From: mwilck@suse.com
To: Christophe Varoqui <christophe.varoqui@opensvc.com>,
Benjamin Marzinski <bmarzins@redhat.com>
Cc: dm-devel@redhat.com, Martin Wilck <mwilck@suse.com>
Subject: [PATCH 26/35] libmultipath: io_err_stat_handle_pathfail(): less error conditions
Date: Thu, 9 Jul 2020 12:16:11 +0200 [thread overview]
Message-ID: <20200709101620.6786-27-mwilck@suse.com> (raw)
In-Reply-To: <20200709101620.6786-1-mwilck@suse.com>
From: Martin Wilck <mwilck@suse.com>
io_err_stat_pathfail() returns an error if marginal path checking is
disabled, and on various other conditions which aren't runtime errors.
Fix that. Also, check the validity of parameters when the configuration
is read, and not on every call.
Signed-off-by: Martin Wilck <mwilck@suse.com>
---
libmultipath/defaults.h | 2 ++
libmultipath/io_err_stat.c | 25 ++++++++-----------------
libmultipath/propsel.c | 6 ++++++
3 files changed, 16 insertions(+), 17 deletions(-)
diff --git a/libmultipath/defaults.h b/libmultipath/defaults.h
index 01a501b..0574e8f 100644
--- a/libmultipath/defaults.h
+++ b/libmultipath/defaults.h
@@ -31,6 +31,8 @@
#define DEFAULT_DEFERRED_REMOVE DEFERRED_REMOVE_OFF
#define DEFAULT_DELAY_CHECKS NU_NO
#define DEFAULT_ERR_CHECKS NU_NO
+/* half of minimum value for marginal_path_err_sample_time */
+#define IOTIMEOUT_SEC 60
#define DEFAULT_UEVENT_STACKSIZE 256
#define DEFAULT_RETRIGGER_DELAY 10
#define DEFAULT_RETRIGGER_TRIES 3
diff --git a/libmultipath/io_err_stat.c b/libmultipath/io_err_stat.c
index 1b9cd6c..58bc1dd 100644
--- a/libmultipath/io_err_stat.c
+++ b/libmultipath/io_err_stat.c
@@ -35,7 +35,6 @@
#include "time-util.h"
#include "io_err_stat.h"
-#define IOTIMEOUT_SEC 60
#define TIMEOUT_NO_IO_NSEC 10000000 /*10ms = 10000000ns*/
#define FLAKY_PATHFAIL_THRESHOLD 2
#define CONCUR_NR_EVENT 32
@@ -301,30 +300,22 @@ int io_err_stat_handle_pathfail(struct path *path)
struct timespec curr_time;
if (uatomic_read(&io_err_thread_running) == 0)
- return 1;
+ return 0;
if (path->io_err_disable_reinstate) {
io_err_stat_log(3, "%s: reinstate is already disabled",
path->dev);
- return 1;
+ return 0;
}
if (path->io_err_pathfail_cnt < 0)
- return 1;
+ return 0;
if (!path->mpp)
- return 1;
- if (path->mpp->marginal_path_double_failed_time <= 0 ||
- path->mpp->marginal_path_err_sample_time <= 0 ||
- path->mpp->marginal_path_err_recheck_gap_time <= 0 ||
- path->mpp->marginal_path_err_rate_threshold < 0) {
- io_err_stat_log(4, "%s: parameter not set", path->mpp->alias);
- return 1;
- }
- if (path->mpp->marginal_path_err_sample_time < (2 * IOTIMEOUT_SEC)) {
- io_err_stat_log(2, "%s: marginal_path_err_sample_time should not less than %d",
- path->mpp->alias, 2 * IOTIMEOUT_SEC);
- return 1;
- }
+ return 0;
+
+ if (!marginal_path_check_enabled(path->mpp))
+ return 0;
+
/*
* The test should only be started for paths that have failed
* repeatedly in a certain time frame, so that we have reason
diff --git a/libmultipath/propsel.c b/libmultipath/propsel.c
index d362beb..2233527 100644
--- a/libmultipath/propsel.c
+++ b/libmultipath/propsel.c
@@ -1066,6 +1066,12 @@ int select_marginal_path_err_sample_time(struct config *conf, struct multipath *
mp_set_conf(marginal_path_err_sample_time);
mp_set_default(marginal_path_err_sample_time, DEFAULT_ERR_CHECKS);
out:
+ if (mp->marginal_path_err_sample_time > 0 &&
+ mp->marginal_path_err_sample_time < 2 * IOTIMEOUT_SEC) {
+ condlog(2, "%s: configuration error: marginal_path_err_sample_time must be >= %d",
+ mp->alias, 2 * IOTIMEOUT_SEC);
+ mp->marginal_path_err_sample_time = 2 * IOTIMEOUT_SEC;
+ }
if (print_off_int_undef(buff, 12, mp->marginal_path_err_sample_time)
!= 0)
condlog(3, "%s: marginal_path_err_sample_time = %s %s",
--
2.26.2
next prev parent reply other threads:[~2020-07-09 10:16 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-09 10:15 [PATCH 00/35] multipath-tools series part I: minor changes mwilck
