[Pacemaker] Frequent SBD triggered server reboots

andrea cuozzo andrea.cuozzo at sysma.it
Thu May 2 17:58:08 UTC 2013


Hi Emmanuel,

it's a merge of the default settings I've taken from the
/usr/share/doc/package/multipath-tools/multipath.conf.annotated   and the
OPEN-* section of the HPDMmultipath-4.4.1 multipath.conf.HPTemplate.SLES11
file, here it is:


defaults {
        udev_dir                /dev
        polling_interval        5
        selector                "round-robin 0"
        prio                    const
        user_friendly_names     no
        getuid_callout          "/lib/udev/scsi_id --whitelisted
--device=/dev/%n"
        failback                manual
        no_path_retry           fail
}
blacklist {
        devnode         "^(ram|raw|loop|fd|md|dm-|sr|scd|st)[0-9]*"
        devnode         "^hd[a-z][[0-9]*]"
}
blacklist_exceptions {
}
multipaths {
        multipath {
                alias "san"
                wwid "360060e8006d2e1000000d2e1000000e6"
        }
}
devices {
# For XP arrays

  device {
        vendor                  "HP"
        product                 "OPEN-.*"
        path_grouping_policy    multibus
        path_selector           "round-robin 0"
        rr_min_io               1000
        rr_weight               uniform
        path_checker            tur
#       no_path_retry           18
#       hardware_handler        "0"
#       failback                immediate
#       getuid_callout          "/lib/udev/scsi_id -g -u /dev/%n"
  }

}

I've commented out some of the HP template device lines due to the
recommended configuration for clustering, read in the SLES 11 Sp1 Storage
Administration Guide:

"We recommend failback setting of "manual" for multipath in cluster
environments in order to prevent multipath failover ping-pong." 

and in the mailing list: " Fourth, don't run SBD on queue_if_no_path MPIO."
>From Mr. Lars at this link: http://en.it-usenet.org/thread/18723/12998/

And this my multipath -ll output:

server1:~ # multipath -ll

san (360060e8006d2e1000000d2e1000000e6) dm-0 HP,OPEN-V
size=50G features='0' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
  |- 1:0:0:0 sdb 8:16 active ready running
  `- 2:0:0:0 sdc 8:32 active ready running

andrea


----- Original Message ----- 
From: emmanuel segura 
Newsgroups: gmane.linux.highavailability.pacemaker
To: The Pacemaker cluster resource manager 
Sent: Thursday, May 02, 2013 4:25 PM
Subject: Re: Frequent SBD triggered server reboots


Hello Andrea


Can you show me your multipath.conf?


Thanks









More information about the Pacemaker mailing list