Mutipath creating multiplath devices on local controller disks, not only storage

Latest response

Hi Community,

I have a DL560 Gen9 system connected with multipath Fibre Channel to a storage MSA 1040 as well 4 pairs of disks in RAID-1 in the internal controller.

The multipath is configured "properly" with find_multipaths to yes as well setup as recommended for a MSA 1040, or:

defaults {
        #polling_interval       10
        #path_selector          "round-robin 0"
        #path_grouping_policy   multibus
        #uid_attribute          ID_SERIAL
        prio                    alua
        #path_checker           readsector0
        #rr_min_io              100
        #max_fds                        8192
        #rr_weight              priorities
        #failback               immediate
        #no_path_retry          fail
        user_friendly_names     yes
        find_multipaths         yes
}
###
blacklist {
       devnode "^(ram|raw|loop|fd|md|dm-|sr|scd|st)[0-9]*"
       devnode "^hd[a-z]"
}
###

##
devices {
        device {
        vendor                  "HP"
        product                 "MSA 1040 SAN"
        path_grouping_policy    group_by_prio
        path_selector           "round-robin 0"
        path_checker            tur
        hardware_handler        "0"
        failback                immediate
        rr_weight               uniform
        rr_min_io               100
        no_path_retry           18
        user_friendly_names     yes
        find_multipaths         yes
        }
}

Although the local RAID disks, with exception of sda which is the boot disk, is being marked as multipath. See below:

mpathd (3600508b1001c4xxxxx) dm-2 HP      ,LOGICAL VOLUME
size=279G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='service-time 0' prio=1 status=active
  `- 6:1:0:3 sdd 8:48 active ready running
mpathc (3600508b1001c4xxxxx) dm-1 HP      ,LOGICAL VOLUME
size=279G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='service-time 0' prio=1 status=active
  `- 6:1:0:2 sdc 8:32 active ready running
mpathb (3600508b1001c4xxxxx) dm-0 HP      ,LOGICAL VOLUME
size=279G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='service-time 0' prio=1 status=active
  `- 6:1:0:1 sdb 8:16 active ready running
bkp01 (3600508b1001c4xxxxx) dm-3 HP      ,MSA 1040 SAN
size=7.6T features='1 queue_if_no_path' hwhandler='0' wp=rw
|-+- policy='round-robin 0' prio=50 status=active
| `- 8:0:0:0 sdf 8:80 active ready running
`-+- policy='round-robin 0' prio=10 status=enabled
  `- 7:0:0:0 sde 8:64 active ready running

We can see there is no real multiple paths on the mpathb, mpathc and mpathd devices.

I do not see reason the multipath is marking these devices since they do not really have multiple paths.

Any comment what could be happening in the RHEL7 configuration logic?

RC

Responses