Invalid Storage Timeout Value with iSCSI+Multipath

(Doc ID 2398211.1)

Last updated on MAY 18, 2018

Applies to:

Linux OS - Version Oracle Linux 6.0 and later
Linux x86-64

Symptoms

Some of iscsi timeout values are accidentally changed to 5(sec) although the value is configured as 120(sec, by default), when the iscsi targets are the part of(paths of) multipath devices.

iscsi timout value is initially described as "node.session.timeo.replacement_timeout" in /etc/iscsi/iscsid.conf, then copyed into /var/lib/iscsi/nodes/*/*/default when first iscsi connection is configured, and then set to /sys/class/iscsi_session*/recovery_tmo by iscsid.

Additionally, some of the values would be changed to 120 after first path recovery.

# multipath -ll
1IET 00020001 dm-3 IET,VIRTUAL-DISK
size=15G features='0' hwhandler='0' wp=rw
|-+- policy='round-robin 0' prio=1 status=active
|  `- 15:0:0:1 sde 8:64 active ready running
`-+- policy='round-robin 0' prio=1 status=enabled
   `- 16:0:0:1 sdd 8:48 active ready running
1IET 00010001 dm-2 IET,VIRTUAL-DISK
size=10G features='0' hwhandler='0' wp=rw
|-+- policy='round-robin 0' prio=1 status=active
|  `- 13:0:0:1 sdb 8:16 active ready running
`-+- policy='round-robin 0' prio=1 status=enabled
   `- 14:0:0:1 sdc 8:32 active ready running
#
# cat /var/lib/iscsi/nodes/*/*/default | grep replacement
node.session.timeo.replacement_timeout = 120
node.session.timeo.replacement_timeout = 120
node.session.timeo.replacement_timeout = 120
node.session.timeo.replacement_timeout = 120
#
# cat /sys/class/iscsi_session/session*/recovery_tmo
5
120
120
5
(all should be 120)
#
# ifconfig eth0 down; sleep 180; ifconfig eth0 up    <<<<< (disconnect and reconnect several paths)
#
# sleep 10
# cat /sys/class/iscsi_session/session*/recovery_tmo
120
120
120
5
(Some of them has been changed to 120)
#
# service multipathd restart
Stopping multipathd daemon: [ OK ]
Starting multipathd daemon: [ OK ]
#
# cat /sys/class/iscsi_session/session*/recovery_tmo
5
120
120
5
(Some of them has been changed to 5)
#

 

Changes

multipathd is started/restarted, and also multipathd has specific configuration in /etc/multipath.conf like:

 

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms