#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 <mpath_persist.h>
+#include <urcu.h>
+#ifdef USE_SYSTEMD
+#include <systemd/sd-daemon.h>
+#endif
+#include <semaphore.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 <pgpolicies.h>
-#include <uevent.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 "pidfile.h"
#include "cli_handlers.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;
-pthread_cond_t exit_cond = PTHREAD_COND_INITIALIZER;
-pthread_mutex_t exit_mutex = PTHREAD_MUTEX_INITIALIZER;
-
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;
/*
* global copy of vecs for use in sig handlers
*/
struct vectors * gvecs;
+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;
- int j;
+ 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)) {
/*
* remove all current maps not allowed by the
/*
* may be just because the device is open
*/
+ if (setup_multipath(vecs, ompp) != 0) {
+ i--;
+ continue;
+ }
if (!vector_alloc_slot(nmpv))
return 1;
vector_set_slot(nmpv, ompp);
- setup_multipath(vecs, ompp);
-
- if ((j = find_slot(ompv, (void *)ompp)) != -1)
- vector_del_slot(ompv, j);
- continue;
+ vector_del_slot(ompv, i);
+ i--;
}
else {
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 ||
- pp->state == PATH_WILD)
+ if (pp->state == PATH_UNCHECKED ||
+ pp->state == PATH_WILD ||
+ pp->state == PATH_DELAYED)
continue;
if ((pp->dmstate == PSTATE_FAILED ||
pp->dmstate == PSTATE_UNDEF) &&
}
static int
-flush_map(struct multipath * mpp, struct vectors * vecs)
+flush_map(struct multipath * mpp, struct vectors * vecs, int nopaths)
{
+ int r;
+
+ if (nopaths)
+ r = dm_flush_map_nopaths(mpp->alias, mpp->deferred_remove);
+ else
+ r = dm_flush_map(mpp->alias);
/*
* clear references to this map before flushing so we can ignore
* the spurious uevent we may generate with the dm_flush_map call below
*/
- if (dm_flush_map(mpp->alias)) {
+ if (r) {
/*
* May not really be an error -- if the map was already flushed
* from the device mapper by dmsetup(8) for instance.
*/
- condlog(0, "%s: can't flush", mpp->alias);
- return 1;
+ if (r == 1)
+ condlog(0, "%s: can't flush", mpp->alias);
+ else {
+ condlog(2, "%s: devmap deferred remove", mpp->alias);
+ mpp->deferred_remove = DEFERRED_REMOVE_IN_PROGRESS;
+ }
+ return r;
}
else {
dm_lib_release();
- condlog(2, "%s: devmap removed", mpp->alias);
+ condlog(2, "%s: map flushed", mpp->alias);
}
orphan_paths(vecs->pathvec, mpp);
return 0;
}
+int
+update_map (struct multipath *mpp, struct vectors *vecs)
+{
+ int retries = 3;
+ char params[PARAMS_SIZE] = {0};
+
+retry:
+ condlog(4, "%s: updating new map", mpp->alias);
+ if (adopt_paths(vecs->pathvec, mpp)) {
+ condlog(0, "%s: failed to adopt paths for new map update",
+ mpp->alias);
+ retries = -1;
+ goto fail;
+ }
+ verify_paths(mpp, vecs);
+ mpp->flush_on_last_del = FLUSH_UNDEF;
+ mpp->action = ACT_RELOAD;
+
+ if (setup_map(mpp, params, PARAMS_SIZE)) {
+ condlog(0, "%s: failed to setup new map in update", mpp->alias);
+ retries = -1;
+ goto fail;
+ }
+ if (domap(mpp, params, 1) <= 0 && retries-- > 0) {
+ condlog(0, "%s: map_udate sleep", mpp->alias);
+ sleep(1);
+ goto retry;
+ }
+ dm_lib_release();
+
+fail:
+ if (setup_multipath(vecs, mpp))
+ return 1;
+
+ sync_map_state(mpp);
+
+ if (retries < 0)
+ condlog(0, "%s: failed reload in new map update", mpp->alias);
+ return 0;
+}
+
static int
uev_add_map (struct uevent * uev, struct vectors * vecs)
{
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 (map_present && dm_type(alias, TGT_MPATH) <= 0) {
+ if (map_present && !dm_is_mpath(alias)) {
condlog(4, "%s: not a multipath map", alias);
return 0;
}
mpp = find_mp_by_alias(vecs->mpvec, 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 */
+ 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 (delayed_reconfig &&
+ !need_to_delay_reconfig(vecs)) {
+ condlog(2, "reconfigure (delayed)");
+ set_config_state(DAEMON_CONFIGURE);
+ return 0;
+ }
+ }
/*
* Not really an error -- we generate our own uevent
* 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);
/*
* now we can register the map
*/
- if (map_present && (mpp = add_map_without_path(vecs, alias))) {
- sync_map_state(mpp);
- condlog(2, "%s: devmap %s registered", alias, dev);
- return 0;
+ if (map_present) {
+ if ((mpp = add_map_without_path(vecs, alias))) {
