure.h>
-#include <prio.h>
-#include <wwids.h>
-#include <pgpolicies.h>
-#include <uevent.h>
-#include <log.h>
+#include "parser.h"
+#include "vector.h"
+#include "memory.h"
+#include "config.h"
+#include "util.h"
+#include "hwtable.h"
+#include "defaults.h"
+#include "structs.h"
+#include "blacklist.h"
+#include "structs_vec.h"
+#include "dmparser.h"
+#include "devmapper.h"
+#include "sysfs.h"
+#include "dict.h"
+#include "discovery.h"
+#include "debug.h"
+#include "propsel.h"
+#include "uevent.h"
+#include "switchgroup.h"
+#include "print.h"
+#include "configure.h"
+#include "prio.h"
+#include "wwids.h"
+#include "pgpolicies.h"
+#include "uevent.h"
+#include "log.h"
+
+#include "mpath_cmd.h"
+#include "mpath_persist.h"
+
#include "prioritizers/alua_rtpg.h"
#include "main.h"
#include "lock.h"
#include "waiter.h"
#include "wwids.h"
+#include "../third-party/valgrind/drd.h"
#define FILE_NAME_SIZE 256
#define CMDSIZE 160
unsigned int mpath_mx_alloc_len;
int logsink;
-enum daemon_status running_state;
+int verbosity;
+int bindings_read_only;
+int ignore_new_devs;
+enum daemon_status running_state = DAEMON_INIT;
pid_t daemon_pid;
+pthread_mutex_t config_lock = PTHREAD_MUTEX_INITIALIZER;
+pthread_cond_t config_cond;
-static sem_t exit_sem;
/*
* global copy of vecs for use in sig handlers
*/
struct udev * udev;
+struct config *multipath_conf;
+
+/* Local variables */
+static volatile sig_atomic_t exit_sig;
+static volatile sig_atomic_t reconfig_sig;
+static volatile sig_atomic_t log_reset_sig;
+
+const char *
+daemon_status(void)
+{
+ switch (running_state) {
+ case DAEMON_INIT:
+ return "init";
+ case DAEMON_START:
+ return "startup";
+ case DAEMON_CONFIGURE:
+ return "configure";
+ case DAEMON_IDLE:
+ return "idle";
+ case DAEMON_RUNNING:
+ return "running";
+ case DAEMON_SHUTDOWN:
+ return "shutdown";
+ }
+ return NULL;
+}
+
+/*
+ * I love you too, systemd ...
+ */
+const char *
+sd_notify_status(void)
+{
+ switch (running_state) {
+ case DAEMON_INIT:
+ return "STATUS=init";
+ case DAEMON_START:
+ return "STATUS=startup";
+ case DAEMON_CONFIGURE:
+ return "STATUS=configure";
+ case DAEMON_IDLE:
+ return "STATUS=idle";
+ case DAEMON_RUNNING:
+ return "STATUS=running";
+ case DAEMON_SHUTDOWN:
+ return "STATUS=shutdown";
+ }
+ return NULL;
+}
+
+static void config_cleanup(void *arg)
+{
+ pthread_mutex_unlock(&config_lock);
+}
+
+void post_config_state(enum daemon_status state)
+{
+ pthread_mutex_lock(&config_lock);
+ if (state != running_state) {
+ running_state = state;
+ pthread_cond_broadcast(&config_cond);
+#ifdef USE_SYSTEMD
+ sd_notify(0, sd_notify_status());
+#endif
+ }
+ pthread_mutex_unlock(&config_lock);
+}
+
+int set_config_state(enum daemon_status state)
+{
+ int rc = 0;
+
+ pthread_cleanup_push(config_cleanup, NULL);
+ pthread_mutex_lock(&config_lock);
+ if (running_state != state) {
+ if (running_state != DAEMON_IDLE) {
+ struct timespec ts;
+
+ clock_gettime(CLOCK_MONOTONIC, &ts);
+ ts.tv_sec += 1;
+ rc = pthread_cond_timedwait(&config_cond,
+ &config_lock, &ts);
+ }
+ if (!rc) {
+ running_state = state;
+ pthread_cond_broadcast(&config_cond);
+#ifdef USE_SYSTEMD
+ sd_notify(0, sd_notify_status());
+#endif
+ }
+ }
+ pthread_cleanup_pop(1);
+ return rc;
+}
+
+struct config *get_multipath_config(void)
+{
+ rcu_read_lock();
+ return rcu_dereference(multipath_conf);
+}
+
+void put_multipath_config(struct config *conf)
+{
+ rcu_read_unlock();
+}
+
static int
need_switch_pathgroup (struct multipath * mpp, int refresh)
{
struct pathgroup * pgp;
struct path * pp;
unsigned int i, j;
+ struct config *conf;
if (!mpp || mpp->pgfailback == -FAILBACK_MANUAL)
return 0;
/*
* Refresh path priority values
*/
- if (refresh)
- vector_foreach_slot (mpp->pg, pgp, i)
- vector_foreach_slot (pgp->paths, pp, j)
- pathinfo(pp, conf->hwtable, DI_PRIO);
+ if (refresh) {
+ vector_foreach_slot (mpp->pg, pgp, i) {
+ vector_foreach_slot (pgp->paths, pp, j) {
+ conf = get_multipath_config();
+ pathinfo(pp, conf, DI_PRIO);
+ put_multipath_config(conf);
+ }
+ }
+ }
+
+ if (!mpp->pg || VECTOR_SIZE(mpp->paths) == 0)
+ return 0;
mpp->bestpg = select_path_group(mpp);
{
struct multipath * ompp;
vector ompv = vecs->mpvec;
- unsigned int i;
+ unsigned int i, reassign_maps;
+ struct config *conf;
+ conf = get_multipath_config();
+ reassign_maps = conf->reassign_maps;
+ put_multipath_config(conf);
vector_foreach_slot (ompv, ompp, i) {
condlog(3, "%s: coalesce map", ompp->alias);
if (!find_mp_by_wwid(nmpv, ompp->wwid)) {
dm_lib_release();
condlog(2, "%s devmap removed", ompp->alias);
}
- } else if (conf->reassign_maps) {
+ } else if (reassign_maps) {
condlog(3, "%s: Reassign existing device-mapper"
" devices", ompp->alias);
dm_reassign(ompp->alias);
vector_foreach_slot (mpp->pg, pgp, i){
vector_foreach_slot (pgp->paths, pp, j){
- if (pp->state == PATH_UNCHECKED ||
+ if (pp->state == PATH_UNCHECKED ||
pp->state == PATH_WILD ||
pp->state == PATH_DELAYED)
continue;
retry:
condlog(4, "%s: updating new map", mpp->alias);
- if (adopt_paths(vecs->pathvec, mpp, 1)) {
+ if (adopt_paths(vecs->pathvec, mpp)) {
condlog(0, "%s: failed to adopt paths for new map update",
mpp->alias);
retries = -1;
retries = -1;
goto fail;
}
- if (domap(mpp, params) <= 0 && retries-- > 0) {
+ if (domap(mpp, params, 1) <= 0 && retries-- > 0) {
condlog(0, "%s: map_udate sleep", mpp->alias);
sleep(1);
goto retry;
return 1;
}
}
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
rc = ev_add_map(uev->kernel, alias, vecs);
+ lock_cleanup_pop(vecs->lock);
FREE(alias);
return rc;
}
char * refwwid;
struct multipath * mpp;
int map_present;
- int r = 1;
+ int r = 1, delayed_reconfig, reassign_maps;
+ struct config *conf;
map_present = dm_map_present(alias);
if (mpp) {
if (mpp->wait_for_udev > 1) {
+ condlog(2, "%s: performing delayed actions",
+ mpp->alias);
if (update_map(mpp, vecs))
- /* setup multipathd removed the map */
+ /* setup multipathd removed the map */
return 1;
}
+ conf = get_multipath_config();
+ delayed_reconfig = conf->delayed_reconfig;
+ reassign_maps = conf->reassign_maps;
+ put_multipath_config(conf);
if (mpp->wait_for_udev) {
mpp->wait_for_udev = 0;
- if (conf->delayed_reconfig &&
