LinuxQuestions.org
Welcome to the most active Linux Forum on the web.
Go Back   LinuxQuestions.org > Forums > Enterprise Linux Forums > Linux - Enterprise
User Name
Password
Linux - Enterprise This forum is for all items relating to using Linux in the Enterprise.

Notices


Reply
  Search this Thread
Old 06-19-2008, 01:14 PM   #1
mkono
Member
 
Registered: May 2008
Posts: 50

Rep: Reputation: 16
ioscan


Ok folks, I have a new one.

I need a command like ioscan that will pick up new LUNS that have been attached to a server.

IBM Blade HS21 running RHEL5.
 
Old 06-19-2008, 06:30 PM   #2
MensaWater
LQ Guru
 
Registered: May 2005
Location: Atlanta Georgia USA
Distribution: Redhat (RHEL), CentOS, Fedora, CoreOS, Debian, FreeBSD, HP-UX, Solaris, SCO
Posts: 7,831
Blog Entries: 15

Rep: Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668
Try the kudzu command. (man kudzu for details).

Not exactly the same as ioscan but may do it for you.

Also how is the storage attached? SCSI? Fibre? Some of the Fibre HBA vendors make tools to rescan for SAN attached things. For example Qlogic has a utility you can download if you're using their cards.
 
Old 06-20-2008, 07:55 AM   #3
mkono
Member
 
Registered: May 2008
Posts: 50

Original Poster
Rep: Reputation: 16
jlightner,
Thanks for the response.

I am using QLogic Fibre HBA's. I downloaded their LUN scan utility and believe it or not It didnt recognize anything... If you are familiar with these utlities I can post the output of that utility.

Im going to check out the kudzu command and see what that does.

Thanks,
Matt


UPDATED SCREEN SHOTS......

$ powermt display
Symmetrix logical device count=0
CLARiiON logical device count=1
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
==============================================================================
3 qla2xxx optimal 2 0 - 0 0




]# ./ql-dynamic-tgt-lun-disc.sh
cat: /proc/scsi/qla2xxx/0: No such file or directory
Scanning HOST: host0
No devices attached to HOST: host0
cat: /proc/scsi/qla2xxx/1: No such file or directory
Scanning HOST: host1
No devices attached to HOST: host1
cat: /proc/scsi/qla2xxx/2: No such file or directory
Scanning HOST: host2
No devices attached to HOST: host2
cat: /proc/scsi/qla2xxx/3: No such file or directory
Scanning HOST: host3


-----There was a 2nd Zone added to the WWN for the HBA Adapter and I can't see it for the life of me. The EMC guy was stumped as well....




Quote:
Originally Posted by jlightner View Post
Try the kudzu command. (man kudzu for details).

Not exactly the same as ioscan but may do it for you.

Also how is the storage attached? SCSI? Fibre? Some of the Fibre HBA vendors make tools to rescan for SAN attached things. For example Qlogic has a utility you can download if you're using their cards.

Last edited by mkono; 06-20-2008 at 08:01 AM.
 
Old 06-20-2008, 08:04 AM   #4
MensaWater
LQ Guru
 
Registered: May 2005
Location: Atlanta Georgia USA
Distribution: Redhat (RHEL), CentOS, Fedora, CoreOS, Debian, FreeBSD, HP-UX, Solaris, SCO
Posts: 7,831
Blog Entries: 15

Rep: Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668
What type of storage is attached? Are you sure you've zoned in the storage to the host?

I've used the utility from Qlogic on both RHEL3 and RHEL5 for EMC Clariion LUNs attached via SAN fabric (over EMC Connectrix fibre switches).

Have you accounted for all the devices you have as /dev/sd*? If you did a reboot after attaching the storage and it was presented correctly that would automatically have rescanned and you should see them then.

The zoning is important - many folks (like us) specifically only present specific LUNs to specific hosts to prevent accidentally using storage in the wrong place. If your zoning is setup that way you won't see anything just by plugging in the host - you have to modify the zone settings to present the LUNs to the host.
 
Old 06-20-2008, 08:21 AM   #5
mkono
Member
 
Registered: May 2008
Posts: 50

Original Poster
Rep: Reputation: 16
We are using CLARiiON for our storage... I guess I should also let you know that this is a boot from SAN setup. When the machine is configured for Single Path only by the EMC guy, the machine boots just fine... When I boot from SAN under Single Path, then have the EMC guy switch it over to Multipath and reboot the Blade, the boot fails and I get a "fsck.ext3 Unable to resolve LABEL=/boot" error.

So we tried booting the blade as single path, adding multipath functionality and scanning for the 2nd Zone. However this is not working either....

To be honest, I dont have direct access to the SAN setup or utlities associated with our SAN.

BTW I am terribly confused on whether to refer to the addition of multipathing as adding a "2nd LUN" or "adding a 2nd zone" Im a little naive on the verbiage to use here.

If I rebbot the blade and go into the QLogic setup utility, I can see my 2 HBA's, if I do a scan for devices, I am able to see both LUNS for each HBA.