+ sync_map_state(mpp);
+ condlog(2, "%s: devmap %s registered", alias, dev);
+ return 0;
+ } else {
+ condlog(2, "%s: uev_add_map failed", dev);
+ return 1;
+ }
}
- refwwid = get_refwwid(dev, DEV_DEVMAP, vecs->pathvec);
+ 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();
}
if (!r)
condlog(2, "%s: devmap %s added", alias, dev);
+ else if (r == 2)
+ condlog(2, "%s: uev_add_map %s blacklisted", alias, dev);
else
condlog(0, "%s: uev_add_map %s failed", alias, dev);
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)
{
if (!mpp) {
condlog(2, "%s: devmap not registered, can't remove",
devname);
- return 0;
+ return 1;
}
if (strcmp(mpp->alias, alias)) {
condlog(2, "%s: minor number mismatch (map %d, event %d)",
mpp->alias, mpp->dmi->minor, minor);
- return 0;
+ return 1;
}
- return flush_map(mpp, vecs);
+ return flush_map(mpp, vecs, 0);
}
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(0, "%s: spurious uevent, path already in pathvec",
uev->kernel);
- if (pp->mpp)
- return 0;
- } else {
- /*
- * get path vital state
- */
- if (!(pp = store_pathinfo(vecs->pathvec, conf->hwtable,
- uev->udev, DI_ALL))) {
- condlog(0, "%s: failed to store path info",
- uev->kernel);
- return 1;
+ if (!pp->mpp && !strlen(pp->wwid)) {
+ condlog(3, "%s: reinitialize path", uev->kernel);
+ udev_device_unref(pp->udev);
+ pp->udev = udev_device_ref(uev->udev);
+ 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, need_do_map);
+ else if (r == PATHINFO_SKIPPED) {
+ condlog(3, "%s: remove blacklisted path",
+ uev->kernel);
+ i = find_slot(vecs->pathvec, (void *)pp);
+ if (i != -1)
+ vector_del_slot(vecs->pathvec, i);
+ free_path(pp);
+ } else {
+ condlog(0, "%s: failed to reinitialize path",
+ uev->kernel);
+ ret = 1;
+ }
}
- pp->checkint = conf->checkint;
}
+ lock_cleanup_pop(vecs->lock);
+ if (pp)
+ return ret;
- return (ev_add_path(pp, vecs) != 1)? 0 : 1;
+ /*
+ * get path vital state
+ */
+ 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;
+ put_multipath_config(conf);
+ ret = ev_add_path(pp, vecs, need_do_map);
+ } else {
+ condlog(0, "%s: failed to store path info, "
+ "dropping event",
+ uev->kernel);
+ free_path(pp);
+ ret = 1;
+ }
+ lock_cleanup_pop(vecs->lock);
+ return ret;
}
/*
* returns:
* 0: added
* 1: error
- * 2: blacklisted
*/
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 empty_buff[WWID_SIZE] = {0};
char params[PARAMS_SIZE] = {0};
int retries = 3;
int start_waiter = 0;
+ int ret;
/*
* need path UID to go any further
*/
- if (memcmp(empty_buff, pp->wwid, WWID_SIZE) == 0) {
+ if (strlen(pp->wwid) == 0) {
condlog(0, "%s: failed to get path uid", pp->dev);
goto fail; /* leave path added to pathvec */
}
- if (filter_path(conf, pp) > 0){
- int i = find_slot(vecs->pathvec, (void *)pp);
- if (i != -1)
- vector_del_slot(vecs->pathvec, i);
- free_path(pp);
- return 2;
+ mpp = find_mp_by_wwid(vecs->mpvec, pp->wwid);
+ 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;
}
- mpp = pp->mpp = find_mp_by_wwid(vecs->mpvec, pp->wwid);
+
+ pp->mpp = mpp;
rescan:
if (mpp) {
- if ((!pp->size) || (mpp->size != pp->size)) {
- if (!pp->size)
- condlog(0, "%s: failed to add new path %s, "
- "device size is 0",
- mpp->alias, pp->dev);
- else
- condlog(0, "%s: failed to add new path %s, "
- "device size mismatch",
- mpp->alias, pp->dev);
+ if (pp->size && mpp->size != pp->size) {
+ condlog(0, "%s: failed to add new path %s, "
+ "device size mismatch",
+ mpp->alias, pp->dev);
int i = find_slot(vecs->pathvec, (void *)pp);
if (i != -1)
vector_del_slot(vecs->pathvec, i);
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, NULL);
+ verify_paths(mpp, vecs);
mpp->flush_on_last_del = FLUSH_UNDEF;
mpp->action = ACT_RELOAD;
- }
- else {
- if (!pp->size) {
- condlog(0, "%s: failed to create new map,"
- " device size is 0 ", pp->dev);
- int i = find_slot(vecs->pathvec, (void *)pp);
- if (i != -1)
- vector_del_slot(vecs->pathvec, i);
- free_path(pp);
- return 1;
+ } else {
+ if (!should_multipath(pp, vecs->pathvec)) {
+ orphan_path(pp, "only one path");
+ return 0;
}
-
condlog(4,"%s: creating new map", pp->dev);
if ((mpp = add_map_with_path(vecs, pp, 1))) {
mpp->action = ACT_CREATE;
*/
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
*/
- if (domap(mpp, params) <= 0) {
+retry:
