redhat-list Digest, Vol 121, Issue 1

[Yong Huang]

What I would do is compare the number of paths with the normal count. For example, my "Script to check multipath failures" at

http://yong321.freeshell.org/computer/Multipath.html

set up as a cron job, alerts me when the number of active paths is below the normal count, 4 in this case. So you establish a baseline first when all active paths are avaialable.


------------------------------

Message: 2
Date: Fri, 28 Feb 2014 20:00:43 +0100
From: emmanuel segura <***@gmail.com>
To: General Red Hat Linux discussion list <redhat-***@redhat.com>
Subject: Re: [rhel6] How-to show failed paths with "multipath"
Message-ID:
    <CAE7pJ3CORoQt-JTEac3VMPpEjnGqz0rHFkBW-1D-***@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

In the newers Suse 11 and Redhat 6, when a path is los, the dead devices
are removed from system, that's what i know

Hello,
With RHEL6 on a SAN environment, is it possible to display dead paths with
"multipath" ?
data (360050768028107fac00000000000009f)
[size=150 GB][features="1 queue_if_no_path"][hwhandler="0"]
\_ round-robin 0 [enabled]
  \_ 3:0:0:2 sde 8:64  [failed][faulty]
  \_ 4:0:0:2 sdm 8:192 [failed][faulty]
\_ round-robin 0 [prio=20][active]
  \_ 3:0:1:2 sdi 8:128 [active][ready]
  \_ 4:0:1:2 sdq 65:0  [active][ready]
sata (360050768028107fac0000000000000d0) dm-3 IBM,2145
size=170G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=10 status=active
  |- 4:0:1:1 sdf 8:80  active ready  running
  `- 3:0:1:1 sdj 8:144 active ready  running
Increasing the verbose level of "multipath" doesn't work (the
"/etc/multipath.conf" file is classic on both systems).
Thanks,
Nicolas