+ if (delayed_reconfig &&
!need_to_delay_reconfig(vecs)) {
condlog(2, "reconfigure (delayed)");
- reconfigure(vecs);
+ set_config_state(DAEMON_CONFIGURE);
return 0;
}
}
* if we create a multipath mapped device as a result
* of uev_add_path
*/
- if (conf->reassign_maps) {
+ if (reassign_maps) {
condlog(3, "%s: Reassign existing device-mapper devices",
alias);
dm_reassign(alias);
return 1;
}
}
- r = get_refwwid(dev, DEV_DEVMAP, vecs->pathvec, &refwwid);
+ r = get_refwwid(CMD_NONE, dev, DEV_DEVMAP, vecs->pathvec, &refwwid);
if (refwwid) {
- r = coalesce_paths(vecs, NULL, refwwid, 0);
+ r = coalesce_paths(vecs, NULL, refwwid, FORCE_RELOAD_NONE,
+ CMD_NONE);
dm_lib_release();
}
return 0;
}
minor = uevent_get_minor(uev);
+
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
mpp = find_mp_by_minor(vecs->mpvec, minor);
if (!mpp) {
orphan_paths(vecs->pathvec, mpp);
remove_map_and_stop_waiter(mpp, vecs, 1);
out:
+ lock_cleanup_pop(vecs->lock);
FREE(alias);
return 0;
}
+/* Called from CLI handler */
int
ev_remove_map (char * devname, char * alias, int minor, struct vectors * vecs)
{
}
static int
-uev_add_path (struct uevent *uev, struct vectors * vecs)
+uev_add_path (struct uevent *uev, struct vectors * vecs, int need_do_map)
{
struct path *pp;
int ret = 0, i;
+ struct config *conf;
condlog(2, "%s: add path (uevent)", uev->kernel);
if (strstr(uev->kernel, "..") != NULL) {
return 1;
}
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
pp = find_path_by_dev(vecs->pathvec, uev->kernel);
if (pp) {
int r;
condlog(3, "%s: reinitialize path", uev->kernel);
udev_device_unref(pp->udev);
pp->udev = udev_device_ref(uev->udev);
- r = pathinfo(pp, conf->hwtable,
+ conf = get_multipath_config();
+ r = pathinfo(pp, conf,
DI_ALL | DI_BLACKLIST);
+ put_multipath_config(conf);
if (r == PATHINFO_OK)
- ret = ev_add_path(pp, vecs);
+ ret = ev_add_path(pp, vecs, need_do_map);
else if (r == PATHINFO_SKIPPED) {
condlog(3, "%s: remove blacklisted path",
uev->kernel);
ret = 1;
}
}
- return ret;
}
+ lock_cleanup_pop(vecs->lock);
+ if (pp)
+ return ret;
/*
* get path vital state
*/
- ret = alloc_path_with_pathinfo(conf->hwtable, uev->udev,
- DI_ALL, &pp);
+ conf = get_multipath_config();
+ ret = alloc_path_with_pathinfo(conf, uev->udev,
+ uev->wwid, DI_ALL, &pp);
+ put_multipath_config(conf);
if (!pp) {
if (ret == PATHINFO_SKIPPED)
return 0;
condlog(3, "%s: failed to get path info", uev->kernel);
return 1;
}
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
ret = store_path(vecs->pathvec, pp);
if (!ret) {
+ conf = get_multipath_config();
pp->checkint = conf->checkint;
- ret = ev_add_path(pp, vecs);
+ put_multipath_config(conf);
+ ret = ev_add_path(pp, vecs, need_do_map);
} else {
condlog(0, "%s: failed to store path info, "
"dropping event",
free_path(pp);
ret = 1;
}
-
+ lock_cleanup_pop(vecs->lock);
return ret;
}
* 1: error
*/
int
-ev_add_path (struct path * pp, struct vectors * vecs)
+ev_add_path (struct path * pp, struct vectors * vecs, int need_do_map)
{
struct multipath * mpp;
char params[PARAMS_SIZE] = {0};
goto fail; /* leave path added to pathvec */
}
mpp = find_mp_by_wwid(vecs->mpvec, pp->wwid);
- if (mpp && mpp->wait_for_udev) {
+ if (mpp && mpp->wait_for_udev &&
+ (pathcount(mpp, PATH_UP) > 0 ||
+ (pathcount(mpp, PATH_GHOST) > 0 && pp->tpgs != TPGS_IMPLICIT))) {
+ /* if wait_for_udev is set and valid paths exist */
+ condlog(2, "%s: delaying path addition until %s is fully initialized", pp->dev, mpp->alias);
mpp->wait_for_udev = 2;
orphan_path(pp, "waiting for create to complete");
return 0;
pp->mpp = mpp;
rescan:
if (mpp) {
- if (mpp->size != pp->size) {
+ if (pp->size && mpp->size != pp->size) {
condlog(0, "%s: failed to add new path %s, "
"device size mismatch",
mpp->alias, pp->dev);
condlog(4,"%s: adopting all paths for path %s",
mpp->alias, pp->dev);
- if (adopt_paths(vecs->pathvec, mpp, 1))
+ if (adopt_paths(vecs->pathvec, mpp))
goto fail; /* leave path added to pathvec */
verify_paths(mpp, vecs);
*/
start_waiter = 1;
}
- else
+ if (!start_waiter)
goto fail; /* leave path added to pathvec */
}
- /* persistent reseravtion check*/
- mpath_pr_event_handle(pp);
+ /* persistent reservation check*/
+ mpath_pr_event_handle(pp);
+
+ if (!need_do_map)
+ return 0;
+ if (!dm_map_present(mpp->alias)) {
+ mpp->action = ACT_CREATE;
+ start_waiter = 1;
+ }
/*
* push the map to the device-mapper
*/
* reload the map for the multipath mapped device
*/
retry:
- ret = domap(mpp, params);
+ ret = domap(mpp, params, 1);
if (ret <= 0) {
if (ret < 0 && retries-- > 0) {
condlog(0, "%s: retry domap for addition of new "
* deal with asynchronous uevents :((
*/
if (mpp->action == ACT_RELOAD && retries-- > 0) {
- condlog(0, "%s: uev_add_path sleep", mpp->alias);
+ condlog(0, "%s: ev_add_path sleep", mpp->alias);
sleep(1);
update_mpp_paths(mpp, vecs->pathvec);
goto rescan;
condlog(2, "%s [%s]: path added to devmap %s",
pp->dev, pp->dev_t, mpp->alias);
return 0;
- }
- else
+ } else
goto fail;
fail_map:
}
static int
-uev_remove_path (struct uevent *uev, struct vectors * vecs)
+uev_remove_path (struct uevent *uev, struct vectors * vecs, int need_do_map)
{
struct path *pp;
+ int ret;
condlog(2, "%s: remove path (uevent)", uev->kernel);
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
pp = find_path_by_dev(vecs->pathvec, uev->kernel);
-
+ if (pp)
+ ret = ev_remove_path(pp, vecs, need_do_map);
+ lock_cleanup_pop(vecs->lock);
if (!pp) {
/* Not an error; path might have been purged earlier */
condlog(0, "%s: path already removed", uev->kernel);
return 0;
}
-
- return ev_remove_path(pp, vecs);
+ return ret;
}
int
-ev_remove_path (struct path *pp, struct vectors * vecs)
+ev_remove_path (struct path *pp, struct vectors * vecs, int need_do_map)
{
struct multipath * mpp;
int i, retval = 0;
mpp->retry_tick = 0;
mpp->no_path_retry = NO_PATH_RETRY_FAIL;
mpp->flush_on_last_del = FLUSH_IN_PROGRESS;
+ mpp->stat_map_failures++;
dm_queue_if_no_path(mpp->alias, 0);
}
if (!flush_map(mpp, vecs, 1)) {
goto out;
}
+ if (!need_do_map)
+ goto out;
/*
* reload the map
*/
mpp->action = ACT_RELOAD;
- if (domap(mpp, params) <= 0) {
+ if (domap(mpp, params, 1) <= 0) {
condlog(0, "%s: failed in domap for "