I fear I might be making this more confusing than it has to be, due to my lack of knowledge on the EMC/SAN solutions.


Quote:
Originally Posted by jlightner View Post
What type of storage is attached? Are you sure you've zoned in the storage to the host?

I've used the utility from Qlogic on both RHEL3 and RHEL5 for EMC Clariion LUNs attached via SAN fabric (over EMC Connectrix fibre switches).

Have you accounted for all the devices you have as /dev/sd*? If you did a reboot after attaching the storage and it was presented correctly that would automatically have rescanned and you should see them then.

The zoning is important - many folks (like us) specifically only present specific LUNs to specific hosts to prevent accidentally using storage in the wrong place. If your zoning is setup that way you won't see anything just by plugging in the host - you have to modify the zone settings to present the LUNs to the host.
 
Old 06-20-2008, 11:13 AM   #6
akvino
Member
 
Registered: May 2007
Posts: 31

Rep: Reputation: 15
Quote:
Originally Posted by mkono View Post
We are using CLARiiON for our storage... I guess I should also let you know that this is a boot from SAN setup. When the machine is configured for Single Path only by the EMC guy, the machine boots just fine... When I boot from SAN under Single Path, then have the EMC guy switch it over to Multipath and reboot the Blade, the boot fails and I get a "fsck.ext3 Unable to resolve LABEL=/boot" error.

So we tried booting the blade as single path, adding multipath functionality and scanning for the 2nd Zone. However this is not working either....

To be honest, I dont have direct access to the SAN setup or utlities associated with our SAN.

BTW I am terribly confused on whether to refer to the addition of multipathing as adding a "2nd LUN" or "adding a 2nd zone" Im a little naive on the verbiage to use here.

If I rebbot the blade and go into the QLogic setup utility, I can see my 2 HBA's, if I do a scan for devices, I am able to see both LUNS for each HBA.

I fear I might be making this more confusing than it has to be, due to my lack of knowledge on the EMC/SAN solutions.
Same thing happened to me, Qlogic to XP SAN - it would not see it until reboot. Then hp_rescan works and I can attach the LUN's.
 
Old 06-20-2008, 06:05 PM   #7
MensaWater
LQ Guru
 
Registered: May 2005
Location: Atlanta Georgia USA
Distribution: Redhat (RHEL), CentOS, Fedora, CoreOS, Debian, FreeBSD, HP-UX, Solaris, SCO
Posts: 7,831
Blog Entries: 15

Rep: Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668Reputation: 1668
In the systems I've done this on we use internal storage on the server or blade to do the boot then add the Clariion storage for our database storage. Of course we make sure the internal storage is hardware RAID (usually Dell PERC cards) so a single drive failure won't cause the server to crash.

I'm assuming you're using EMC PowerPath and Navisphere on the server? PowerPath is what sets up the multipathing (the EMC way). Navisphere is software designed to interact with the Clariion.

When you use PowerPath it would present a virtual device such as /dev/emcpowera and the different paths to the physical device would each have standard names.

Say you had presented two different LUNs from the Clariion. If you have two fibre HBAs (Qlogics) in the server and have run cables to two different fibre connections on the fibre switch you would actually have 4 total paths. In such a case you would see two virtual drives composed of 4 physical paths each.
e.g.
/dev/emcpowera = /dev/sda, /dev/sdc, /dev/sde & /dev/sdg
/dev/emcpowerb = /dev/sdb, /dev/sdd, /dev/sdf & /dev/sdh

For normal usage you'd want to use the virtual names rather than the physical names. I'm not sure you can do that because PowerPath that presents the virtual device.

I did see a couple of links that might be relevant on EMC's PowerLink site. You may be able to find others.

http://knowledgebase.emc.com/emcice/documentDisplay.do;jsessionid=04F673358FD3A2510BE7AC025605BEFE?docType=1006&clusterName=DefaultClust er&resultType=5002&groupId=1&page=&docProp=$solution_id&docPropValue=emc183989&passedTitle=null
http://knowledgebase.emc.com/emcice/documentDisplay.do?docType=1006&clusterName=DefaultCluster&resultType=5002&groupId=1&page=&docProp=$ solution_id&docPropValue=emc134693&passedTitle=null

The site is powerlink.emc.com - Its been a while since I signed up on that site so I forget what it required. I think I needed to know the array serial number or the site ID.
 
Old 06-21-2008, 08:53 PM   #8
zQUEz
Member
 
Registered: Jun 2007
Distribution: Fedora, RHEL, Centos
Posts: 294

Rep: Reputation: 54
When you use the qlogic utils, you should be able to add a "-p" switch like this:
./ql-dynamic-tgt-lun-disc.sh -p
This tells it to use a proc filesystem. I found this was needed for our servers attached to a HP-EVA frame.

