[dm-devel] [PATCH v3]multipath-tools: prevent unnecessary reinstate of stand-by paths with implicit tags mode and no active array paths
Shiva Krishna
Shiva.Krishna at nimblestorage.com
Sat Feb 27 20:44:22 UTC 2016
multipathd treats "stand-by" path as active(ghost) and reinstate path.This
causes I/O hang issues and lots of "change" udev events in cases where only
stand-by paths are present in multipath map and target supports only
implicit tpgs mode(active/passive arrays)
This can happen during system boot where only stand-by paths are discovered
first and continuous retry of I/O's by dm-multipath and change(failed)
events are hogging multipathd and slowing down the entire boot process with
large number of volumes mapped (~100s).
Selecting path checkers other than tur is not a solution as well, as they
will continuously throw messages saying paths are failed for stand-by
state.
This patch will prevent re-instate of stand-by paths in this situation and
thus prevent unnecessary i/o with failed events during boot or when active
array goes down temporarily.
With this, for targets supporting only alua and implicit tpgs mode,
stand-by paths are not reinstated by path checker unless there is an active
path in the map.
Detailed steps to hit the I/O hang issue even with no_path_retry:
devices {
device {
vendor "Nimble"
product "Server"
path_grouping_policy group_by_prio
prio alua
hardware_handler "1 alua"
path_checker tur
failback immediate
fast_io_fail_tmo 10
no_path_retry 30
path_selector "round-robin 0"
}
}
1. Delete all active paths.
mpathal (2e0176ad6309077166c9ce90033bfa248_1) dm-2 Nimble,Server
size=20G features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
|- 8:0:5:1 sdg 8:96 active ghost running
`- 8:0:6:1 sdh 8:112 active ghost running
2. Issue I/O on mpath with zero active paths.
dd if=/dev/mapper/mpathal of=/dev/null bs=512 count=1 iflag=direct &
Nov 23 15:10:36 hitdev-rhel67 multipathd: mpathal: sdg - tur checker
reports path is in standby state
Nov 23 15:10:36 hitdev-rhel67 multipathd: 8:96: reinstated
Nov 23 15:10:36 hitdev-rhel67 multipathd: mpathal: queue_if_no_path enabled
Nov 23 15:10:36 hitdev-rhel67 multipathd: mpathal: Recovered to normal mode
Nov 23 15:10:36 hitdev-rhel67 multipathd: mpathal: remaining active paths:1
Nov 23 15:10:36 hitdev-rhel67 kernel: sd 8:0:5:1: alua: port group 02
state S non-preferred supports tolusna
Nov 23 15:10:36 hitdev-rhel67 kernel: device-mapper: multipath: Failing
path 8:96.
Nov 23 15:10:36 hitdev-rhel67 multipathd: 8:96: mark as failed
Nov 23 15:10:36 hitdev-rhel67 multipathd: mpathal: Entering recovery mode:
max_retries=20
Nov 23 15:10:36 hitdev-rhel67 multipathd: mpathal: remaining active paths:
