I'm not a pro at dm-multipath but if you are looking for failover dont you want your path_grouping_policy line to be
default_path_grouping_policy failover
that way your multipath -ll output would look something like
archives (30690a018a0e3a3178c46f4510900307f)
[size=16 GB][features="0"][hwhandler="0"]
\_ round-robin 0 [active]
\_ 2:0:1:0 sdc 8:32 [active][ready]
\_ round-robin 0 [enabled]
\_ 3:0:1:0 sdo 8:224 [active][ready]
hope some of that made sense..
--
:wq!
kevin.foote
On 6/27/06, Andrew Elwell <andrewe@xxxxxxxxxxxxx> wrote:
Hi folks,
I'm trying to set up access to our SAN, 5 cabinets of 6320 storage, from
CentOS 4.3 using dm-multipath and two Emulex lpfc cards.
So far I can see the correct arrays through the zoning and have setup
multipath.conf as below:
defaults {
user_friendly_names yes
polling_interval 1
path_checker tur
path_grouping_policy group_by_serial
}
devnode_blacklist {
devnode "^(ram|raw|loop|fd|md|dm-|sr|scd|st)[0-9]*"
devnode "^hd[a-z]"
devnode "^sd[a-b]"
# sda and sdb are internal 3ware raid and sata system disk
devnode "^cciss!c[0-9]d[0-9]*"
}
multipaths {
multipath {
wwid 360003baccb976000426f26d30004c45a
alias cab2a01t0
}
multipath {
wwid 360003baccb976000426f27a00006ae93
alias cab2a01t1
}
... snip more multipath definitions....
}
multipath -ll shows..
cab2a01t2 (360003baccb976000426f289000049adb)
[size=1638 GB][features="0"][hwhandler="0"]
\_ round-robin 0 [active]
\_ 3:0:0:2 sde 8:64 [failed][faulty]
\_ 4:0:0:2 sdj 8:144 [active][ready]
cab2a01t1 (360003baccb976000426f27a00006ae93)
[size=1638 GB][features="0"][hwhandler="0"]
\_ round-robin 0 [active]
\_ 3:0:0:1 sdd 8:48 [active][ready]
\_ 4:0:0:1 sdi 8:128 [active][ready]
cab2a01t0 (360003baccb976000426f26d30004c45a)
[size=1638 GB][features="0"][hwhandler="0"]
\_ round-robin 0 [active]
\_ 3:0:0:0 sdc 8:32 [active][ready]
\_ 4:0:0:0 sdh 8:112 [active][ready]
but the error log is filling up with
kernel: Device sde not ready.
kernel: end_request: I/O error, dev sde, sector 2181037632
multipathd: 8:64: mark as failed
multipathd: cab2a01t2: remaining active paths: 1
multipathd: 8:64: tur checker reports path is up
multipathd: 8:64: reinstated
multipathd: cab2a01t2: remaining active paths: 2
kernel: Device sde not ready.
kernel: end_request: I/O error, dev sde, sector 3187670215
kernel: device-mapper: dm-multipath: Failing path 8:64.
etc. I guess I should be using something other than the default for
prio_callout, but what?
Please can someone give me hints on the correct incantation for
path_checker, path_grouping_policy, prio_callout
Many thanks
Andrew
--
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
-- dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