"removal of path %s",
mpp->alias, pp->dev);
{
int ro, retval = 0;
struct path * pp;
+ struct config *conf;
+ int disable_changed_wwids;
+ int needs_reinit = 0;
- pp = find_path_by_dev(vecs->pathvec, uev->kernel);
- if (!pp) {
- condlog(0, "%s: spurious uevent, path not found",
- uev->kernel);
- return 1;
- }
-
- if (pp->initialized == INIT_REQUESTED_UDEV)
- return uev_add_path(uev, vecs);
+ conf = get_multipath_config();
+ disable_changed_wwids = conf->disable_changed_wwids;
+ put_multipath_config(conf);
ro = uevent_get_disk_ro(uev);
- if (ro >= 0) {
- condlog(2, "%s: update path write_protect to '%d' (uevent)",
- uev->kernel, ro);
- if (pp->mpp) {
- if (pp->mpp->wait_for_udev) {
- pp->mpp->wait_for_udev = 2;
- return 0;
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
+
+ pp = find_path_by_dev(vecs->pathvec, uev->kernel);
+ if (pp) {
+ struct multipath *mpp = pp->mpp;
+
+ if (disable_changed_wwids &&
+ (strlen(pp->wwid) || pp->wwid_changed)) {
+ char wwid[WWID_SIZE];
+
+ strcpy(wwid, pp->wwid);
+ get_uid(pp, pp->state, uev->udev);
+ if (strcmp(wwid, pp->wwid) != 0) {
+ condlog(0, "%s: path wwid changed from '%s' to '%s'. disallowing", uev->kernel, wwid, pp->wwid);
+ strcpy(pp->wwid, wwid);
+ if (!pp->wwid_changed) {
+ pp->wwid_changed = 1;
+ pp->tick = 1;
+ if (pp->mpp)
+ dm_fail_path(pp->mpp->alias, pp->dev_t);
+ }
+ goto out;
+ } else
+ pp->wwid_changed = 0;
+ }
+
+ if (pp->initialized == INIT_REQUESTED_UDEV)
+ needs_reinit = 1;
+ else if (mpp && ro >= 0) {
+ condlog(2, "%s: update path write_protect to '%d' (uevent)", uev->kernel, ro);
+
+ if (mpp->wait_for_udev)
+ mpp->wait_for_udev = 2;
+ else {
+ if (ro == 1)
+ pp->mpp->force_readonly = 1;
+ retval = reload_map(vecs, mpp, 0, 1);
+ pp->mpp->force_readonly = 0;
+ condlog(2, "%s: map %s reloaded (retval %d)",
+ uev->kernel, mpp->alias, retval);
}
+ }
+ }
+out:
+ lock_cleanup_pop(vecs->lock);
+ if (!pp) {
+ /* If the path is blacklisted, print a debug/non-default verbosity message. */
+ if (uev->udev) {
+ int flag = DI_SYSFS | DI_WWID;
- retval = reload_map(vecs, pp->mpp, 0);
+ conf = get_multipath_config();
+ retval = alloc_path_with_pathinfo(conf, uev->udev, uev->wwid, flag, NULL);
+ put_multipath_config(conf);
- condlog(2, "%s: map %s reloaded (retval %d)",
- uev->kernel, pp->mpp->alias, retval);
+ if (retval == PATHINFO_SKIPPED) {
+ condlog(3, "%s: spurious uevent, path is blacklisted", uev->kernel);
+ return 0;
+ }
}
+ condlog(0, "%s: spurious uevent, path not found", uev->kernel);
}
-
+ if (needs_reinit)
+ retval = uev_add_path(uev, vecs, 1);
return retval;
}
vector_foreach_slot (vecs->mpvec, mpp, i)
if (setup_multipath(vecs, mpp))
- return 1;
+ i--;
return 0;
}
int
-uxsock_trigger (char * str, char ** reply, int * len, void * trigger_data)
+uxsock_trigger (char * str, char ** reply, int * len, bool is_root,
+ void * trigger_data)
{
struct vectors * vecs;
int r;
*len = 0;
vecs = (struct vectors *)trigger_data;
- pthread_cleanup_push(cleanup_lock, &vecs->lock);
- lock(vecs->lock);
- pthread_testcancel();
+ if ((str != NULL) && (is_root == false) &&
+ (strncmp(str, "list", strlen("list")) != 0) &&
+ (strncmp(str, "show", strlen("show")) != 0)) {
+ *reply = STRDUP("permission deny: need to be root");
+ if (*reply)
+ *len = strlen(*reply) + 1;
+ return 1;
+ }
- r = parse_cmd(str, reply, len, vecs);
+ r = parse_cmd(str, reply, len, vecs, uxsock_timeout);
if (r > 0) {
- *reply = STRDUP("fail\n");
- *len = strlen(*reply) + 1;
+ if (r == ETIMEDOUT)
+ *reply = STRDUP("timeout\n");
+ else
+ *reply = STRDUP("fail\n");
+ if (*reply)
+ *len = strlen(*reply) + 1;
r = 1;
}
else if (!r && *len == 0) {
*reply = STRDUP("ok\n");
- *len = strlen(*reply) + 1;
+ if (*reply)
+ *len = strlen(*reply) + 1;
r = 0;
}
/* else if (r < 0) leave *reply alone */
- lock_cleanup_pop(vecs->lock);
return r;
}
-static int
-uev_discard(char * devpath)
-{
- char *tmp;
- char a[11], b[11];
-
- /*
- * keep only block devices, discard partitions
- */
- tmp = strstr(devpath, "/block/");
- if (tmp == NULL){
- condlog(4, "no /block/ in '%s'", devpath);
- return 1;
- }
- if (sscanf(tmp, "/block/%10s", a) != 1 ||
- sscanf(tmp, "/block/%10[^/]/%10s", a, b) == 2) {
- condlog(4, "discard event on %s", devpath);
- return 1;
- }
- return 0;
-}
-
int
uev_trigger (struct uevent * uev, void * trigger_data)
{
int r = 0;
struct vectors * vecs;
+ struct uevent *merge_uev, *tmp;
vecs = (struct vectors *)trigger_data;
- if (uev_discard(uev->devpath))
- return 0;
+ pthread_cleanup_push(config_cleanup, NULL);
+ pthread_mutex_lock(&config_lock);
+ if (running_state != DAEMON_IDLE &&
+ running_state != DAEMON_RUNNING)
+ pthread_cond_wait(&config_cond, &config_lock);
+ pthread_cleanup_pop(1);
- pthread_cleanup_push(cleanup_lock, &vecs->lock);
- lock(vecs->lock);
- pthread_testcancel();
+ if (running_state == DAEMON_SHUTDOWN)
+ return 0;
/*
* device map event
}
/*
- * path add/remove event
+ * path add/remove/change event, add/remove maybe merged
*/
- if (filter_devnode(conf->blist_devnode, conf->elist_devnode,
- uev->kernel) > 0)
- goto out;
-
- if (!strncmp(uev->action, "add", 3)) {
- r = uev_add_path(uev, vecs);
- goto out;
- }
- if (!strncmp(uev->action, "remove", 6)) {
- r = uev_remove_path(uev, vecs);
- goto out;
- }
- if (!strncmp(uev->action, "change", 6)) {
- r = uev_update_path(uev, vecs);
- goto out;
+ list_for_each_entry_safe(merge_uev, tmp, &uev->merge_node, node) {
+ if (!strncmp(merge_uev->action, "add", 3))
+ r += uev_add_path(merge_uev, vecs, 0);
+ if (!strncmp(merge_uev->action, "remove", 6))
+ r += uev_remove_path(merge_uev, vecs, 0);
}
+ if (!strncmp(uev->action, "add", 3))
+ r += uev_add_path(uev, vecs, 1);
+ if (!strncmp(uev->action, "remove", 6))
+ r += uev_remove_path(uev, vecs, 1);
+ if (!strncmp(uev->action, "change", 6))
+ r += uev_update_path(uev, vecs);
+
out:
- lock_cleanup_pop(vecs->lock);
return r;
}
+static void rcu_unregister(void *param)
+{
+ rcu_unregister_thread();
+}
+
static void *
ueventloop (void * ap)
{
struct udev *udev = ap;
+ pthread_cleanup_push(rcu_unregister, NULL);
+ rcu_register_thread();
if (uevent_listen(udev))
condlog(0, "error starting uevent listener");
-