Also, I didn't use the above tool till recently, and the way I typically scan for new LUN's is as follows:
# echo "1" > /sys/class/fc_host/host0/issue_lip
# echo "- - -" > /sys/class/scsi_host/host0/scan
# echo "1" > /sys/class/fc_host/host1/issue_lip
# echo "- - -" > /sys/class/scsi_host/host1/scan

This assumes you are using 2 hba's and they are host0 and host1. Sometimes my hba's show up as host2 and host3 so the above commands would be adjusted for that.
 
Old 06-23-2008, 11:48 AM   #9
akvino
Member
 
Registered: May 2007
Posts: 31

Rep: Reputation: 15
Quote:
Originally Posted by zQUEz View Post
When you use the qlogic utils, you should be able to add a "-p" switch like this:
./ql-dynamic-tgt-lun-disc.sh -p
This tells it to use a proc filesystem. I found this was needed for our servers attached to a HP-EVA frame.

Also, I didn't use the above tool till recently, and the way I typically scan for new LUN's is as follows:
# echo "1" > /sys/class/fc_host/host0/issue_lip
# echo "- - -" > /sys/class/scsi_host/host0/scan
# echo "1" > /sys/class/fc_host/host1/issue_lip
# echo "- - -" > /sys/class/scsi_host/host1/scan

This assumes you are using 2 hba's and they are host0 and host1. Sometimes my hba's show up as host2 and host3 so the above commands would be adjusted for that.
Hello zQUEz. Can you tell me what does "echo 1 > issue_lip" result in, and what does "echo "---"> scan" do?
 
Old 06-23-2008, 12:36 PM   #10
mkono
Member
 
Registered: May 2008
Posts: 50

Original Poster
Rep: Reputation: 16
Just to give you all an update of the issue.

Status: RESOLVED! :-)

I ended up working with our EMC guy to resolve the issue which ended up being the following:

- the host was registered manually via the navi agent, so CLARiiON was confused as to what paths belonged to what host.

- clean up the paths by removing the host and adding it back with the correct paths.

At this point I can successfully boot into multipath mode and view the 2 zones in the "powermt display dev=all" output.

# => powermt display dev=all
Pseudo name=emcpowera
CLARiiON ID=APM00043204754 [sap2b103]
Logical device ID=60060160A0991100799803B5D810DD11 [LUN 921]
state=alive; policy=CLAROpt; priority=0; queued-IOs=0
Owner: default=SP A, current=SP A Array failover mode: 1
==============================================================================
---------------- Host --------------- - Stor - -- I/O Path - -- Stats ---
### HW Path I/O Paths Interf. Mode State Q-IOs Errors
==============================================================================
1 qla2xxx sda SP A2 active alive 0 0
1 qla2xxx sdb SP B2 active alive 0 0
3 qla2xxx sdc SP A3 active alive 0 0
3 qla2xxx sdd SP B3 active alive 0 0


I am now able to boot into the Blade with multipath enabled from the SAN side.

Essentially RHEL5 was configured properly, the QLOGIC adapters were configured properly and the Communication breakdown was on the SAN side.

Thanks to everyone for their advice and I learned some great stuff from this posting, especially the scanning tips/utlities that zQUEz and jlightner educated me on....Thanks you guys. Another group of educational responses. Thanks for all your help. See you on the next post!

-Matt

Last edited by mkono; 06-23-2008 at 01:03 PM.
 
1 members found this post helpful.
Old 05-31-2011, 01:32 AM   #11
hahacc
Member
 
Registered: Oct 2010
Posts: 93

Rep: Reputation: 1
Quote:
Originally Posted by zQUEz View Post
When you use the qlogic utils, you should be able to add a "-p" switch like this:
./ql-dynamic-tgt-lun-disc.sh -p
This tells it to use a proc filesystem. I found this was needed for our servers attached to a HP-EVA frame.

Also, I didn't use the above tool till recently, and the way I typically scan for new LUN's is as follows:
# echo "1" > /sys/class/fc_host/host0/issue_lip
# echo "- - -" > /sys/class/scsi_host/host0/scan
# echo "1" > /sys/class/fc_host/host1/issue_lip
# echo "- - -" > /sys/class/scsi_host/host1/scan

This assumes you are using 2 hba's and they are host0 and host1. Sometimes my hba's show up as host2 and host3 so the above commands would be adjusted for that.
I think before issue_lip, there're some pre-checks we should go through. And as lip is asynchronous operation, we should keep an eye to /var/log/messages or vxdisk path if we're using vxvm. And after issue_lip, we should use vxdisk scandisks fabric to rescan the device bus . Details in the following two links:
http://www.doxer.org/learn-linux/ext...hannel-storage
http://docs.redhat.com/docs/en-US/Re...rconnects.html
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
HP-UX ioscan CDROM device path gctaylor1 Other *NIX 3 11-08-2007 09:11 AM

LinuxQuestions.org > Forums > Enterprise Linux Forums > Linux - Enterprise

All times are GMT -5. The time now is 06:15 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Open Source Consulting | Domain Registration