ioscan -funC disk NO_HW - getting rid of

[silusan]

last few lines from ioscan -funC disk from a hpvm host (11.31)

disk    280  2/0/14/1/0/4/0.16.11.0.0.7.0  sdisk   NO_HW       DEVICE       HP      HSV300
                           /dev/dsk/c10t7d0   /dev/rdsk/c10t7d0
disk    283  2/0/14/1/0/4/0.16.11.0.0.7.1  sdisk   NO_HW       DEVICE       HP      HSV300
                           /dev/dsk/c10t7d1   /dev/rdsk/c10t7d1
disk    286  2/0/14/1/0/4/0.16.11.0.0.7.2  sdisk   NO_HW       DEVICE       HP      HSV300
                           /dev/dsk/c10t7d2   /dev/rdsk/c10t7d2
vmhost #

vmhost  # ioscan -m dsf /dev/dsk/c10t7d2
Persistent DSF           Legacy DSF(s)
========================================
/dev/disk/disk290        /dev/dsk/c10t7d2

vmhost #

vmhost #
vmhost # ioscan -m dsf /dev/disk/disk290
Persistent DSF           Legacy DSF(s)
========================================
/dev/disk/disk290        /dev/dsk/c10t7d2
                         /dev/dsk/c12t7d2
                         /dev/dsk/c8t7d2
                         /dev/dsk/c6t7d2
vmhost #

vmhost #

did:

rmsf /dev/dsk/c10t7d2   /dev/rdsk/c10t7d2
rmsf /dev/dsk/c12t7d2   /dev/rdsk/c12t7d2
rmsf /dev/dsk/c8t7d2  /dev/rdsk/c8t7d2
rmsf /dev/dsk/c6t7d2  /dev/rdsk/c6t7d2
rmsf /dev/disk/disk290  /dev/rdisk/disk290

but i see this in ioscan -funC disk

disk    286  2/0/14/1/0/4/0.16.11.0.0.7.2  sdisk   NO_HW       DEVICE       HP      HSV300

How do I get roid of these lines without rebooting the host?

Thank you.

[Torsten.]

Try

# rmsf -k -H 2/0/14/1/0/4/0.16.11.0.0.7.0

or

# rmsf -x

[silusan]

Thank you very much.

vmhost # ioscan -funC disk| grep NO_HW | awk '{print $3}' > /tmp/NO_HW

vmhost # while read line
> do
> echo "rmsf -k -H $line" >> /tmp/test1
> done < /tmp/NO_HW

vmhost # chmod u+x /tmp/test1

vmhost # ./tmp/test1

vmhost # ioscan -funC disk| grep NO_HW

vmhost #

vmhost # ioscan

vmhost # insf

vmhost # insf -e

I still cant find the eva luns presented from HP EVA 4400, onto our vm host...this is another issue I need to work out

[Johnson Punniyalingam]

For disk arrays, if the WWID of the LUN representing the array controller (that is, LUN 0) changes, all LUN paths of that target
port go to the NO_HW state on a subsequent ioscan, resulting in a loss of access to the array controller. This causes the host
operating system to no longer probe for LUNs behind that array controller so the LUNs behind the controller also go to the
NO_HW state.

Run scsimgr replace_wwid to correct the problem:
scsimgr [-f] replace_wwid identifier [dsf]


or

Schedule downtime and reboot the server.

[OScar123]

If all disks in no_hw have  been removed then

check once again through below command.

ioscan -m lun | grep -i no_hw

If everything ok then go to /dev/disk and execute ls -alrt.

CAn you see the devices ? If not then

unpresent the disks . again remove the no_hw path and present the disks again.

[silusan]

ioscan -m lun | grep -i no_hw gave so many NO_HW lines

deleted the hardware path

then the new luns appeared! thanks much