+ pthread_cleanup_pop(1);
return NULL;
}
static void *
uevqloop (void * ap)
{
+ pthread_cleanup_push(rcu_unregister, NULL);
+ rcu_register_thread();
if (uevent_dispatch(&uev_trigger, ap))
condlog(0, "error starting uevent dispatcher");
-
+ pthread_cleanup_pop(1);
return NULL;
}
static void *
uxlsnrloop (void * ap)
{
- if (cli_init())
+ if (cli_init()) {
+ condlog(1, "Failed to init uxsock listener");
return NULL;
-
+ }
+ pthread_cleanup_push(rcu_unregister, NULL);
+ rcu_register_thread();
set_handler_callback(LIST+PATHS, cli_list_paths);
set_handler_callback(LIST+PATHS+FMT, cli_list_paths_fmt);
set_handler_callback(LIST+PATHS+RAW+FMT, cli_list_paths_raw);
set_handler_callback(LIST+PATH, cli_list_path);
set_handler_callback(LIST+MAPS, cli_list_maps);
- set_handler_callback(LIST+STATUS, cli_list_status);
- set_handler_callback(LIST+DAEMON, cli_list_daemon);
+ set_unlocked_handler_callback(LIST+STATUS, cli_list_status);
+ set_unlocked_handler_callback(LIST+DAEMON, cli_list_daemon);
set_handler_callback(LIST+MAPS+STATUS, cli_list_maps_status);
set_handler_callback(LIST+MAPS+STATS, cli_list_maps_stats);
set_handler_callback(LIST+MAPS+FMT, cli_list_maps_fmt);
set_handler_callback(LIST+MAPS+RAW+FMT, cli_list_maps_raw);
set_handler_callback(LIST+MAPS+TOPOLOGY, cli_list_maps_topology);
set_handler_callback(LIST+TOPOLOGY, cli_list_maps_topology);
+ set_handler_callback(LIST+MAPS+JSON, cli_list_maps_json);
set_handler_callback(LIST+MAP+TOPOLOGY, cli_list_map_topology);
+ set_handler_callback(LIST+MAP+FMT, cli_list_map_fmt);
+ set_handler_callback(LIST+MAP+RAW+FMT, cli_list_map_fmt);
+ set_handler_callback(LIST+MAP+JSON, cli_list_map_json);
set_handler_callback(LIST+CONFIG, cli_list_config);
set_handler_callback(LIST+BLACKLIST, cli_list_blacklist);
set_handler_callback(LIST+DEVICES, cli_list_devices);
set_handler_callback(LIST+WILDCARDS, cli_list_wildcards);
+ set_handler_callback(RESET+MAPS+STATS, cli_reset_maps_stats);
+ set_handler_callback(RESET+MAP+STATS, cli_reset_map_stats);
set_handler_callback(ADD+PATH, cli_add_path);
set_handler_callback(DEL+PATH, cli_del_path);
set_handler_callback(ADD+MAP, cli_add_map);
set_handler_callback(DEL+MAP, cli_del_map);
set_handler_callback(SWITCH+MAP+GROUP, cli_switch_group);
- set_handler_callback(RECONFIGURE, cli_reconfigure);
+ set_unlocked_handler_callback(RECONFIGURE, cli_reconfigure);
set_handler_callback(SUSPEND+MAP, cli_suspend);
set_handler_callback(RESUME+MAP, cli_resume);
set_handler_callback(RESIZE+MAP, cli_resize);
set_handler_callback(RESTOREQ+MAP, cli_restore_queueing);
set_handler_callback(DISABLEQ+MAPS, cli_disable_all_queueing);
set_handler_callback(RESTOREQ+MAPS, cli_restore_all_queueing);
- set_handler_callback(QUIT, cli_quit);
- set_handler_callback(SHUTDOWN, cli_shutdown);
+ set_unlocked_handler_callback(QUIT, cli_quit);
+ set_unlocked_handler_callback(SHUTDOWN, cli_shutdown);
set_handler_callback(GETPRSTATUS+MAP, cli_getprstatus);
set_handler_callback(SETPRSTATUS+MAP, cli_setprstatus);
set_handler_callback(UNSETPRSTATUS+MAP, cli_unsetprstatus);
umask(077);
uxsock_listen(&uxsock_trigger, ap);
-
+ pthread_cleanup_pop(1);
return NULL;
}
void
exit_daemon (void)
{
- sem_post(&exit_sem);
-}
-
-const char *
-daemon_status(void)
-{
- switch (running_state) {
- case DAEMON_INIT:
- return "init";
- case DAEMON_START:
- return "startup";
- case DAEMON_CONFIGURE:
- return "configure";
- case DAEMON_RUNNING:
- return "running";
- case DAEMON_SHUTDOWN:
- return "shutdown";
- }
- return NULL;
+ post_config_state(DAEMON_SHUTDOWN);
}
static void
{
struct multipath * mpp;
unsigned int i;
- int timed_out = 0;
+ int timed_out = 0, delayed_reconfig;
+ struct config *conf;
vector_foreach_slot (vecs->mpvec, mpp, i) {
if (mpp->wait_for_udev && --mpp->uev_wait_tick <= 0) {
}
}
- if (timed_out && conf->delayed_reconfig &&
+ conf = get_multipath_config();
+ delayed_reconfig = conf->delayed_reconfig;
+ put_multipath_config(conf);
+ if (timed_out && delayed_reconfig &&
!need_to_delay_reconfig(vecs)) {
condlog(2, "reconfigure (delayed)");
- reconfigure(vecs);
+ set_config_state(DAEMON_CONFIGURE);
}
}
unsigned int i;
vector_foreach_slot (mpvec, mpp, i) {
- if (mpp->retry_tick) {
+ if (mpp->retry_tick > 0) {
mpp->stat_total_queueing_time++;
condlog(4, "%s: Retrying.. No active path", mpp->alias);
if(--mpp->retry_tick == 0) {
+ mpp->stat_map_failures++;
dm_queue_if_no_path(mpp->alias, 0);
condlog(2, "%s: Disable queueing", mpp->alias);
}
struct path *pp1;
struct pathgroup * pgp;
int i, j, changed = 0;
+ struct config *conf;
if (refresh_all) {
vector_foreach_slot (pp->mpp->pg, pgp, i) {
vector_foreach_slot (pgp->paths, pp1, j) {
oldpriority = pp1->priority;
- pathinfo(pp1, conf->hwtable, DI_PRIO);
+ conf = get_multipath_config();
+ pathinfo(pp1, conf, DI_PRIO);
+ put_multipath_config(conf);
if (pp1->priority != oldpriority)
changed = 1;
}
return changed;
}
oldpriority = pp->priority;
- pathinfo(pp, conf->hwtable, DI_PRIO);
+ conf = get_multipath_config();
+ if (pp->state != PATH_DOWN)
+ pathinfo(pp, conf, DI_PRIO);
+ put_multipath_config(conf);
if (pp->priority == oldpriority)
return 0;
int update_path_groups(struct multipath *mpp, struct vectors *vecs, int refresh)
{
- if (reload_map(vecs, mpp, refresh))
+ if (reload_map(vecs, mpp, refresh, 1))
return 1;
dm_lib_release();
return 0;
}
+void repair_path(struct path * pp)
+{
+ if (pp->state != PATH_DOWN)
+ return;
+
+ checker_repair(&pp->checker);
+ LOG_MSG(1, checker_message(&pp->checker));
+}
+
+static int check_path_reinstate_state(struct path * pp) {
+ struct timespec curr_time;
+ if (!((pp->mpp->san_path_err_threshold > 0) &&
+ (pp->mpp->san_path_err_forget_rate > 0) &&
+ (pp->mpp->san_path_err_recovery_time >0))) {
+ return 0;
+ }
+
+ if (pp->disable_reinstate) {
+ /* If we don't know how much time has passed, automatically
+ * reinstate the path, just to be safe. Also, if there are
+ * no other usable paths, reinstate the path
+ */
+ if (clock_gettime(CLOCK_MONOTONIC, &curr_time) != 0 ||
+ pp->mpp->nr_active == 0) {
+ condlog(2, "%s : reinstating path early", pp->dev);
+ goto reinstate_path;
+ }
+ if ((curr_time.tv_sec - pp->dis_reinstate_time ) > pp->mpp->san_path_err_recovery_time) {