+ ret = domap(mpp, params, 1);
+ if (ret <= 0) {
+ if (ret < 0 && retries-- > 0) {
+ condlog(0, "%s: retry domap for addition of new "
+ "path %s", mpp->alias, pp->dev);
+ sleep(1);
+ goto retry;
+ }
condlog(0, "%s: failed in domap for addition of new "
"path %s", mpp->alias, pp->dev);
/*
* 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;
goto fail_map;
if (retries >= 0) {
- condlog(2, "%s path added to devmap %s", pp->dev, mpp->alias);
+ condlog(2, "%s [%s]: path added to devmap %s",
+ pp->dev, pp->dev_t, mpp->alias);
return 0;
- }
- else
- return 1;
+ } else
+ goto fail;
fail_map:
remove_map(mpp, vecs, 1);
fail:
- orphan_path(pp);
+ orphan_path(pp, "failed to add path");
return 1;
}
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)) {
+ if (!flush_map(mpp, vecs, 1)) {
condlog(2, "%s: removed map after"
" removing all paths",
alias);
" removal of path %s", mpp->alias, pp->dev);
goto fail;
}
+
+ if (mpp->wait_for_udev) {
+ mpp->wait_for_udev = 2;
+ 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);
/*
* update our state from kernel
*/
- if (setup_multipath(vecs, mpp)) {
- goto fail;
- }
+ if (setup_multipath(vecs, mpp))
+ return 1;
sync_map_state(mpp);
- condlog(2, "%s: path removed from map %s",
- pp->dev, mpp->alias);
+ condlog(2, "%s [%s]: path removed from map %s",
+ pp->dev, pp->dev_t, mpp->alias);
}
}
uev_update_path (struct uevent *uev, struct vectors * vecs)
{
int ro, retval = 0;
+ struct path * pp;
+ struct config *conf;
+ int disable_changed_wwids;
+ int needs_reinit = 0;
+
+ conf = get_multipath_config();
+ disable_changed_wwids = conf->disable_changed_wwids;
+ put_multipath_config(conf);
ro = uevent_get_disk_ro(uev);
- if (ro >= 0) {
- struct path * pp;
+ pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ lock(&vecs->lock);
+ pthread_testcancel();
- condlog(2, "%s: update path write_protect to '%d' (uevent)",
- uev->kernel, ro);
- pp = find_path_by_dev(vecs->pathvec, uev->kernel);
- if (!pp) {
- condlog(0, "%s: spurious uevent, path not found",
- uev->kernel);
- return 1;
+ 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->mpp)
- retval = reload_map(vecs, pp->mpp);
- condlog(2, "%s: map %s reloaded (retval %d)",
- uev->kernel, pp->mpp->alias, retval);
+ 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;
+
+ conf = get_multipath_config();
+ retval = alloc_path_with_pathinfo(conf, uev->udev, uev->wwid, flag, NULL);
+ put_multipath_config(conf);
+
+ 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)
{
- block_signal(SIGUSR1, NULL);
- block_signal(SIGHUP, NULL);
+ struct udev *udev = ap;
- if (uevent_listen())
+ 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)
{
- block_signal(SIGUSR1, NULL);
- block_signal(SIGHUP, NULL);
-
+ 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)
{
- block_signal(SIGUSR1, NULL);
- block_signal(SIGHUP, NULL);
-
- 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);
+ set_handler_callback(FORCEQ+DAEMON, cli_force_no_daemon_q);
+ set_handler_callback(RESTOREQ+DAEMON, cli_restore_no_daemon_q);
umask(077);
uxsock_listen(&uxsock_trigger, ap);
-
+ pthread_cleanup_pop(1);
return NULL;
}
-int
-exit_daemon (int status)
-{
- if (status != 0)
- fprintf(stderr, "bad exit status. see daemon.log\n");
-
- condlog(3, "unlink pidfile");
- unlink(DEFAULT_PIDFILE);
-
- pthread_mutex_lock(&exit_mutex);
- pthread_cond_signal(&exit_cond);
- pthread_mutex_unlock(&exit_mutex);
-
- return status;
-}
-
-const char *
-daemon_status(void)
+void
+exit_daemon (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
/*
* caller must have locked the path list before calling that function
*/
-static void
+static int
reinstate_path (struct path * pp, int add_active)
{
+ int ret = 0;
+
if (!pp->mpp)
- return;
+ return 0;
- if (dm_reinstate_path(pp->mpp->alias, pp->dev_t))
+ if (dm_reinstate_path(pp->mpp->alias, pp->dev_t)) {
condlog(0, "%s: reinstate failed", pp->dev_t);
- else {
+ ret = 1;
+ } else {
condlog(2, "%s: reinstated", pp->dev_t);
if (add_active)
update_queue_mode_add_path(pp->mpp);
}
+ return ret;
}
static void
}
}
+/* This is called after a path has started working again. It the multipath
+ * device for this path uses the followover failback type, and this is the
+ * best pathgroup, and this is the first path in the pathgroup to come back
