Scsi device discovery in Cisco MDS switches

I was using this device discovery to find out how many luns can be seen on that server.
I tried to use discover scsi-target command and it does not give much useful information. and I do not want to add the OS WWNS on my LUN security in addition to HBA WWNS for the host group on the array end.
The problem that I am having is that I can see 7 LUNs out of 9 when I connect to Cisco MDS where as in brocade I can see all the LUNs.
I checked the LUN security and zoning and everything seems to be correct.
If Cisco SAN OS has commands where by I can see what LUNs can be seen on the switch, that would be really helpful. But It looks like LUN/LDEV info cannot be seen through discover command.
Please let me know how to troubleshoot this problem
Thanks

In the array, I tried to add switch wwn to the LUN Security and zoned the same switch wwn on the MDS. Is there anything I am missing. Thanks in advance
But still I could not see any LUNs. When I do discover command, NLUNS shows 0.
and the command sh scsi-target lun os win shows nothing.
discover scsi-target vsan 21 fcid 0xea000b os win
VSAN: 21 FCID: 0xea000b
PRLI RSP: 0x01 SPARM: 0x0012
SCSI TYPE: 31 NLUNS: 0
Vendor: HITACHI Model: DISK-SUBSYSTEM Rev: 5007
Other: 7f:00:03:12:cf:00:00:02

Similar Messages

AAA for Cisco MDS Switches

I have configured Cisco ACS 4.0 (TACACS) with Windows AD for all Cisco MDS switches and it is working fine. But local "admin" access to the Cisco MDS switches via telnet is not working. At the same time , if I create a user with "network-admin" role locally, that works but not the default admin user.
Could anyone help me in this regard.

You have two options.
1. Configure an "admin" user in AD. (note that you don't have to use the account named admin, you can just as easily assign a local user with the network-admin role).One thing to note, is that you normally use this local account in case the tacacs+ or radius authentication server goes down.
You can have users configured locally and AD at the same time. If you are running AAA the default config is to check your AAA servers first, if they are not available, then to default to a local account
2. Configure your local network-admin role user and then specify that say console access is authenticated locally, while ssh and telnet is authenticated through tacacs. This will allow you to always get in with a local account through the console, while it will force SSH and Telnet connections to authenticate through the AAA servers.
You can find this option in Device Manager > Security > AAA > Applications
If you found this helpful, please give it a rating.

AAA and Cisco MDS switches.........

have configured Cisco ACS 4.0 (TACACS) with Windows AD for all Cisco MDS switches and it is working fine. But local "admin" access to the Cisco MDS switches via telnet is not working. At the same time , if I create a user with "network-admin" role locally, that works but not the default admin user.
Could anyone help me in this regard.

local. Below is the script I used to configure TACACS (Cisco ACS 4.0) on Cisco MDS switches.
config t
# Enable TACACS+
tacacs+ enable
tacacs-server host nnn.nnn.nnn.nnn key 0 xxxxxx
tacacs-server host mmm.mmm.mmm.mmm key 0 xxxxx
# Specify TACACS+ Server groups
aaa group server tacacs+ tacgrp
server nnn.nnn.nnn.nnn
server mmm.mmm.mmm.mmm
aaa authentication login default group tacgrp
aaa authentication login console local
# Enable TACACS+ Accounting
aaa accounting default group tacgrp local
end
copy running-config startup-config
Thanks
MOhan

Device discovery in Cisco Prime LMS 4.1

Hi,
im using CISCO PRIME LMS 4.1 for manage my network, i ran a device discovery job using a ping sweep on the ip ranges of my network, ie 192.168.1.0/24, 192.168.2.0/24 and so on... the discovery found all my devices, but it discovered duplicate devices with illegal ip Addresses, ie
ip                                    device
192.168.1.255               S1
192.168.2.0                    S1
192.168.2.1                    S1 (the real address)
all in reachable status.
i dont undestad why the program adds that repeated illegal addresses.. someone can help me with that issue??
how can i run the device discovery to avoid this problem?
Many Thanks.

Julio,
Don't let LMS decide on what IP address it will manage the device.
For example, let LMS do a lookup of the sysname in DNS, then it will use the IP address coming back from the DNS as management IP.
This is what I do to avoid this.
But I also feel that LMS should be bright enough to never manage the same device twice. I think Campus DFM or RME will eventually figure out the 2 devices are the same, but they can't "fix" what is wrong in the DCR.
Cheers,
Michel

Scripting for Cisco MDS

Though I know writing Scripts for Cisco MDs switches (as I don't rely on FM/DM GUI), I am wondering how to include comments in a script for better readability. Would appreciate if anyone could point me to the direction where it is been documented or how to do it.
Thanks
Mohan

With regards to your questions:
1- No, you must create the script externally to the MDS as specified in the doc listed below.
http://www.cisco.com/univercd/cc/td/doc/product/sn5000/mds9000/1_3_1/sw_confg/cnfig.pdf
You can however create a cron job on a host and login to the MDS via SSH providing Passwordless access to run a script. Let me know if this is something you would like to pursue.
2- Yes, this is a feature in version 2.0(1b) which was released recently to Cisco.com, look under "Command Scheduler" in the following link.
http://www.cisco.com/univercd/cc/td/doc/product/sn5000/mds9000/2_0/ol624901.htm

Cisco MDS 9418 switch doubt

Hi Team,
We have an issue, whether the cisco MDS 9418 switch supports FCIP feature.
We are planning to migrate the data from our old data center over WAN to the EMC VNX storage box in other location.
In our old data center we have 2 FC cisco switches connected to the EMC clariion storage array. Inorder to replicate data from clariion box to EMC VNX box over WAN ,it can be done via FCIP.
I wanted to know if we can use insert the FCIP converters into these MDS 9418 switch or do we need to have separate FCIP converters.
Do we need FCIP converters on both sides for doing replication over FCIP.
For clear understanding please see the attachment.
Will waiting for your response......
Regards,
Pranav.

Hi,
Since the 9148 does not have IP interfaces, something will be needed to tunnel the FC into IP (FCIP) such as an MDS 9222i, which has both FC interfaces and GigE interfaces and supports FCIP.
Regards,
David

Cisco MDS series switch details

Hello I 'm new to CISCO Switch world.
I 'm working as Clariion Admin and have have taken up the responsibilty to manage switches.Since,i am a novice i need little guidance on how work on cisco switches.I know the Zoning (how zoning done on MDS 9000 series switches) part ,but would like to venture deep into the details on CISCO MDS series switches.
Like, the use of Port channels etc.
Will really appreciate if some one can help me with a link or place ( not the configuration guide) where in i can find details on concentrated more on the theory part .
Thanks in advance!

Actually the best info is in the configuration guides. There are some external companies that run training sessions on the MDS to provide in depth training. There is the MDS cookbook available on CCO that is also a good point of information. If you search on 'MDS white papers' after you login into Cisco.com, there are several good documents that might meet your needs.
Hope this hels,
Mike

Does cisco MDS 9000 switch by default set a little higher R_T_TOV?

Hi, All:
Not sure anyone has done LOS for cisco MDS 9000 switch before. I tried to insert code violation which is equal to 103ms but only see link reset but not NOS/OLS. is it because that Cisco MDS has default higher ? Or can we set R_T_TOV on switch?

Hi Larry - [this|http://lists.apple.com/archives/usb/2008/Oct/msg00021.html] was the only thing I could find about transactions timing out, but since you don't have anything plugged in, it really doesn't apply. Have you tried resetting your SMU?
If that doesn't work, run Disk Utility from your install disc and see what it comes up with. Your suspicion that the MB may be faulty is probably correct, and if you are still within the 90 day warrant period, you should call the repair place and have them replace it yet again.

Cisco Prime LMS - Device discovery

How do you configure Cisco Prime LMS Device Discovery, so that you can have two credentials sets for specific "special devices" (host based statement) and one for the of the devices (global subnet statement)?
Thanks in advance.

Hi,
Here is the link to the document containing the detailed steps regarding Credential Sets.
After defining your sets then set up Credential Set Policies.
Thanks!

Using Cisco MDS 9148 switch for switching and routing

Hi Gurus,
Can you please advice me! Can i configure interface trunking, routing and dhcp services on the Cisco MDS 9148 switch?
Thanks for your response!!

Tommy,
MDS9148 is a Storage SAN Fibre Channel switch, it doesn't support Ethernet, IP, VLANs, VLAN trunking, 802.1Q, IP routing, DHCP. It's meant for Fibre Channel connectivity between Fibre Channel server HBAs and Fibre Channel storage.
Roman

Configuring Radius server with Cisco MDS - 9606 switch

Need help in configuring Radius server with cisco MDS - 9606
please let me know if any document available