+ condlog(2,"%s : reinstate the path after err recovery time", pp->dev);
+ goto reinstate_path;
+ }
+ return 1;
+ }
+ /* forget errors on a working path */
+ if ((pp->state == PATH_UP || pp->state == PATH_GHOST) &&
+ pp->path_failures > 0) {
+ if (pp->san_path_err_forget_rate > 0){
+ pp->san_path_err_forget_rate--;
+ } else {
+ /* for every san_path_err_forget_rate number of
+ * successful path checks decrement path_failures by 1
+ */
+ pp->path_failures--;
+ pp->san_path_err_forget_rate = pp->mpp->san_path_err_forget_rate;
+ }
+ return 0;
+ }
+
+ /* If the path isn't recovering from a failed state, do nothing */
+ if (pp->state != PATH_DOWN && pp->state != PATH_SHAKY &&
+ pp->state != PATH_TIMEOUT)
+ return 0;
+
+ if (pp->path_failures == 0)
+ pp->san_path_err_forget_rate = pp->mpp->san_path_err_forget_rate;
+
+ pp->path_failures++;
+
+ /* if we don't know the currently time, we don't know how long to
+ * delay the path, so there's no point in checking if we should
+ */
+
+ if (clock_gettime(CLOCK_MONOTONIC, &curr_time) != 0)
+ return 0;
+ /* when path failures has exceeded the san_path_err_threshold
+ * place the path in delayed state till san_path_err_recovery_time
+ * so that the cutomer can rectify the issue within this time. After
+ * the completion of san_path_err_recovery_time it should
+ * automatically reinstate the path
+ */
+ if (pp->path_failures > pp->mpp->san_path_err_threshold) {
+ condlog(2, "%s : hit error threshold. Delaying path reinstatement", pp->dev);
+ pp->dis_reinstate_time = curr_time.tv_sec;
+ pp->disable_reinstate = 1;
+ return 1;
+ } else {
+ return 0;
+ }
+
+reinstate_path:
+ pp->path_failures = 0;
+ pp->disable_reinstate = 0;
+ pp->san_path_err_forget_rate = 0;
+ return 0;
+}
+
/*
- * Returns '1' if the path has been checked, '0' otherwise
+ * Returns '1' if the path has been checked, '-1' if it was blacklisted
+ * and '0' otherwise
*/
int
-check_path (struct vectors * vecs, struct path * pp)
+check_path (struct vectors * vecs, struct path * pp, int ticks)
{
int newstate;
int new_path_up = 0;
int add_active;
int disable_reinstate = 0;
int oldchkrstate = pp->chkrstate;
+ int retrigger_tries, checkint;
+ struct config *conf;
+ int ret;
if ((pp->initialized == INIT_OK ||
pp->initialized == INIT_REQUESTED_UDEV) && !pp->mpp)
return 0;
- if (pp->tick && --pp->tick)
+ if (pp->tick)
+ pp->tick -= (pp->tick > ticks) ? ticks : pp->tick;
+ if (pp->tick)
return 0; /* don't check this path yet */
+ conf = get_multipath_config();
+ retrigger_tries = conf->retrigger_tries;
+ checkint = conf->checkint;
+ put_multipath_config(conf);
if (!pp->mpp && pp->initialized == INIT_MISSING_UDEV &&
- pp->retriggers < conf->retrigger_tries) {
+ pp->retriggers < retrigger_tries) {
condlog(2, "%s: triggering change event to reinitialize",
pp->dev);
pp->initialized = INIT_REQUESTED_UDEV;
sysfs_attr_set_value(pp->udev, "uevent", "change",
strlen("change"));
return 0;
- }
+ }
/*
* provision a next check soonest,
* in case we exit abnormaly from here
*/
- pp->tick = conf->checkint;
+ pp->tick = checkint;
newstate = path_offline(pp);
/*
if (newstate == PATH_REMOVED)
newstate = PATH_DOWN;
- if (newstate == PATH_UP)
- newstate = get_state(pp, 1);
- else
+ if (newstate == PATH_UP) {
+ conf = get_multipath_config();
+ newstate = get_state(pp, conf, 1);
+ put_multipath_config(conf);
+ } else
checker_clear_message(&pp->checker);
+ if (pp->wwid_changed) {
+ condlog(2, "%s: path wwid has changed. Refusing to use",
+ pp->dev);
+ newstate = PATH_DOWN;
+ }
+
if (newstate == PATH_WILD || newstate == PATH_UNCHECKED) {
condlog(2, "%s: unusable path", pp->dev);
- pathinfo(pp, conf->hwtable, 0);
+ conf = get_multipath_config();
+ pathinfo(pp, conf, 0);
+ put_multipath_config(conf);
return 1;
}
if (!pp->mpp) {
if (!strlen(pp->wwid) && pp->initialized != INIT_MISSING_UDEV &&
(newstate == PATH_UP || newstate == PATH_GHOST)) {
condlog(2, "%s: add missing path", pp->dev);
- if (pathinfo(pp, conf->hwtable, DI_ALL) == 0) {
- ev_add_path(pp, vecs);
+ conf = get_multipath_config();
+ ret = pathinfo(pp, conf, DI_ALL | DI_BLACKLIST);
+ if (ret == PATHINFO_OK) {
+ ev_add_path(pp, vecs, 1);
pp->tick = 1;
+ } else if (ret == PATHINFO_SKIPPED) {
+ put_multipath_config(conf);
+ return -1;
}
+ put_multipath_config(conf);
}
return 0;
}
/*
* Synchronize with kernel state
*/
- if (update_multipath_strings(pp->mpp, vecs->pathvec)) {
+ if (update_multipath_strings(pp->mpp, vecs->pathvec, 1)) {
condlog(1, "%s: Could not synchronize with kernel state",
pp->dev);
pp->dmstate = PSTATE_UNDEF;
return 0;
if ((newstate == PATH_UP || newstate == PATH_GHOST) &&
+ check_path_reinstate_state(pp)) {
+ pp->state = PATH_DELAYED;
+ return 1;
+ }
+
+ if ((newstate == PATH_UP || newstate == PATH_GHOST) &&
pp->wait_checks > 0) {
- if (pp->mpp && pp->mpp->nr_active > 0) {
+ if (pp->mpp->nr_active > 0) {
pp->state = PATH_DELAYED;
pp->wait_checks--;
return 1;
int oldstate = pp->state;
pp->state = newstate;
- if (strlen(checker_message(&pp->checker)))
- LOG_MSG(1, checker_message(&pp->checker));
+ LOG_MSG(1, checker_message(&pp->checker));
/*
* upon state change, reset the checkint
* to the shortest delay
*/
+ conf = get_multipath_config();
pp->checkint = conf->checkint;
+ put_multipath_config(conf);
- if (newstate == PATH_DOWN || newstate == PATH_SHAKY) {
+ if (newstate != PATH_UP && newstate != PATH_GHOST) {
/*
* proactively fail path in the DM
*/
pp->mpp->failback_tick = 0;
pp->mpp->stat_path_failures++;
+ repair_path(pp);
return 1;
}
}
if (!disable_reinstate && reinstate_path(pp, add_active)) {
condlog(3, "%s: reload map", pp->dev);
- ev_add_path(pp, vecs);
+ ev_add_path(pp, vecs, 1);
pp->tick = 1;
return 0;
}
/* Clear IO errors */
if (reinstate_path(pp, 0)) {
condlog(3, "%s: reload map", pp->dev);
- ev_add_path(pp, vecs);
+ ev_add_path(pp, vecs, 1);
pp->tick = 1;
return 0;
}
} else {
+ unsigned int max_checkint;
LOG_MSG(4, checker_message(&pp->checker));
- if (pp->checkint != conf->max_checkint) {
+ conf = get_multipath_config();
+ max_checkint = conf->max_checkint;
+ put_multipath_config(conf);
+ if (pp->checkint != max_checkint) {
/*
* double the next check delay.
