dm-devel.redhat.com archive mirror
 help / color / mirror / Atom feed
From: "Benjamin Marzinski" <bmarzins@redhat.com>
To: device-mapper development <dm-devel@redhat.com>
Cc: Martin Wilck <mwilck@suse.com>
Subject: [PATCH 10/12] multipathd: clean up set_no_path_retry
Date: Thu,  7 Dec 2017 12:49:04 -0600	[thread overview]
Message-ID: <1512672546-12785-11-git-send-email-bmarzins@redhat.com> (raw)
In-Reply-To: <1512672546-12785-1-git-send-email-bmarzins@redhat.com>

set_no_path_retry() was always updating the queue_if_no_path setting,
even if it was unnecessary.  This caused problems when no_path_retry was
set to a number, where any time __setup_multipath() was called with
reset, queuing automatically restarted, even if there were no usable
paths.

Instead, set_no_path_retry() should just fix the queueing if it is set
to an incorrect value.  This is simple to detect if no_path_retry is set
to "queue" or "fail".  When it is set to a number, multipathd needs to
see if there are usable paths. If so, queueing should be enabled.  If
not, queueing may be either enabled or disabled, but if it is enabled
and the device isn't already in recovery mode, it should be.

Also, calling dm_map_present() is pointless in __setup_multipath(),
since the existence of the map was just checked in dm_get_info().

Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
---
 libmultipath/structs_vec.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/libmultipath/structs_vec.c b/libmultipath/structs_vec.c
index 2f1f277..fbab61f 100644
--- a/libmultipath/structs_vec.c
+++ b/libmultipath/structs_vec.c
@@ -325,21 +325,27 @@ void enter_recovery_mode(struct multipath *mpp)
 
 static void set_no_path_retry(struct multipath *mpp)
 {
-	mpp->retry_tick = 0;
+	char is_queueing = 0;
+
 	mpp->nr_active = pathcount(mpp, PATH_UP) + pathcount(mpp, PATH_GHOST);
+	if (mpp->features && strstr(mpp->features, "queue_if_no_path"))
+		is_queueing = 1;
 
 	switch (mpp->no_path_retry) {
 	case NO_PATH_RETRY_UNDEF:
 		break;
 	case NO_PATH_RETRY_FAIL:
-		dm_queue_if_no_path(mpp->alias, 0);
+		if (is_queueing)
+			dm_queue_if_no_path(mpp->alias, 0);
 		break;
 	case NO_PATH_RETRY_QUEUE:
-		dm_queue_if_no_path(mpp->alias, 1);
+		if (!is_queueing)
+			dm_queue_if_no_path(mpp->alias, 1);
 		break;
 	default:
-		dm_queue_if_no_path(mpp->alias, 1);
-		if (mpp->nr_active == 0)
+		if (mpp->nr_active > 0)
+			dm_queue_if_no_path(mpp->alias, 1);
+		else if (is_queueing && mpp->retry_tick == 0)
 			enter_recovery_mode(mpp);
 		break;
 	}
@@ -354,12 +360,6 @@ int __setup_multipath(struct vectors *vecs, struct multipath *mpp,
 		goto out;
 	}
 
-	if (!dm_map_present(mpp->alias)) {
-		/* Table has been removed */
-		condlog(3, "%s: table does not exist", mpp->alias);
-		goto out;
-	}
-
 	if (update_multipath_strings(mpp, vecs->pathvec, 1)) {
 		condlog(0, "%s: failed to setup multipath", mpp->alias);
 		goto out;
-- 
2.7.4

  parent reply	other threads:[~2017-12-07 18:49 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-07 18:48 [PATCH 00/12] Misc fixes Benjamin Marzinski
2017-12-07 18:48 ` [PATCH 01/12] multipath: add "ghost_delay" parameter Benjamin Marzinski
2017-12-07 22:08   ` Martin Wilck
2017-12-07 18:48 ` [PATCH 02/12] kpartx: don't delete partitions from partitions Benjamin Marzinski
2017-12-07 22:09   ` Martin Wilck
2017-12-07 18:48 ` [PATCH 03/12] multipath: fix hwhandler check in select_action Benjamin Marzinski
2017-12-07 22:09   ` Martin Wilck
2017-12-07 18:48 ` [PATCH 04/12] libmultipath: cleanup features handling code Benjamin Marzinski
2017-12-07 22:10   ` Martin Wilck
2017-12-08 15:24   ` Martin Wilck
2017-12-08 21:12     ` Benjamin Marzinski
2017-12-07 18:48 ` [PATCH 05/12] multipathd: move helper functions to libmultipath Benjamin Marzinski
2017-12-07 22:11   ` Martin Wilck
2017-12-07 18:49 ` [PATCH 06/12] multipathd: fix device creation issues Benjamin Marzinski
2017-12-08 17:26   ` Martin Wilck
2018-01-30 16:51   ` Martin Wilck
2018-01-30 18:34     ` Benjamin Marzinski
2018-01-30 19:02       ` Martin Wilck
2017-12-07 18:49 ` [PATCH 07/12] multipathd: remove select_* from setup_multipath Benjamin Marzinski
2017-12-08 20:08   ` Martin Wilck
2017-12-07 18:49 ` [PATCH 08/12] libmultipath: __setup_multipath param cleanup Benjamin Marzinski
2017-12-08 20:13   ` Martin Wilck
2017-12-07 18:49 ` [PATCH 09/12] multipathd: move recovery mode code to function Benjamin Marzinski
2017-12-07 22:13   ` Martin Wilck
2017-12-07 18:49 ` Benjamin Marzinski [this message]
2017-12-07 22:14   ` [PATCH 10/12] multipathd: clean up set_no_path_retry Martin Wilck
2017-12-07 18:49 ` [PATCH 11/12] multipath: check failed path dmstate in check_path Benjamin Marzinski
2017-12-07 22:14   ` Martin Wilck
2017-12-07 18:49 ` [PATCH 12/12] multipathd: marginal path code fixes Benjamin Marzinski
2017-12-07 22:15   ` Martin Wilck
2018-01-13  9:19 ` [PATCH 00/12] Misc fixes Christophe Varoqui

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=1512672546-12785-11-git-send-email-bmarzins@redhat.com \
    --to=bmarzins@redhat.com \
    --cc=dm-devel@redhat.com \
    --cc=mwilck@suse.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).