+ * up, then switch to this pathgroup */
+static int
+followover_should_failback(struct path * pp)
+{
+ struct pathgroup * pgp;
+ struct path *pp1;
+ int i;
+
+ if (pp->mpp->pgfailback != -FAILBACK_FOLLOWOVER ||
+ !pp->mpp->pg || !pp->pgindex ||
+ pp->pgindex != pp->mpp->bestpg)
+ return 0;
+
+ pgp = VECTOR_SLOT(pp->mpp->pg, pp->pgindex - 1);
+ vector_foreach_slot(pgp->paths, pp1, i) {
+ if (pp1 == pp)
+ continue;
+ if (pp1->chkrstate != PATH_DOWN && pp1->chkrstate != PATH_SHAKY)
+ return 0;
+ }
+ return 1;
+}
+
+static void
+missing_uev_wait_tick(struct vectors *vecs)
+{
+ struct multipath * mpp;
+ unsigned int i;
+ 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) {
+ timed_out = 1;
+ condlog(0, "%s: timeout waiting on creation uevent. enabling reloads", mpp->alias);
+ if (mpp->wait_for_udev > 1 && update_map(mpp, vecs)) {
+ /* update_map removed map */
+ i--;
+ continue;
+ }
+ mpp->wait_for_udev = 0;
+ }
+ }
+
+ 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)");
+ set_config_state(DAEMON_CONFIGURE);
+ }
+}
+
static void
defered_failback_tick (vector mpvec)
{
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;
+ return 1;
+}
+
+int update_path_groups(struct multipath *mpp, struct vectors *vecs, int refresh)
+{
+ if (reload_map(vecs, mpp, refresh, 1))
+ return 1;
+
+ dm_lib_release();
+ if (setup_multipath(vecs, mpp) != 0)
+ return 1;
+ sync_map_state(mpp);
+
+ 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 (pp->priority == oldpriority)
+ /* 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;
- return 1;
-}
-int update_path_groups(struct multipath *mpp, struct vectors *vecs, int refresh)
-{
- int i;
- struct path * pp;
- char params[PARAMS_SIZE];
+ if (pp->path_failures == 0)
+ pp->san_path_err_forget_rate = pp->mpp->san_path_err_forget_rate;
- update_mpp_paths(mpp, vecs->pathvec);
- if (refresh) {
- vector_foreach_slot (mpp->paths, pp, i)
- pathinfo(pp, conf->hwtable, DI_PRIO);
- }
- params[0] = '\0';
- if (setup_map(mpp, params, PARAMS_SIZE))
- return 1;
+ pp->path_failures++;
- mpp->action = ACT_RELOAD;
- if (domap(mpp, params) <= 0) {
- condlog(0, "%s: failed to update map : %s", mpp->alias,
- strerror(errno));
+ /* 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;
}
- dm_lib_release();
- if (setup_multipath(vecs, mpp) != 0)
- return 1;
- sync_map_state(mpp);
+reinstate_path:
+ pp->path_failures = 0;
+ pp->disable_reinstate = 0;
+ pp->san_path_err_forget_rate = 0;
return 0;
}
-void
-check_path (struct vectors * vecs, struct path * pp)
+/*
+ * 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, int ticks)
{
int newstate;
int new_path_up = 0;
+ int chkr_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->mpp)
- return;
-
- if (pp->tick && --pp->tick)
- return; /* don't check this path yet */
+ 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 < retrigger_tries) {
+ condlog(2, "%s: triggering change event to reinitialize",
+ pp->dev);
+ pp->initialized = INIT_REQUESTED_UDEV;
+ pp->retriggers++;
+ 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_UP)
- newstate = get_state(pp, 1);
+ /*
+ * Wait for uevent for removed paths;
+ * some LLDDs like zfcp keep paths unavailable
+ * without sending uevents.
+ */
+ if (newstate == PATH_REMOVED)
+ newstate = PATH_DOWN;
+
+ 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);
- return;
+ 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);
+ 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;
}
/*
* Async IO in flight. Keep the previous path state
*/
if (newstate == PATH_PENDING) {
pp->tick = 1;
- return;
+ return 0;
}
/*
* Synchronize with kernel state
*/
- if (update_multipath_strings(pp->mpp, vecs->pathvec)) {
- condlog(1, "%s: Could not synchronize with kernel state\n",
+ if (update_multipath_strings(pp->mpp, vecs->pathvec, 1)) {
+ condlog(1, "%s: Could not synchronize with kernel state",
pp->dev);
pp->dmstate = PSTATE_UNDEF;
}
+ /* if update_multipath_strings orphaned the path, quit early */
+ if (!pp->mpp)
+ 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->nr_active > 0) {
+ pp->state = PATH_DELAYED;
+ pp->wait_checks--;
+ return 1;
+ } else
+ pp->wait_checks = 0;
+ }
+
+ /*
+ * don't reinstate failed path, if its in stand-by
+ * and if target supports only implicit tpgs mode.