* max at conf->max_checkint
*/
- if (pp->checkint < (conf->max_checkint / 2))
+ if (pp->checkint < (max_checkint / 2))
pp->checkint = 2 * pp->checkint;
else
- pp->checkint = conf->max_checkint;
+ pp->checkint = max_checkint;
condlog(4, "%s: delay next check %is",
pp->dev_t, pp->checkint);
pp->tick = pp->checkint;
}
}
- else if (newstate == PATH_DOWN &&
- strlen(checker_message(&pp->checker))) {
- if (conf->log_checker_err == LOG_CHKR_ERR_ONCE)
+ else if (newstate == PATH_DOWN) {
+ int log_checker_err;
+
+ conf = get_multipath_config();
+ log_checker_err = conf->log_checker_err;
+ put_multipath_config(conf);
+ if (log_checker_err == LOG_CHKR_ERR_ONCE)
LOG_MSG(3, checker_message(&pp->checker));
else
LOG_MSG(2, checker_message(&pp->checker));
}
pp->state = newstate;
-
+ repair_path(pp);
if (pp->mpp->wait_for_udev)
return 1;
return 1;
}
+static void init_path_check_interval(struct vectors *vecs)
+{
+ struct config *conf;
+ struct path *pp;
+ unsigned int i;
+
+ vector_foreach_slot (vecs->pathvec, pp, i) {
+ conf = get_multipath_config();
+ pp->checkint = conf->checkint;
+ put_multipath_config(conf);
+ }
+}
+
static void *
checkerloop (void *ap)
{
struct path *pp;
int count = 0;
unsigned int i;
+ struct itimerval timer_tick_it;
+ struct timespec last_time;
+ struct config *conf;
+ pthread_cleanup_push(rcu_unregister, NULL);
+ rcu_register_thread();
mlockall(MCL_CURRENT | MCL_FUTURE);
vecs = (struct vectors *)ap;
condlog(2, "path checkers start up");
- /*
- * init the path check interval
- */
- vector_foreach_slot (vecs->pathvec, pp, i) {
- pp->checkint = conf->checkint;
- }
+ /* Tweak start time for initial path check */
+ if (clock_gettime(CLOCK_MONOTONIC, &last_time) != 0)
+ last_time.tv_sec = 0;
+ else
+ last_time.tv_sec -= 1;
while (1) {
- struct timeval diff_time, start_time, end_time;
- int num_paths = 0;
+ struct timespec diff_time, start_time, end_time;
+ int num_paths = 0, ticks = 0, signo, strict_timing, rc = 0;
+ sigset_t mask;
- if (gettimeofday(&start_time, NULL) != 0)
+ if (clock_gettime(CLOCK_MONOTONIC, &start_time) != 0)
start_time.tv_sec = 0;
- pthread_cleanup_push(cleanup_lock, &vecs->lock);
- lock(vecs->lock);
- pthread_testcancel();
- condlog(4, "tick");
+ if (start_time.tv_sec && last_time.tv_sec) {
+ timespecsub(&start_time, &last_time, &diff_time);
+ condlog(4, "tick (%lu.%06lu secs)",
+ diff_time.tv_sec, diff_time.tv_nsec / 1000);
+ last_time = start_time;
+ ticks = diff_time.tv_sec;
+ } else {
+ ticks = 1;
+ condlog(4, "tick (%d ticks)", ticks);
+ }
#ifdef USE_SYSTEMD
if (use_watchdog)
sd_notify(0, "WATCHDOG=1");
#endif
- if (vecs->pathvec) {
- vector_foreach_slot (vecs->pathvec, pp, i) {
- num_paths += check_path(vecs, pp);
- }
+ rc = set_config_state(DAEMON_RUNNING);
+ if (rc == ETIMEDOUT) {
+ condlog(4, "timeout waiting for DAEMON_IDLE");
+ continue;
}
- if (vecs->mpvec) {
- defered_failback_tick(vecs->mpvec);
- retry_count_tick(vecs->mpvec);
- missing_uev_wait_tick(vecs);
+
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
+ vector_foreach_slot (vecs->pathvec, pp, i) {
+ rc = check_path(vecs, pp, ticks);
+ if (rc < 0) {
+ vector_del_slot(vecs->pathvec, i);
+ free_path(pp);
+ i--;
+ } else
+ num_paths += rc;
}
+ lock_cleanup_pop(vecs->lock);
+
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
+ defered_failback_tick(vecs->mpvec);
+ retry_count_tick(vecs->mpvec);
+ missing_uev_wait_tick(vecs);
+ lock_cleanup_pop(vecs->lock);
+
if (count)
count--;
else {
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
condlog(4, "map garbage collection");
mpvec_garbage_collector(vecs);
count = MAPGCINT;
+ lock_cleanup_pop(vecs->lock);
}
- lock_cleanup_pop(vecs->lock);
+ diff_time.tv_nsec = 0;
if (start_time.tv_sec &&
- gettimeofday(&end_time, NULL) == 0 &&
- num_paths) {
- timersub(&end_time, &start_time, &diff_time);
- condlog(3, "checked %d path%s in %lu.%06lu secs",
- num_paths, num_paths > 1 ? "s" : "",
- diff_time.tv_sec, diff_time.tv_usec);
+ clock_gettime(CLOCK_MONOTONIC, &end_time) == 0) {
+ timespecsub(&end_time, &start_time, &diff_time);
+ if (num_paths) {
+ unsigned int max_checkint;
+
+ condlog(3, "checked %d path%s in %lu.%06lu secs",
+ num_paths, num_paths > 1 ? "s" : "",
+ diff_time.tv_sec,
+ diff_time.tv_nsec / 1000);
+ conf = get_multipath_config();
+ max_checkint = conf->max_checkint;
+ put_multipath_config(conf);
+ if (diff_time.tv_sec > max_checkint)
+ condlog(1, "path checkers took longer "
+ "than %lu seconds, consider "
+ "increasing max_polling_interval",
+ diff_time.tv_sec);
+ }
+ }
+
+ post_config_state(DAEMON_IDLE);
+ conf = get_multipath_config();
+ strict_timing = conf->strict_timing;
+ put_multipath_config(conf);
+ if (!strict_timing)
+ sleep(1);
+ else {
+ timer_tick_it.it_interval.tv_sec = 0;
+ timer_tick_it.it_interval.tv_usec = 0;
+ if (diff_time.tv_nsec) {
+ timer_tick_it.it_value.tv_sec = 0;
+ timer_tick_it.it_value.tv_usec =
+ 1000UL * 1000 * 1000 - diff_time.tv_nsec;
+ } else {
+ timer_tick_it.it_value.tv_sec = 1;
+ timer_tick_it.it_value.tv_usec = 0;
+ }
+ setitimer(ITIMER_REAL, &timer_tick_it, NULL);
+
+ sigemptyset(&mask);
+ sigaddset(&mask, SIGALRM);
+ condlog(3, "waiting for %lu.%06lu secs",
+ timer_tick_it.it_value.tv_sec,
+ timer_tick_it.it_value.tv_usec);
+ if (sigwait(&mask, &signo) != 0) {
+ condlog(3, "sigwait failed with error %d",
+ errno);
+ conf = get_multipath_config();
+ conf->strict_timing = 0;
+ put_multipath_config(conf);
+ break;
+ }
}
- sleep(1);
}
+ pthread_cleanup_pop(1);
return NULL;
}
struct path * pp;
vector mpvec;
int i, ret;
+ struct config *conf;
+ static int force_reload = FORCE_RELOAD_WEAK;
- if (!vecs->pathvec && !(vecs->pathvec = vector_alloc()))
+ if (!vecs->pathvec && !(vecs->pathvec = vector_alloc())) {
+ condlog(0, "couldn't allocate path vec in configure");
return 1;
+ }
- if (!vecs->mpvec && !(vecs->mpvec = vector_alloc()))
+ if (!vecs->mpvec && !(vecs->mpvec = vector_alloc())) {
+ condlog(0, "couldn't allocate multipath vec in configure");
return 1;
+ }
- if (!(mpvec = vector_alloc()))
+ if (!(mpvec = vector_alloc())) {
+ condlog(0, "couldn't allocate new maps vec in configure");
return 1;
+ }
/*
* probe for current path (from sysfs) and map (from dm) sets
*/
- ret = path_discovery(vecs->pathvec, conf, DI_ALL);
- if (ret < 0)
+ ret = path_discovery(vecs->pathvec, DI_ALL);
+ if (ret < 0) {
+ condlog(0, "configure failed at path discovery");
return 1;
+ }
vector_foreach_slot (vecs->pathvec, pp, i){
+ conf = get_multipath_config();
if (filter_path(conf, pp) > 0){
vector_del_slot(vecs->pathvec, i);
free_path(pp);
}
else
pp->checkint = conf->checkint;
+ put_multipath_config(conf);
}
- if (map_discovery(vecs))
+ if (map_discovery(vecs)) {
+ condlog(0, "configure failed at map discovery");
return 1;
+ }
/*
* create new set of maps & push changed ones into dm
+ * In the first call, use FORCE_RELOAD_WEAK to avoid making
+ * superfluous ACT_RELOAD ioctls. Later calls are done
+ * with FORCE_RELOAD_YES.