0
3. Monitor udev events every 5 seconds.
# udevadm monitor
monitor will print the received events for:
monitor will print the received events for:
UDEV - the event which udev sends out after rule processing
KERNEL - the kernel uevent
KERNEL[1448320131.061837] change /devices/virtual/block/dm-2 (block)
KERNEL[1448320131.064802] change /devices/virtual/block/dm-2 (block)
UDEV [1448320131.082838] change /devices/virtual/block/dm-2 (block)
UDEV [1448320131.102134] change /devices/virtual/block/dm-2 (block)
KERNEL[1448320135.551737] change /devices/virtual/block/dm-2 (block)
KERNEL[1448320135.552701] change /devices/virtual/block/dm-2 (block)
UDEV [1448320135.571634] change /devices/virtual/block/dm-2 (block)
UDEV [1448320135.591017] change /devices/virtual/block/dm-2 (block)
KERNEL[1448320136.553368] change /devices/virtual/block/dm-2 (block)
KERNEL[1448320136.554298] change /devices/virtual/block/dm-2 (block)
UDEV [1448320136.572733] change /devices/virtual/block/dm-2 (block)
UDEV [1448320136.592089] change /devices/virtual/block/dm-2 (block)
KERNEL[1448320140.555389] change /devices/virtual/block/dm-2 (block)
KERNEL[1448320140.556369] change /devices/virtual/block/dm-2 (block)
UDEV [1448320140.574944] change /devices/virtual/block/dm-2 (block)
UDEV [1448320140.594076] change /devices/virtual/block/dm-2 (block)
Signed-off-by: ShivaKrishna Merla <shiva.krishna at nimblestorage.com>
---
libmultipath/propsel.c | 13 ++++++++++++-
libmultipath/structs.h | 1 +
multipathd/main.c | 19 ++++++++++++++++---
3 files changed, 29 insertions(+), 4 deletions(-)
diff --git a/libmultipath/propsel.c b/libmultipath/propsel.c
index f64d5e4..890d4b1 100644
--- a/libmultipath/propsel.c
+++ b/libmultipath/propsel.c
@@ -373,9 +373,11 @@ detect_prio(struct path * pp)
{
int ret;
struct prio *p = &pp->prio;
+ int tpgs = 0;
- if (get_target_port_group_support(pp->fd) <= 0)
+ if ((tpgs = get_target_port_group_support(pp->fd)) <= 0)
return;
+ pp->tpgs = tpgs;
ret = get_target_port_group(pp->fd);
if (ret < 0)
return;
@@ -415,6 +417,15 @@ select_prio (struct path * pp)
prio_get(p, DEFAULT_PRIO, DEFAULT_PRIO_ARGS);
origin = "(internal default)";
out:
+ /*
+ * fetch tpgs mode for alua, if its not already obtained
+ */
+ if (!strncmp(prio_name(p), PRIO_ALUA, PRIO_NAME_LEN)) {
+ int tpgs = 0;
+ if(!pp->tpgs &&
+ (tpgs = get_target_port_group_support(pp->fd)) >= 0)
+ pp->tpgs = tpgs;
+ }
condlog(3, "%s: prio = %s %s", pp->dev, prio_name(p), origin);
condlog(3, "%s: prio args = \"%s\" %s", pp->dev, prio_args(p), origin);
return 0;
diff --git a/libmultipath/structs.h b/libmultipath/structs.h
index 5f68a8e..ef5fb7e 100644
--- a/libmultipath/structs.h
+++ b/libmultipath/structs.h
@@ -193,6 +193,7 @@ struct path {
int detect_prio;
int watch_checks;
int wait_checks;
+ int tpgs;
char * uid_attribute;
char * getuid;
struct prio prio;
diff --git a/multipathd/main.c b/multipathd/main.c
index 04f6d02..b6b0053 100644
--- a/multipathd/main.c
+++ b/multipathd/main.c
@@ -62,6 +62,7 @@ static int use_watchdog;
#include <pgpolicies.h>
#include <uevent.h>
#include <log.h>
+#include "prioritizers/alua_rtpg.h"
#include "main.h"
#include "pidfile.h"
@@ -1161,6 +1162,7 @@ check_path (struct vectors * vecs, struct path * pp)
int new_path_up = 0;
int chkr_new_path_up = 0;
int add_active;
+ int ignore_reinstate = 0;
int oldchkrstate = pp->chkrstate;
if (pp->initialized && !pp->mpp)
@@ -1235,6 +1237,16 @@ check_path (struct vectors * vecs, struct path * pp)
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.
+ */
+ ignore_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;
@@ -1297,7 +1309,7 @@ check_path (struct vectors * vecs, struct path * pp)
pp->watch_checks--;
add_active = 0;
}
- if (reinstate_path(pp, add_active)) {
+ if (!ignore_reinstate && reinstate_path(pp, add_active)) {
condlog(3, "%s: reload map", pp->dev);
ev_add_path(pp, vecs);
pp->tick = 1;
@@ -1316,8 +1328,9 @@ check_path (struct vectors * vecs, struct path * pp)
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) &&
+ !ignore_reinstate) {
/* Clear IO errors */
if (reinstate_path(pp, 0)) {
condlog(3, "%s: reload map", pp->dev);
--
More information about the dm-devel
mailing list