+ * this will prevent unnecessary i/o by dm on stand-by
+ * paths if there are no other active paths in map.
+ */
+ disable_reinstate = (newstate == PATH_GHOST &&
+ pp->mpp->nr_active == 0 &&
+ pp->tpgs == TPGS_IMPLICIT) ? 1 : 0;
+
+ pp->chkrstate = newstate;
if (newstate != pp->state) {
int oldstate = pp->state;
pp->state = newstate;
+
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
*/
if (oldstate == PATH_UP ||
- oldstate == PATH_GHOST)
+ oldstate == PATH_GHOST) {
fail_path(pp, 1);
- else
+ if (pp->mpp->delay_wait_checks > 0 &&
+ pp->watch_checks > 0) {
+ pp->wait_checks = pp->mpp->delay_wait_checks;
+ pp->watch_checks = 0;
+ }
+ }else
fail_path(pp, 0);
/*
pp->mpp->failback_tick = 0;
pp->mpp->stat_path_failures++;
- return;
+ repair_path(pp);
+ return 1;
}
if(newstate == PATH_UP || newstate == PATH_GHOST){
* reinstate this path
*/
if (oldstate != PATH_UP &&
- oldstate != PATH_GHOST)
- reinstate_path(pp, 1);
- else
- reinstate_path(pp, 0);
-
+ oldstate != PATH_GHOST) {
+ if (pp->mpp->delay_watch_checks > 0)
+ pp->watch_checks = pp->mpp->delay_watch_checks;
+ add_active = 1;
+ } else {
+ if (pp->watch_checks > 0)
+ pp->watch_checks--;
+ add_active = 0;
+ }
+ if (!disable_reinstate && reinstate_path(pp, add_active)) {
+ condlog(3, "%s: reload map", pp->dev);
+ ev_add_path(pp, vecs, 1);
+ pp->tick = 1;
+ return 0;
+ }
new_path_up = 1;
+ if (oldchkrstate != PATH_UP && oldchkrstate != PATH_GHOST)
+ chkr_new_path_up = 1;
+
/*
* if at least one path is up in a group, and
* the group is disabled, re-enable it
enable_group(pp);
}
else if (newstate == PATH_UP || newstate == PATH_GHOST) {
- if (pp->dmstate == PSTATE_FAILED ||
- pp->dmstate == PSTATE_UNDEF) {
+ if ((pp->dmstate == PSTATE_FAILED ||
+ pp->dmstate == PSTATE_UNDEF) &&
+ !disable_reinstate) {
/* Clear IO errors */
- reinstate_path(pp, 0);
+ if (reinstate_path(pp, 0)) {
+ condlog(3, "%s: reload map", pp->dev);
+ ev_add_path(pp, vecs, 1);
+ pp->tick = 1;
+ return 0;
+ }
} else {
+ unsigned int max_checkint;
LOG_MSG(4, checker_message(&pp->checker));
- /*
- * double the next check delay.
- * max at conf->max_checkint
- */
- if (pp->checkint < (conf->max_checkint / 2))
- pp->checkint = 2 * pp->checkint;
- else
- 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 < (max_checkint / 2))
+ pp->checkint = 2 * pp->checkint;
+ else
+ pp->checkint = max_checkint;
+ condlog(4, "%s: delay next check %is",
+ pp->dev_t, pp->checkint);
+ }
+ if (pp->watch_checks > 0)
+ pp->watch_checks--;
pp->tick = pp->checkint;
- condlog(4, "%s: delay next check %is",
- pp->dev_t, pp->tick);
}
}
else if (newstate == PATH_DOWN) {
- if (conf->log_checker_err == LOG_CHKR_ERR_ONCE)
+ 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;
/*
* path prio refreshing
*/
(new_path_up || pp->mpp->failback_tick <= 0))
pp->mpp->failback_tick =
pp->mpp->pgfailback + 1;
- else if (pp->mpp->pgfailback == -FAILBACK_IMMEDIATE)
+ else if (pp->mpp->pgfailback == -FAILBACK_IMMEDIATE ||
+ (chkr_new_path_up && followover_should_failback(pp)))
switch_pathgroup(pp->mpp);
}
+ 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 *
struct path *pp;
int count = 0;
unsigned int i;
- sigset_t old;
+ 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) {
- block_signal(SIGHUP, &old);
+ struct timespec diff_time, start_time, end_time;
+ int num_paths = 0, ticks = 0, signo, strict_timing, rc = 0;
+ sigset_t mask;
+
+ if (clock_gettime(CLOCK_MONOTONIC, &start_time) != 0)
+ start_time.tv_sec = 0;
+ 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
+ rc = set_config_state(DAEMON_RUNNING);
+ if (rc == ETIMEDOUT) {
+ condlog(4, "timeout waiting for DAEMON_IDLE");
+ continue;
+ }
+
pthread_cleanup_push(cleanup_lock, &vecs->lock);
- lock(vecs->lock);
+ lock(&vecs->lock);
pthread_testcancel();
- condlog(4, "tick");
-
- if (vecs->pathvec) {