*/
- if (coalesce_paths(vecs, mpvec, NULL, 1))
+ ret = coalesce_paths(vecs, mpvec, NULL, force_reload, CMD_NONE);
+ if (force_reload == FORCE_RELOAD_WEAK)
+ force_reload = FORCE_RELOAD_YES;
+ if (ret) {
+ condlog(0, "configure failed while coalescing paths");
return 1;
+ }
/*
* may need to remove some maps which are no longer relevant
* e.g., due to blacklist changes in conf file
*/
- if (coalesce_maps(vecs, mpvec))
+ if (coalesce_maps(vecs, mpvec)) {
+ condlog(0, "configure failed while coalescing maps");
return 1;
+ }
dm_lib_release();
* start dm event waiter threads for these new maps
*/
vector_foreach_slot(vecs->mpvec, mpp, i) {
- if (setup_multipath(vecs, mpp))
- return 1;
- if (start_waiters)
- if (start_waiter_thread(mpp, vecs))
- return 1;
+ if (setup_multipath(vecs, mpp)) {
+ i--;
+ continue;
+ }
+ if (start_waiters) {
+ if (start_waiter_thread(mpp, vecs)) {
+ remove_map(mpp, vecs, 1);
+ i--;
+ }
+ }
}
return 0;
}
return 0;
}
+void rcu_free_config(struct rcu_head *head)
+{
+ struct config *conf = container_of(head, struct config, rcu);
+
+ free_config(conf);
+}
+
int
reconfigure (struct vectors * vecs)
{
- struct config * old = conf;
- int retval = 1;
+ struct config * old, *conf;
- running_state = DAEMON_CONFIGURE;
+ conf = load_config(DEFAULT_CONFIGFILE);
+ if (!conf)
+ return 1;
/*
* free old map and path vectors ... they use old conf state
if (VECTOR_SIZE(vecs->mpvec))
remove_maps_and_stop_waiters(vecs);
- if (VECTOR_SIZE(vecs->pathvec))
- free_pathvec(vecs->pathvec, FREE_PATHS);
-
+ free_pathvec(vecs->pathvec, FREE_PATHS);
vecs->pathvec = NULL;
- conf = NULL;
/* Re-read any timezone changes */
tzset();
- if (!load_config(DEFAULT_CONFIGFILE, udev)) {
- dm_drv_version(conf->version, TGT_MPATH);
- conf->verbosity = old->verbosity;
- conf->bindings_read_only = old->bindings_read_only;
- conf->ignore_new_devs = old->ignore_new_devs;
- conf->daemon = 1;
- configure(vecs, 1);
- free_config(old);
- retval = 0;
+ dm_drv_version(conf->version, TGT_MPATH);
+ if (verbosity)
+ conf->verbosity = verbosity;
+ if (bindings_read_only)
+ conf->bindings_read_only = bindings_read_only;
+ if (conf->find_multipaths) {
+ condlog(2, "find_multipaths is set: -n is implied");
+ ignore_new_devs = 1;
}
+ if (ignore_new_devs)
+ conf->ignore_new_devs = ignore_new_devs;
+ uxsock_timeout = conf->uxsock_timeout;
- running_state = DAEMON_RUNNING;
+ old = rcu_dereference(multipath_conf);
+ rcu_assign_pointer(multipath_conf, conf);
+ call_rcu(&old->rcu, rcu_free_config);
- return retval;
+ configure(vecs, 1);
+
+
+ return 0;
}
static struct vectors *
if (!vecs)
return NULL;
- vecs->lock.mutex =
- (pthread_mutex_t *)MALLOC(sizeof(pthread_mutex_t));
-
- if (!vecs->lock.mutex)
- goto out;
-
- pthread_mutex_init(vecs->lock.mutex, NULL);
- vecs->lock.depth = 0;
+ pthread_mutex_init(&vecs->lock.mutex, NULL);
return vecs;
-
-out:
- FREE(vecs);
- condlog(0, "failed to init paths");
- return NULL;
}
static void *
void
handle_signals(void)
{
- if (reconfig_sig && running_state == DAEMON_RUNNING) {
- pthread_cleanup_push(cleanup_lock,
- &gvecs->lock);
- lock(gvecs->lock);
- pthread_testcancel();
- if (need_to_delay_reconfig(gvecs)) {
- conf->delayed_reconfig = 1;
- condlog(2, "delaying reconfigure (signal)");
- }
- else {
- condlog(2, "reconfigure (signal)");
- reconfigure(gvecs);
- }
- lock_cleanup_pop(gvecs->lock);
+ if (exit_sig) {
+ condlog(2, "exit (signal)");
+ exit_daemon();
+ }
+ if (reconfig_sig) {
+ condlog(2, "reconfigure (signal)");
+ set_config_state(DAEMON_CONFIGURE);
}
if (log_reset_sig) {
condlog(2, "reset log (signal)");
log_reset("multipathd");
pthread_mutex_unlock(&logq_lock);
}
+ exit_sig = 0;
reconfig_sig = 0;
log_reset_sig = 0;
}
static void
sigend (int sig)
{
- exit_daemon();
+ exit_sig = 1;
}
static void
sigset_t set;
sigemptyset(&set);
- sigaddset(&set, SIGHUP);
- sigaddset(&set, SIGUSR1);
sigaddset(&set, SIGUSR2);
- pthread_sigmask(SIG_BLOCK, &set, NULL);
+ pthread_sigmask(SIG_SETMASK, &set, NULL);
signal_set(SIGHUP, sighup);
signal_set(SIGUSR1, sigusr1);
signal_set(SIGUSR2, sigusr2);
signal_set(SIGINT, sigend);
signal_set(SIGTERM, sigend);
- signal(SIGPIPE, SIG_IGN);
+ signal_set(SIGPIPE, sigend);
}
static void
int i;
#ifdef USE_SYSTEMD
unsigned long checkint;
+ int startup_done = 0;
#endif
int rc;
+ int pid_fd = -1;
+ struct config *conf;
char *envp;
mlockall(MCL_CURRENT | MCL_FUTURE);
- sem_init(&exit_sem, 0, 0);
signal_init();
+ rcu_init();
- udev = udev_new();
-
- setup_thread_attr(&misc_attr, 64 * 1024, 1);
- setup_thread_attr(&uevent_attr, DEFAULT_UEVENT_STACKSIZE * 1024, 1);
+ setup_thread_attr(&misc_attr, 64 * 1024, 0);
+ setup_thread_attr(&uevent_attr, DEFAULT_UEVENT_STACKSIZE * 1024, 0);
setup_thread_attr(&waiter_attr, 32 * 1024, 1);
if (logsink == 1) {
log_thread_start(&log_attr);
pthread_attr_destroy(&log_attr);
}
- if (pidfile_create(DEFAULT_PIDFILE, daemon_pid)) {
+ pid_fd = pidfile_create(DEFAULT_PIDFILE, daemon_pid);
+ if (pid_fd < 0) {
condlog(1, "failed to create pidfile");
if (logsink == 1)
log_thread_stop();
exit(1);
}
- running_state = DAEMON_START;
+ post_config_state(DAEMON_START);
-#ifdef USE_SYSTEMD
- sd_notify(0, "STATUS=startup");
-#endif
condlog(2, "--------start up--------");
condlog(2, "read " DEFAULT_CONFIGFILE);
- if (load_config(DEFAULT_CONFIGFILE, udev))
+ conf = load_config(DEFAULT_CONFIGFILE);
+ if (!conf)
goto failed;
+ if (verbosity)
+ conf->verbosity = verbosity;
+ if (bindings_read_only)
+ conf->bindings_read_only = bindings_read_only;
+ if (ignore_new_devs)
+ conf->ignore_new_devs = ignore_new_devs;
+ uxsock_timeout = conf->uxsock_timeout;
+ rcu_assign_pointer(multipath_conf, conf);
+ dm_init(conf->verbosity);
dm_drv_version(conf->version, TGT_MPATH);
- if (init_checkers()) {
+ if (init_checkers(conf->multipath_dir)) {