rtt min/avg/max/mdev = 0.260/0.327/0.468/0.077 ms
IFCBCCEMCSW2# sh version
Cisco Storage Area Networking Operating System (SAN-OS) Software
TAC support: http://www.cisco.com/tac
Copyright (c) 2002-2008, Cisco Systems, Inc. All rights reserved.
The copyrights to certain works contained herein are owned by
other third parties and are used and distributed under license.
Some parts of this software may be covered under the GNU Public
License or the GNU Lesser General Public License. A copy of
each such license is available at
http://www.gnu.org/licenses/gpl.html and
http://www.gnu.org/licenses/lgpl.html
Software
BIOS: version 1.1.0
loader: version 1.2(2)
kickstart: version 3.3(1c)
system: version 3.3(1c)
BIOS compile time: 10/24/03
kickstart image file is: bootflash:/m9500-sf1ek9-kickstart-mz.3.3.1c.bin
kickstart compile time: 5/23/2008 19:00:00 [06/19/2008 23:56:56]
system image file is: bootflash:/m9500-sf1ek9-mz.3.3.1c.bin
system compile time: 5/23/2008 19:00:00 [06/20/2008 00:26:51]
Hardware
cisco MDS 9506 ("Supervisor/Fabric-1")
Intel(R) Pentium(R) III CPU with 1028596 kB of memory.
Processor Board ID JAB094300ER
bootflash: 250368 kB
slot0: 0 kB

LMS 3.1 Device Discovery

Hi, Having had a fully operational version of Ciscoworks on a very old server we were advised to move it to a virtual server, with a new IP address. Unfortunately since then the device discovery is not collecting data on all devices on our network, and showing them as unreachable. Even though I have replaced the IP address in the configs on the network switches.
The odd thing is that it is finding 6 devices but not all of them. Just wondered if this is something anyone out there has seen before that may be able to shed some light for me.
Thanks
Jo

Hi,
Kindly make sure all the device are snmp reachable with the new IP from the CiscoWorks Server. However more information will be found by enabling the debugging for Device Discovery and then try to run the Device Discovery again.
To enable the debug for Device Discovery , kindly to Common Services > Device & Credentials > Device Discovery > Discovery Logging Configuration > select the modules you are using for device discovery along with Data Collector, Discovery DeviceInfo , Discovery Framework, Discovery util and run the device discovery again and then send the ngdiscovery.log from CSCOpx\log location.
Also kindly make sure you have make the required changes on the Schedule discovery settings as well since Schedule discovery have its own settings for Device Discovery as compared with Ad hoc Discovery settings.
Here is the good doc link created by Sir Joseph Clarke to understand the Device Discovery and for troubleshooting purposes.
https://supportforums.cisco.com/docs/DOC-9005
Note :- ngdiscovery.log contain information about the IP address of your device in the network.
Thanks,
Gaganjeet

LMS 4.2 Device Discovery

I have used every version of LMS since 1.0 and I'm very familiar with the software. My problem is with the last few versions and the device discovery feature. It seems that in earlier versions, once the seed device was configured and devices had the appropriate community strings and cdp was enabled, the device was discovered. With these earlier versions, all of my devices were discovered. Any new device added to the network was also discovered. I never had to deal with devices that were "Unreachable" as all my devices were "Reachable". My network has not changed. I still have the same core switch ip address since version 1.0. My edge switches are still configured for the network as they always have. Same community strings. What has changed is the LMS discovery process. What's going on with these "modules"? I have 120 devices on my network and 117 will be "Unreachable". Even the core switch is unreachable. I have tried CDP, Global, seed device, DCR as seed. Still the devices are unreachable. There are no firewalls or access lists. The LMS server has complete access to all devices. Does anybody have any thoughts, comments??? Or am I just being too picky expecting the software to just work? TIA

Nothing has chnaged as considerable in discovery module since inception. Addition of new modules and algorithm has just made it stronger and smarter.
If all the devices are unknown there is certainly some issue with communication using snmp. I would request to modify and if you think multiple communities may be there try to configure all and keep the target *.*.*.*.
Following guide on LMS discovery stands from LMS old versions to till date similar on how it works :
https://supportforums.cisco.com/docs/DOC-9005
-Thanks