- vector_foreach_slot (vecs->pathvec, pp, i) {
- check_path(vecs, pp);
- }
- }
- if (vecs->mpvec) {
- defered_failback_tick(vecs->mpvec);
- retry_count_tick(vecs->mpvec);
+ 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);
- pthread_sigmask(SIG_SETMASK, &old, NULL);
- sleep(1);
+ diff_time.tv_nsec = 0;
+ if (start_time.tv_sec &&
+ 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;
+ }
+ }
}
+ pthread_cleanup_pop(1);
return NULL;
}
struct multipath * mpp;
struct path * pp;
vector mpvec;
- int i;
+ 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
*/
- path_discovery(vecs->pathvec, conf, DI_ALL);
+ 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();
sync_maps_state(mpvec);
vector_foreach_slot(mpvec, mpp, i){
+ remember_wwid(mpp->wwid);
update_map_pr(mpp);
}
* start dm event waiter threads for these new maps
*/
vector_foreach_slot(vecs->mpvec, mpp, i) {
- if (setup_multipath(vecs, mpp))
+ if (setup_multipath(vecs, mpp)) {
+ i--;
+ continue;
+ }
+ if (start_waiters) {
+ if (start_waiter_thread(mpp, vecs)) {
+ remove_map(mpp, vecs, 1);
+ i--;
+ }
+ }
+ }
+ return 0;
+}
+
+int
+need_to_delay_reconfig(struct vectors * vecs)
+{
+ struct multipath *mpp;
+ int i;
+
+ if (!VECTOR_SIZE(vecs->mpvec))
+ return 0;
+
+ vector_foreach_slot(vecs->mpvec, mpp, i) {
+ if (mpp->wait_for_udev)
return 1;
- if (start_waiters)
- if (start_waiter_thread(mpp, vecs))
- return 1;
}
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;
+
+ 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;
- if (!load_config(DEFAULT_CONFIGFILE)) {
- conf->verbosity = old->verbosity;
- conf->daemon = 1;
- configure(vecs, 1);
- free_config(old);
- retval = 0;
+ /* Re-read any timezone changes */
+ tzset();
+
+ 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;
- return retval;
+ old = rcu_dereference(multipath_conf);
+ rcu_assign_pointer(multipath_conf, conf);
+ call_rcu(&old->rcu, rcu_free_config);
+
+ 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 *
return (osig.sa_handler);
}
+void
+handle_signals(void)
+{
+ 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)");
+ pthread_mutex_lock(&logq_lock);
+ log_reset("multipathd");
+ pthread_mutex_unlock(&logq_lock);
+ }
+ exit_sig = 0;
+ reconfig_sig = 0;
+ log_reset_sig = 0;
+}
+
static void
sighup (int sig)
{
- condlog(2, "reconfigure (SIGHUP)");
-
- if (running_state != DAEMON_RUNNING)
- return;
-
- lock(gvecs->lock);
- reconfigure(gvecs);
- unlock(gvecs->lock);
-
-#ifdef _DEBUG_
- dbg_free_final(NULL);
-#endif
+ reconfig_sig = 1;
}
static void
sigend (int sig)
{
- exit_daemon(0);
+ exit_sig = 1;
}
static void
sigusr1 (int sig)
{
- condlog(3, "SIGUSR1 received");
+ log_reset_sig = 1;
+}
+
+static void
+sigusr2 (int sig)
+{
+ condlog(3, "SIGUSR2 received");
}
static void
signal_init(void)
{
+ sigset_t set;
+
+ sigemptyset(&set);
+ sigaddset(&set, SIGUSR2);
+ 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
#endif
FILE *fp;
struct stat st;
+ char *envp;
+ envp = getenv("OOMScoreAdjust");
+ if (envp) {
+ condlog(3, "Using systemd provided OOMScoreAdjust");
+ return;
+ }
do {
if (stat(file, &st) == 0){
fp = fopen(file, "w");
strerror(errno));
return;
}
+#ifdef OOM_ADJUST_MIN
file = "/proc/self/oom_adj";
score = OOM_ADJUST_MIN;
+#else
+ retry = 0;
+#endif
} while (retry--);
condlog(0, "couldn't adjust oom score");
}
child (void * param)
{
pthread_t check_thr, uevent_thr, uxlsnr_thr, uevq_thr;
- pthread_attr_t log_attr, misc_attr;
+ pthread_attr_t log_attr, misc_attr, uevent_attr;
struct vectors * vecs;
struct multipath * mpp;
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);
+ signal_init();
+ rcu_init();
- setup_thread_attr(&misc_attr, 64 * 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) {
+ if (logsink == 1) {
setup_thread_attr(&log_attr, 64 * 1024, 0);
log_thread_start(&log_attr);
pthread_attr_destroy(&log_attr);
}