condlog(0, "failed to initialize checkers");
goto failed;
}
- if (init_prio()) {
+ if (init_prio(conf->multipath_dir)) {
condlog(0, "failed to initialize prioritizers");
goto failed;
}
setscheduler();
set_oom_adj();
- conf->daemon = 1;
dm_udev_set_sync_support(0);
#ifdef USE_SYSTEMD
envp = getenv("WATCHDOG_USEC");
}
#endif
/*
+ * Startup done, invalidate configuration
+ */
+ conf = NULL;
+
+ /*
+ * Signal start of configuration
+ */
+ post_config_state(DAEMON_CONFIGURE);
+
+ init_path_check_interval(vecs);
+
+ /*
* Start uevent listener early to catch events
*/
if ((rc = pthread_create(&uevent_thr, &uevent_attr, ueventloop, udev))) {
condlog(0, "failed to create cli listener: %d", rc);
goto failed;
}
- /*
- * fetch and configure both paths and multipaths
- */
-#ifdef USE_SYSTEMD
- sd_notify(0, "STATUS=configure");
-#endif
- running_state = DAEMON_CONFIGURE;
-
- lock(vecs->lock);
- if (configure(vecs, 1)) {
- unlock(vecs->lock);
- condlog(0, "failure during configuration");
- goto failed;
- }
- unlock(vecs->lock);
/*
* start threads
}
pthread_attr_destroy(&misc_attr);
- running_state = DAEMON_RUNNING;
+ while (running_state != DAEMON_SHUTDOWN) {
+ pthread_cleanup_push(config_cleanup, NULL);
+ pthread_mutex_lock(&config_lock);
+ if (running_state != DAEMON_CONFIGURE &&
+ running_state != DAEMON_SHUTDOWN) {
+ pthread_cond_wait(&config_cond, &config_lock);
+ }
+ pthread_cleanup_pop(1);
+ if (running_state == DAEMON_CONFIGURE) {
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
+ if (!need_to_delay_reconfig(vecs)) {
+ reconfigure(vecs);
+ } else {
+ conf = get_multipath_config();
+ conf->delayed_reconfig = 1;
+ put_multipath_config(conf);
+ }
+ lock_cleanup_pop(vecs->lock);
+ post_config_state(DAEMON_IDLE);
#ifdef USE_SYSTEMD
- sd_notify(0, "READY=1\nSTATUS=running");
+ if (!startup_done) {
+ sd_notify(0, "READY=1");
+ startup_done = 1;
+ }
#endif
+ }
+ }
- /*
- * exit path
- */
- while(sem_wait(&exit_sem) != 0); /* Do nothing */
-
-#ifdef USE_SYSTEMD
- sd_notify(0, "STATUS=shutdown");
-#endif
- running_state = DAEMON_SHUTDOWN;
- lock(vecs->lock);
+ lock(&vecs->lock);
+ conf = get_multipath_config();
if (conf->queue_without_daemon == QUE_NO_DAEMON_OFF)
vector_foreach_slot(vecs->mpvec, mpp, i)
dm_queue_if_no_path(mpp->alias, 0);
+ put_multipath_config(conf);
remove_maps_and_stop_waiters(vecs);
- unlock(vecs->lock);
+ unlock(&vecs->lock);
pthread_cancel(check_thr);
pthread_cancel(uevent_thr);
pthread_cancel(uxlsnr_thr);
pthread_cancel(uevq_thr);
- lock(vecs->lock);
+ pthread_join(check_thr, NULL);
+ pthread_join(uevent_thr, NULL);
+ pthread_join(uxlsnr_thr, NULL);
+ pthread_join(uevq_thr, NULL);
+
+ lock(&vecs->lock);
free_pathvec(vecs->pathvec, FREE_PATHS);
vecs->pathvec = NULL;
- unlock(vecs->lock);
- /* Now all the waitevent threads will start rushing in. */
- while (vecs->lock.depth > 0) {
- sleep (1); /* This is weak. */
- condlog(3, "Have %d wait event checkers threads to de-alloc,"
- " waiting...", vecs->lock.depth);
- }
- pthread_mutex_destroy(vecs->lock.mutex);
- FREE(vecs->lock.mutex);
- vecs->lock.depth = 0;
- vecs->lock.mutex = NULL;
+ unlock(&vecs->lock);
+
+ pthread_mutex_destroy(&vecs->lock.mutex);
FREE(vecs);
vecs = NULL;
* because logging functions like dlog() and dm_write_log()
* reference the config.
*/
- free_config(conf);
- conf = NULL;
+ conf = rcu_dereference(multipath_conf);
+ rcu_assign_pointer(multipath_conf, NULL);
+ call_rcu(&conf->rcu, rcu_free_config);
udev_unref(udev);
udev = NULL;
+ pthread_attr_destroy(&waiter_attr);
#ifdef _DEBUG_
dbg_free_final(NULL);
#endif
#ifdef USE_SYSTEMD
sd_notify(0, "ERRNO=1");
#endif
+ if (pid_fd >= 0)
+ close(pid_fd);
exit(1);
}
int arg;
int err;
int foreground = 0;
+ struct config *conf;
+
+ ANNOTATE_BENIGN_RACE_SIZED(&multipath_conf, sizeof(multipath_conf),
+ "Manipulated through RCU");
+ ANNOTATE_BENIGN_RACE_SIZED(&running_state, sizeof(running_state),
+ "Suppress complaints about unprotected running_state reads");
+ ANNOTATE_BENIGN_RACE_SIZED(&uxsock_timeout, sizeof(uxsock_timeout),
+ "Suppress complaints about this scalar variable");
logsink = 1;
- running_state = DAEMON_INIT;
- dm_init();
if (getuid() != 0) {
fprintf(stderr, "need to be root\n");
strerror(errno));
umask(umask(077) | 022);
- conf = alloc_config();
+ pthread_cond_init_mono(&config_cond);
- if (!conf)
- exit(1);
+ udev = udev_new();
while ((arg = getopt(argc, argv, ":dsv:k::Bn")) != EOF ) {
- switch(arg) {
+ switch(arg) {
case 'd':
foreground = 1;
if (logsink > 0)
!isdigit(optarg[0]))
exit(1);
- conf->verbosity = atoi(optarg);
+ verbosity = atoi(optarg);
break;
case 's':
logsink = -1;
break;
case 'k':
- if (load_config(DEFAULT_CONFIGFILE, udev_new()))
+ conf = load_config(DEFAULT_CONFIGFILE);
+ if (!conf)
exit(1);
- uxclnt(optarg, conf->uxsock_timeout);
+ if (verbosity)
+ conf->verbosity = verbosity;
+ uxsock_timeout = conf->uxsock_timeout;
+ uxclnt(optarg, uxsock_timeout + 100);
+ free_config(conf);
exit(0);
case 'B':
- conf->bindings_read_only = 1;
+ bindings_read_only = 1;
break;
case 'n':
- conf->ignore_new_devs = 1;
+ ignore_new_devs = 1;
break;
default:
- ;
+ fprintf(stderr, "Invalid argument '-%c'\n",
+ optopt);
+ exit(1);
}
}
if (optind < argc) {
char * s = cmd;
char * c = s;
- if (load_config(DEFAULT_CONFIGFILE, udev_new()))
+ conf = load_config(DEFAULT_CONFIGFILE);
+ if (!conf)
exit(1);
+ if (verbosity)
+ conf->verbosity = verbosity;
+ uxsock_timeout = conf->uxsock_timeout;
+ memset(cmd, 0x0, CMDSIZE);
while (optind < argc) {
if (strchr(argv[optind], ' '))
c += snprintf(c, s + CMDSIZE - c, "\"%s\" ", argv[optind]);
optind++;
}
c += snprintf(c, s + CMDSIZE - c, "\n");
- uxclnt(s, conf->uxsock_timeout);
+ uxclnt(s, uxsock_timeout + 100);
+ free_config(conf);
exit(0);
}
rc = pthread_join(thread, NULL);
return 0;
}
-