2020-07-09 10:15 ` [PATCH 01/35] multipath-tools tests/util: separate group for bitmask tests mwilck
2020-07-09 10:15 ` [PATCH 02/35] multipath-tools tests/directio: fix missing initializers mwilck
2020-07-09 10:15 ` [PATCH 03/35] tests: __wrap_dlog: use check_expected() mwilck
2020-07-09 10:15 ` [PATCH 04/35] multipath tools tests: add strchop() test mwilck
2020-07-09 10:15 ` [PATCH 05/35] libmultipath: improve strchop() mwilck
2020-07-09 10:15 ` [PATCH 06/35] multipath-tools tests: add test for devt2devname mwilck
2020-07-09 10:15 ` [PATCH 07/35] libmultipath: devt2devname(): simplify using libudev mwilck
2020-07-09 10:15 ` [PATCH 08/35] libmultipath: create bitfield abstraction mwilck
2020-07-16 21:17 ` Benjamin Marzinski
2020-08-04 15:04 ` Martin Wilck
2020-08-04 15:18 ` Martin Wilck
2020-08-04 16:26 ` Benjamin Marzinski
2020-08-04 19:35 ` Martin Wilck
2020-08-10 18:05 ` Benjamin Marzinski
2020-08-10 18:59 ` Martin Wilck
2020-07-09 10:15 ` [PATCH 09/35] libmultipath: use bitfields in group_by_match() mwilck
2020-07-09 10:15 ` [PATCH 10/35] libmultipath: util: constify function arguments mwilck
2020-07-09 10:15 ` [PATCH 11/35] multipath-tools tests: add unit tests for strlcat mwilck
2020-07-09 10:15 ` [PATCH 12/35] libmultipath: strlcpy()/strlcat(): use restrict qualifier mwilck
2020-07-16 22:18 ` Benjamin Marzinski
2020-08-04 15:36 ` Martin Wilck
2020-08-04 17:29 ` Benjamin Marzinski
2020-08-04 20:15 ` Martin Wilck
2020-07-09 10:15 ` [PATCH 13/35] libmultipath: constify blacklist code mwilck
2020-07-09 10:15 ` [PATCH 14/35] libmultipath: rlookup_binding(): remove newline in log message mwilck
2020-07-09 10:16 ` [PATCH 15/35] libmultipath: fix missing initializer warning from clang 3.9 mwilck
2020-07-09 10:16 ` [PATCH 16/35] libmultipath: fix gcc -Wstringop-overflow warning mwilck
2020-07-09 10:16 ` [PATCH 17/35] libmultipath: remove uevent listener failback mwilck
2020-07-09 10:16 ` [PATCH 18/35] libmultipath: uevent: use static linkage where possible mwilck
2020-07-09 10:16 ` [PATCH 19/35] libmultipath: uevent: inline trivial functions mwilck
2020-07-09 10:16 ` [PATCH 20/35] libmultipath: decrease log level of "SCSI target" log message mwilck
2020-07-09 10:16 ` [PATCH 21/35] libmultipath: get_udev_uid(): more appropriate error code mwilck
2020-07-09 10:16 ` [PATCH 22/35] libmultipath: get_uid(): improve log message on udev failure mwilck
2020-07-09 10:16 ` [PATCH 23/35] libmultipath: make sysfs_pathinfo() static and use const mwilck
2020-07-09 10:16 ` [PATCH 24/35] libmultipath: pathinfo(): improve a log message mwilck
2020-07-09 10:16 ` [PATCH 25/35] libmultipath: pathinfo(): don't filter emtpy devnode names mwilck
2020-07-09 10:16 ` mwilck [this message]
2020-07-09 10:16 ` [PATCH 27/35] libmultipath: improve libdm logging mwilck
2020-07-09 10:16 ` [PATCH 28/35] libmultipath: snprint_devices(): use udev_enumerate mwilck
2020-07-09 10:16 ` [PATCH 29/35] libmultipath: snprint_devices(): print hidden/foreign status mwilck
2020-07-09 10:16 ` [PATCH 30/35] libmultipath: alloc_path(): initialize pp->initialized mwilck
2020-07-09 10:16 ` [PATCH 31/35] libmultipath: alloc_path_with_pathinfo(): treat devname overflow as error mwilck
2020-07-09 10:16 ` [PATCH 32/35] libmultipath: log table params in addmap() mwilck
2020-07-09 10:16 ` [PATCH 33/35] multipathd: remove set_multipath_wwid() mwilck
2020-07-09 10:16 ` [PATCH 34/35] kpartx: print an error message if removing a partition fails mwilck
2020-07-09 10:16 ` [PATCH 35/35] kpartx: add missing newline mwilck
2020-07-20 21:09 ` [PATCH 00/35] multipath-tools series part I: minor changes Benjamin Marzinski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200709101620.6786-27-mwilck@suse.com \
--to=mwilck@suse.com \
--cc=bmarzins@redhat.com \
--cc=christophe.varoqui@opensvc.com \
--cc=dm-devel@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.