Cisco MDS 9148 Inventory Collection

Hi
I get the following error on an inventory collection of a MDS 9148 (NOS 5.0.4). I have tried deleting and recreating the device.
RME 4.3.1
RICS0001:Internal Error,unable to process the collected data from the device.
[ Thu Dec 30 13:48:46 GMT 2010 ],INFO ,[Thread-15],com.cisco.nm.rmeng.util.logger.XDILogger,77,com.cisco.nm.xms.xdi.pkgs.LibInventory.PortInterfaceAGI_RFC1213_HelperMethods,getIfTableEntriesFromDevice,759,Entry speed after if block: 1000000000 Index value : 101191680
[ Thu Dec 30 13:48:46 GMT 2010 ],INFO ,[Thread-15],com.cisco.nm.rmeng.util.logger.XDILogger,77,com.cisco.nm.xms.xdi.pkgs.SharedInventoryWireless.AssetAGI_CISCO_ENTITY_ASSET_Mib,g$eval,58,com.cisco.nm.xms.xdi.pkgs.SharedInventoryWireless:AssetAGI_CISCO_ENTITY_ASSET_MIB:g$eval:populating AssetAG attributes, begins...
[ Thu Dec 30 13:48:53 GMT 2010 ],ERROR,[Thread-15],com.cisco.nm.rmeng.util.logger.XDILogger,57,com.cisco.nm.xms.xdi.pkgs.SharedInventoryWireless.AssetAGI_CISCO_ENTITY_ASSET_Mib,g$eval,64,com.cisco.nm.xms.xdi.pkgs.SharedInventoryWireless:AssetAGI_CISCO_ENTITY_ASSET_MIB:g$eval:AssetAG Collection aborted com.cisco.nm.xms.xdi.ags.system.CollectionFailedDeviceSensed: java.lang.ArrayIndexOutOfBoundsException: 5 >= 5
[ Thu Dec 30 13:48:55 GMT 2010 ],INFO ,[Thread-15],com.cisco.nm.rmeng.inventory.ics.core.CollectionController,549,DP time is 20 seconds for 172.23.32.129
[ Thu Dec 30 13:48:55 GMT 2010 ],ERROR,[Thread-15],com.cisco.nm.rmeng.inventory.ics.server.InvDataProcessor,458,
java.lang.NullPointerException
at com.cisco.nm.rmeng.inventory.ics.invdata.containment.ContainmentAGInvData.compare(ContainmentAGInvData.java:103)
at com.cisco.nm.rmeng.inventory.ics.server.InvDataProcessor.processContainmentAG(InvDataProcessor.java:864)
at com.cisco.nm.rmeng.inventory.ics.server.InvDataProcessor.processInvData(InvDataProcessor.java:330)
at com.cisco.nm.rmeng.inventory.ics.core.CollectionController.run(CollectionController.java:923)
at java.lang.Thread.run(Thread.java:595)
[ Thu Dec 30 13:48:55 GMT 2010 ],ERROR,[Thread-15],com.cisco.nm.rmeng.inventory.ics.core.CollectionController,935, Exception occured in process method while processing: 172.23.32.129 null
ICSException
at com.cisco.nm.rmeng.inventory.ics.server.InvDataProcessor.processInvData(InvDataProcessor.java:473)
at com.cisco.nm.rmeng.inventory.ics.core.CollectionController.run(CollectionController.java:923)
at java.lang.Thread.run(Thread.java:595)
[ Thu Dec 30 13:48:55 GMT 2010 ],INFO ,[Thread-15],com.cisco.nm.rmeng.inventory.ics.core.CollectionController,1191,No Trap
[ Thu Dec 30 13:48:55 GMT 2010 ],INFO ,[Thread-15],com.cisco.nm.rmeng.inventory.ics.core.CollectionController,1203,CALLING SELF TEST
[ Thu Dec 30 13:48:55 GMT 2010 ],INFO ,[Thread-15],com.cisco.nm.rmeng.inventory.ics.core.CollectionController,1207,DONE WITH SELF TEST

Hi Naem,
have just read the othre reply's, the intresting bit it is in the MDF.
- <DEVICE NAME="Cisco MDS 9148 Multilayer Fabric Switch" MDFID="282867260" CISCOMDFID="282867260">
<SYSOID OID="1.3.6.1.4.1.9.12.3.1.3.841" NAME="OID:1.3.6.1.4.1.9.12.3.1.3.841" />
The device adds ok, I just can not run an Inventory but can run a backup !
Thanks

Migrating from Brocade 2800/ 3900 to Cisco MDS 9509

What is the best procedure from migrating from Brocade 2800/3900/12000 to Cisco MDS 9509 especially connected to HP-UX and AIX server.
Without any Downtime I should be able to migrate these servers.
I thought about these options:
1. Use vgexport and vgimport or exportvg and importvg (AIX) after connecting to Cisco MDS. But this requires complete downtime on the application.
2. Take one path down or HBA down and switch the cable and vgextend the devices.
Please let me know if somebody has procedure.
I was successful in HP-UX server using the second options but I cant see all the LUNs. That might be array specific problem also.
If anybody has detailed procedure on migrating this scenario. please let me know.
We are using Persistent FCIDs on our MDS switches.
Thanks in Advance

After connecting 1 cable from server HBA to Cisco S/w, why you are not able to see all LUNs, did you cross-check that...I mean HBA Configuration (max 256 LUNs) or Disk Array library driver to be installed on host side.if that can be sorted out, you can mirror the volumes across the disk arrays.are you using HDS Arrays? I don't know but what is vgextend command you are mentioning...
also did you take a reboot of server or is it online addition on new LUNs...
Also any how if you are not able to see all LUNs that is going to be an issue later as well for migration...pls cross-check that...

Scsi device discovery in Cisco MDS switches

Similar Messages

Maybe you are looking for