Skip to main content
Start a Conversation

This post is more than 5 years old

Solved!

Go to Solution

zainal1

127 Posts

7929

July 27th, 2009 01:00

1 HBA missing in Powerpath

Hi
I have 2 RedHat Enterprise 4 Linux nodes running on a RedHat cluster on a CX300.
In each nodes, there are 2x HBA Qlogic cards.

One of the node, is showing just one HBA card, even though in reality there shld be 1 more HBA card, total 2 HBA cards.

[root@misln01 scsi]# powermt display
Symmetrix logical device count=0
CLARiiON logical device count=6
==============================================================================
----- Host Bus Adapters --------- ------ I/O Paths ----- ------ Stats ------
### HW Path Summary Total Dead IO/Sec Q-IOs Errors
==============================================================================
1 qla2xxx optimal 12 0 - 0 0


A run of scli Qlogic utlity confirm this.

Host Name : misln01.pen.comm.mot.com
OS Type : Linux - Red Hat Enterprise Linux AS release 4 i686
OS Version : 2.6.9-42.0.10.ELsmp
FO API Version : 2.2.9
SDM API Version : v2.02 beta14
-----------------------------------------------------------------------------
HBA Model QLA2340:
Port 1: WWPN 21-00-00-E0-8B-8F-26-92 SN K20454 FW 3.03.19 Online
HBA Model QLA2340:
Port 0: WWPN 21-00-00-E0-8B-8F-73-97 SN K21811 FW 3.03.19 Online
-----------------------------------------------------------------------------
Total QLogic HBA(s) : 2

I have tried to run powermt config, but still its showing 1 HBA card.

What shld I do to get the other HBA card shown in PP without affecting the I/O.

Thanks

Zainal

dynamox

1 Rookie

 • 

20.4K Posts

July 29th, 2009 04:00

try these commands, on my system hbas are 1 and 2, on yours they could be 3 and 4. Cd into appropriate directory to see what you have. Reboot will be your last option:


RHEL4 native qla driver:

echo "1" > /sys/class/fc_host/host1/issue_lip
echo "1" > /sys/class/fc_host/host2/issue_lip
echo "- - -" > /sys/class/scsi_host/host1/scan
echo "- - -" > /sys/class/scsi_host/host2/scan

qlogic driver for 2.6 kernel (RHEL4):

echo "scsi-qlascan" > /proc/scsi/qla2xxx/1
echo "scsi-qlascan" > /proc/scsi/qla2xxx/2
echo "- - -" > /sys/class/scsi_host/host1/scan
echo "- - -" > /sys/class/scsi_host/host2/scan

dynamox

1 Rookie

 • 

20.4K Posts

July 27th, 2009 03:00

can you check Connectivity Status on Clariion to make sure that HBA is logged-in and it's green color (part of as storage group)

zainal1

127 Posts

July 28th, 2009 23:00

Connectivity status shows 1 HBA connected to A1 and B0 ports as greyed out.
There is ~management under Connected Storage groups as green color.

Conor

341 Posts

July 28th, 2009 23:00

As per emc199088 the initiator is not registered correctly on the CX array, De-register the initiator and manually register it again.

zainal1

127 Posts

July 29th, 2009 00:00

Hi Conor

I forgot to mention that this is a 2-node Linux cluster.
Each node having 2 HBA's.

My guess is that the HBA which was greyed out was faulty and replaced with a new HBA which is shown under the ~management.
Can I click on the Reconnect button to refresh the affected node so that the WWN no get refresh with a new WWN no. I check somewhere in this PowerPath forum that this can be done.


Thanks

Conor

341 Posts

July 29th, 2009 00:00

Well if you are not currently seeing devices down this path then "Reconnect" will not have any impact, and may bring the path back online (Registered and Logged in on the array)

zainal1

127 Posts

July 29th, 2009 02:00

Shld I highlight the the HBA on ~management first before I click on reconnect?

zainal1

127 Posts

July 29th, 2009 02:00

After doing reconnect of the HBA, i went to the host and did a powermt config.
Somehow the devices on the 2nd HBA is only 2 devices and not 12 as it shld be.


[root@misln01 log]# powermt display
Symmetrix logical device count=0
CLARiiON logical device count=6
==============================================================================
----- Host Bus Adapters --------- ------ I/O Paths ----- ------ Stats ------
### HW Path Summary Total Dead IO/Sec Q-IOs Errors
==============================================================================
1 qla2xxx optimal 12 0 - 2 0
2 qla2xxx optimal 2 0 - 0 0

Conor

341 Posts

July 29th, 2009 02:00

Yes - thats it, let us know how it goes

Conor

341 Posts

July 29th, 2009 02:00

That¿s a good sign, at least the HBA is showing as logging in. You may need to rescan for new devices at the OS level, there are Qlogic rescan scripts available as part of their driver package, personally I think a reboot is the most consistent way of picking up the devices.

zainal1

127 Posts

July 29th, 2009 19:00

Hi Dynamox

The commands below work for me. Thanks

zainal1

127 Posts

August 5th, 2009 00:00

Hi dynamox

I have another red hat enterprise 4 linux server on kernel 2.6 also facing same problem.HBA card is Emulex lpfc. THis is a 2-node veritas cluster.
What shld be the equivalent command for lpfc drivers?

[root@zmy02spc01 emc]# ./lun_scan.sh all
Scanning lpfc HBA instance with scsi host number : 0
Scanning lpfc HBA instance with scsi host number : 1

[root@zmy02spc01 emc]# powermt display
Symmetrix logical device count=0
CLARiiON logical device count=5
Hitachi logical device count=0
Invista logical device count=0
HP xp logical device count=0
Ess logical device count=0
HP HSx logical device count=0
==============================================================================
----- Host Bus Adapters --------- ------ I/O Paths ----- ------ Stats ------
### HW Path Summary Total Dead IO/Sec Q-IOs Errors
==============================================================================
0 lpfc optimal 10 0 - 0 0

dynamox

1 Rookie

 • 

20.4K Posts

August 5th, 2009 03:00

give this a shot

RHEL4 native driver:

echo "1" > /sys/class/fc_host/host1/issue_lip
echo "1" > /sys/class/fc_host/host2/issue_lip
echo "- - -" > /sys/class/scsi_host/host1/scan
echo "- - -" > /sys/class/scsi_host/host2/scan

zainal1

127 Posts

August 5th, 2009 20:00

Hi Dynamox

I've tried, still i cannot get the 2nd HBA.
I even unplug and plug the 2nd HBA, still cannot see the HBA online
My colleague was telling me that she has to reboot the server a few times
in order to get both HBA's online.
Is this a sign that the HBA is bad?
Is there any Emulex utlitity that I can get the HBA online, just like scli in Qlogic.

[root@zmy02spc01 lpfc]# pwd
/proc/scsi/lpfc

[root@zmy02spc01 lpfc]# more 0
lpfc0t00 DID 010000 WWPN 50:06:01:60:41:e0:0a:4e WWNN 50:06:01:60:c1:e0:0a:4e
lpfc0t01 DID 010800 WWPN 50:06:01:69:41:e0:0a:4e WWNN 50:06:01:60:c1:e0:0a:4e

[root@zmy02spc01 lpfc]# more 1

dynamox

1 Rookie

 • 

20.4K Posts

August 5th, 2009 21:00

did you install HBAnywhere utility ?

http://www.emulex.com/downloads/emc/drivers/linux.html

View More

View All

No Events found!

Top