[prev in list] [next in list] [prev in thread] [next in thread] 

List:       aix-l
Subject:    Re: Migrating to FastT
From:       Soós_Zoltán <soos.zoltan () gmail ! com>
Date:       2005-07-22 18:57:13
Message-ID: 81330cc9050722115721ef618a () mail ! gmail ! com
[Download RAW message or body]

> Been there done that.  Will try again.
> 
> #./t1
> hdisk2 deleted
> hdisk3 deleted
> dar0 deleted
> dar1 deleted
> dac0 deleted
> dac1 deleted
> fcnet0 deleted
> fcnet1 deleted
> fscsi0 deleted
> fcs0 deleted
> fscsi1 deleted
> fcs1 deleted
> 
> Where t1 is :
> 
> for i in `lsdev -Cc disk|grep Array|awk '{print $1'}`
> do
>         rmdev -dl $i
> done
> 
> rmdev -dl dar0
> rmdev -dl dar1
> 
> rmdev -dl dac0
> rmdev -dl dac1
> 
> rmdev -dl fcnet0
> rmdev -dl fcnet1
> 
> rmdev -Rdl fcs0
> rmdev -Rdl fcs1
> 
> So now ...
> 
> # fget_config -Av
> #
> 
> Ok .. I'll spare you the cfgmgr -v ...
> 
> # fget_config -Av
> 
> ---dar0---
> 
> User array name = 'VSC_DS4300_1'
> dac0 ACTIVE dacNONE ACTIVE
> 
> Disk     DAC   LUN Logical Drive
> utm             31
> hdisk2   dacNONE   0 colleague2
> 
> ---dar1---
> 
> User array name = 'VSC_DS4300_1'
> dac1 ACTIVE dacNONE ACTIVE
> 
> Disk     DAC   LUN Logical Drive
> utm             31
> hdisk3   dac1    0 colleague2
 
Nice try, but it's still sick. Hmmmm. 

> > still have two dar devices), then it may be controller firmware
> > related. You may want to check if there is a newer firmware available
> > for your FAStT box (probably 6.12 or 6.15). Since you have two
> 
> Ok, that's the next step.  We're running 6.12.03 ... so I'll look for a
> newer version.

In fact, I have never seen this kind of behavior with any of the 6.xx
firmwares. So I do not anticipate that an upgrade would solve the
issue. However, if you have the will and the means you can go ahead
and try a new firmware. It may also prove to be helpful if you are
willing to escalate the problem to IBM since they always insist on
using the newest version available.

> > switches and the are not connected via an ISL (they are not, aren't
> > they? should not be), zoning is not much of an issue because your
> > paths are running like parallel lines, which is the correct way.
> 
> Right.  No ISL's that I know of.

Good, good. If all else fails you may want to rewire the config for a
minute, just to make some tests. Try to attach the FAStT directly to
your FC adapters (F/C 6239, I assume) and see if it solves the issue.
Other than that, you can still check the microcode on the adapters and
the level of AIX drivers (devices.fcp.disk.rte and
devices.fcp.disk.array.rte). I have no other ideas.
   Check the connections. Unplug the LC connectors and check them, you
should see red light coming out of both connectors (one wire per
connector, of course).
[prev in list] [next in list] [prev in thread] [next in thread]