+ 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);
condlog(2, "--------start up--------");
condlog(2, "read " DEFAULT_CONFIGFILE);
- if (load_config(DEFAULT_CONFIGFILE))
- exit(1);
-
- if (init_checkers()) {
+ 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(conf->multipath_dir)) {
condlog(0, "failed to initialize checkers");
- exit(1);
+ goto failed;
}
- if (init_prio()) {
+ if (init_prio(conf->multipath_dir)) {
condlog(0, "failed to initialize prioritizers");
- exit(1);
+ goto failed;
}
setlogmask(LOG_UPTO(conf->verbosity + 3));
- if (conf->max_fds) {
+ envp = getenv("LimitNOFILE");
+
+ if (envp) {
+ condlog(2,"Using systemd provided open fds limit of %s", envp);
+ } else if (conf->max_fds) {
struct rlimit fd_limit;
if (getrlimit(RLIMIT_NOFILE, &fd_limit) < 0) {
- condlog(0, "can't get open fds limit: %s\n",
+ condlog(0, "can't get open fds limit: %s",
strerror(errno));
fd_limit.rlim_cur = 0;
fd_limit.rlim_max = 0;
fd_limit.rlim_max = conf->max_fds;
if (setrlimit(RLIMIT_NOFILE, &fd_limit) < 0) {
condlog(0, "can't set open fds limit to "
- "%lu/%lu : %s\n",
+ "%lu/%lu : %s",
fd_limit.rlim_cur, fd_limit.rlim_max,
strerror(errno));
} else {
- condlog(3, "set open fds limit to %lu/%lu\n",
+ condlog(3, "set open fds limit to %lu/%lu",
fd_limit.rlim_cur, fd_limit.rlim_max);
}
}
}
- signal_init();
+ vecs = gvecs = init_vecs();
+ if (!vecs)
+ goto failed;
+
setscheduler();
set_oom_adj();
- vecs = gvecs = init_vecs();
- if (!vecs)
- exit(1);
+ dm_udev_set_sync_support(0);
+#ifdef USE_SYSTEMD
+ envp = getenv("WATCHDOG_USEC");
+ if (envp && sscanf(envp, "%lu", &checkint) == 1) {
+ /* Value is in microseconds */
+ conf->max_checkint = checkint / 1000000;
+ /* Rescale checkint */
+ if (conf->checkint > conf->max_checkint)
+ conf->checkint = conf->max_checkint;
+ else
+ conf->checkint = conf->max_checkint / 4;
+ condlog(3, "enabling watchdog, interval %d max %d",
+ conf->checkint, conf->max_checkint);
+ use_watchdog = conf->checkint;
+ }
+#endif
+ /*
+ * Startup done, invalidate configuration
+ */
+ conf = NULL;
+
+ /*
+ * Signal start of configuration
+ */
+ post_config_state(DAEMON_CONFIGURE);
+
+ init_path_check_interval(vecs);
- conf->daemon = 1;
- udev_set_sync_support(0);
/*
* Start uevent listener early to catch events
*/
- if ((rc = pthread_create(&uevent_thr, &misc_attr, ueventloop, vecs))) {
+ if ((rc = pthread_create(&uevent_thr, &uevent_attr, ueventloop, udev))) {
condlog(0, "failed to create uevent thread: %d", rc);
- exit(1);
+ goto failed;
}
+ pthread_attr_destroy(&uevent_attr);
if ((rc = pthread_create(&uxlsnr_thr, &misc_attr, uxlsnrloop, vecs))) {
condlog(0, "failed to create cli listener: %d", rc);
- exit(1);
- }
- /*
- * fetch and configure both paths and multipaths
- */
- lock(vecs->lock);
- running_state = DAEMON_CONFIGURE;
-
- if (configure(vecs, 1)) {
- unlock(vecs->lock);
- condlog(0, "failure during configuration");
- exit(1);
+ goto failed;
}
- unlock(vecs->lock);
/*
* start threads
*/
if ((rc = pthread_create(&check_thr, &misc_attr, checkerloop, vecs))) {
condlog(0,"failed to create checker loop thread: %d", rc);
- exit(1);
+ goto failed;
}
if ((rc = pthread_create(&uevq_thr, &misc_attr, uevqloop, vecs))) {
condlog(0, "failed to create uevent dispatcher: %d", rc);
- exit(1);
+ goto failed;
}
pthread_attr_destroy(&misc_attr);
- pthread_mutex_lock(&exit_mutex);
- /* Startup complete, create logfile */
- if (pidfile_create(DEFAULT_PIDFILE, daemon_pid))
- /* Ignore errors, we can live without */
- condlog(1, "failed to create pidfile");
-
- running_state = DAEMON_RUNNING;
- pthread_cond_wait(&exit_cond, &exit_mutex);
+ 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
+ if (!startup_done) {
+ sd_notify(0, "READY=1");
+ startup_done = 1;
+ }
+#endif
+ }
+ }
- /*
- * exit path
- */
- running_state = DAEMON_SHUTDOWN;
- block_signal(SIGHUP, NULL);
- 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..\n", 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;
condlog(2, "--------shut down-------");
- if (logsink)
+ if (logsink == 1)
log_thread_stop();
/*
* 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=0");
+#endif
exit(0);
+
+failed:
+#ifdef USE_SYSTEMD
+ sd_notify(0, "ERRNO=1");
+#endif
+ if (pid_fd >= 0)
+ close(pid_fd);
+ exit(1);
}
static int
extern int optind;
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, ":dv:k::")) != EOF ) {
- switch(arg) {
+ while ((arg = getopt(argc, argv, ":dsv:k::Bn")) != EOF ) {
+ switch(arg) {
case 'd':
- logsink = 0;
+ foreground = 1;
+ if (logsink > 0)
+ logsink = 0;
//debug=1; /* ### comment me out ### */
break;
case 'v':
!isdigit(optarg[0]))
exit(1);
- conf->verbosity = atoi(optarg);
+ verbosity = atoi(optarg);
+ break;
+ case 's':
+ logsink = -1;
break;
case 'k':
- uxclnt(optarg);
+ conf = load_config(DEFAULT_CONFIGFILE);
+ if (!conf)
+ exit(1);
+ if (verbosity)
+ conf->verbosity = verbosity;
+ uxsock_timeout = conf->uxsock_timeout;
+ uxclnt(optarg, uxsock_timeout + 100);
+ free_config(conf);
exit(0);
+ case 'B':
+ bindings_read_only = 1;
+ break;
+ case 'n':
+ ignore_new_devs = 1;
+ break;
default:
- ;
+ fprintf(stderr, "Invalid argument '-%c'\n",
+ optopt);
+ exit(1);
}
}
if (optind < argc) {
char * s = cmd;
char * c = s;
+ 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);
+ uxclnt(s, uxsock_timeout + 100);
+ free_config(conf);
exit(0);
}
- if (!logsink)
+ if (foreground) {
+ if (!isatty(fileno(stdout)))
+ setbuf(stdout, NULL);
err = 0;
- else
+ daemon_pid = getpid();
+ } else
err = daemonize();
if (err < 0)
resp = mpath_alloc_prin_response(MPATH_PRIN_RKEY_SA);
if (!resp){
- condlog(0,"%s Alloc failed for prin response \n", pp->dev);
+ condlog(0,"%s Alloc failed for prin response", pp->dev);
return NULL;
}
ret = prin_do_scsi_ioctl(pp->dev, MPATH_PRIN_RKEY_SA, resp, 0);
if (ret != MPATH_PR_SUCCESS )
{
- condlog(0,"%s : pr in read keys service action failed. Error=%d\n", pp->dev, ret);
+ condlog(0,"%s : pr in read keys service action failed. Error=%d", pp->dev, ret);
goto out;
}
- condlog(3, " event pr=%d addlen=%d\n",resp->prin_descriptor.prin_readkeys.prgeneration,
+ condlog(3, " event pr=%d addlen=%d",resp->prin_descriptor.prin_readkeys.prgeneration,
resp->prin_descriptor.prin_readkeys.additional_length );
if (resp->prin_descriptor.prin_readkeys.additional_length == 0 )
isFound =0;
for (i = 0; i < resp->prin_descriptor.prin_readkeys.additional_length/8; i++ )
{
- condlog(2, "PR IN READKEYS[%d] reservation key:\n",i);
+ condlog(2, "PR IN READKEYS[%d] reservation key:",i);
dumpHex((char *)&resp->prin_descriptor.prin_readkeys.key_list[i*8], 8 , -1);
if (!memcmp(mpp->reservation_key, &resp->prin_descriptor.prin_readkeys.key_list[i*8], 8))
{
if (!isFound)
{
condlog(0, "%s: Either device not registered or ", pp->dev);
- condlog(0, "host is not authorised for registration. Skip path\n");
+ condlog(0, "host is not authorised for registration. Skip path");
ret = MPATH_PR_OTHER;
goto out;
}
}
param->num_transportid = 0;
- condlog(3, "device %s:%s \n", pp->dev, pp->mpp->wwid);
+ condlog(3, "device %s:%s", pp->dev, pp->mpp->wwid);
ret = prout_do_scsi_ioctl(pp->dev, MPATH_PROUT_REG_IGN_SA, 0, 0, param, 0);
if (ret != MPATH_PR_SUCCESS )
{
- condlog(0,"%s: Reservation registration failed. Error: %d\n", pp->dev, ret);
+ condlog(0,"%s: Reservation registration failed. Error: %d", pp->dev, ret);
}
mpp->prflag = 1;
rc = pthread_create(&thread, NULL , mpath_pr_event_handler_fn, pp);
if (rc) {
- condlog(0, "%s: ERROR; return code from pthread_create() is %d\n", pp->dev, rc);
+ condlog(0, "%s: ERROR; return code from pthread_create() is %d", pp->dev, rc);
return -1;
}
pthread_attr_destroy(&attr);
rc = pthread_join(thread, NULL);
return 0;
}
-