The physical connectivity of fabric interconnect with MDS and failover please suggest

Dear team

We have 2 FI and 2 MDS 1 SAN

Currenlty connectivity is

2 physical connections of FI - has direct MDS1

2 direct physical links to FI - B MDS2

SDM 1 connection to the main controller of SAN

SDM 1 connection to the secondary controller SAN

Connection of 2 MDS to the PDC of SAN

2 MDS to the secondary controller SAN connection

Hope the above connectivity is good?

We had watched

http://www.Cisco.com/en/US/prod/Collateral/ps4159/ps6409/ps5990/white_paper_c11_586100.html (as a team always preferred to go through the standards of cisco)

We could consolidate the FI and the primary and subordinate, even here for MDS in we need FI. or it will be through FIS?

If not these 2 MDs they work as independent but share the configs that happens between them?

Now this never changes (creation area) we do on MDS1 hope that will get relpicated in MDS2 through FIS?

If MDS1 fails will be all the configuration is available on MDS2 and yet infra smooth it won't work? and vice versa?

All the additional steps to be performed to achieve this?

Which is the best way to do this please suggest

Thanks and greetings

Jose

Hi Jose.

Physically, your connection is good.

On the side of the UCS, the role of "Primary" and "Subordinate" refer only to the management of the system and the device that actually performs UCS Manager.

Each MDS device will have a separate configuration (zoning).  It's different between the 2 devices.

The blade itself will be a connection on each side, or the "fabric SAN.

for example

+----+

+---+    +-----+

|   |    |     |

|   |    |     |

|   | SAN |     |

| +->----<--+ >

| |         |  |

| |         |  |

+---+-+-+    +--+--+--+

|       |    |        |

| MDS - A |    | MDS - B |

+--^-^--+    +---^-^--+

| |           | |

+--+-+--+    +---+-+--+

|       |    |        |

| UCS - A |    | UCS - B |

+--+----+    +-----+--+

|               |

|               |

VSAN100 +-+ | VSAN200

|    |     |    |

+----><>

| Blade |

+-----+

The blade will have a HBA on FabricA (VSAN100) and FabricB (VSAN200)

Each HBA has a different WWPN, and on the Bay of SAN, each controller will have a WWPN

So on MDS - A, the zoning will be:

Blade WWPN

WWPN primary storage has

WWPN secondary storage has

MDS - B, the zoning will be:

Blade WWPN B

Storage WWPN primary B

Storage WWPN secondary B

So the configuration is * not * synchronized between both MDS devices, but each of them have visibility to the device of vHBA of blades.  At the level of the blade, the software of multiple paths on the operating system will handle any failover.

On the UCS, we would usually use a channel from Port to the MDS.

Tags: Cisco DataCenter

Similar Questions

  • Connection of fabric interconnect with the Cisco MDS switch

    I'm new to the unified fabric and try to understand it.

    There are clients who connect fabric connects directly to switch MDS without any Nexus 5 K between the two.  If the fabric of interconnections are able to provide connectivity to the LAN and SAN Islands then why you install Nexus 5 K between two switches to add an extra jump.

    Thank you

    Hello

    Most of the clients do the following:

    -connect FI upstream for example N5k with vPC for Ethernet (IP)

    -connect FI upstream for example MDS for the classic FC

    This implies that between the blade into the chassis and FI you run FCoE, integrating CF and Ethernet on the same means of transport; then, on the FI, FC and Ethernet is disintegrated.

    The other choice would be Multihop FCoE, linking FCoE also FI and N5k and make the disagregation it.

    However, if your storage subsystem is FC and FCoE, I wouldn't recommend Multihop FCoE (I see only a minority of clients to do).

    Walter.

  • Fabric 6248 with MDS

    Hi all

    We are at the stage of inter-connecting simple IBM V3700 SAN box with double FI 6248 (redundant) via unique Cisco MDS 9000 SAN Switch with single UCS 5108 chasis.

    We have four blades B200 M3 in UCS 5108. The blades have no local hard disks. Our requirement is to install hyperV/Windows on IBM storage.

    I did the physical connections between IBM and MDS & MDS FI 6248 and FI 6248 dual Core 6509 VSS enabled.

    I don't have issues with part FI 6248 and LAN, I have doubts about FI 6248 with SAN kindly advise.

    6248 FI are default mode (Host End). I need to change FI on switch mode?
    I change ports Uplink FC, unified FI ports connected to the MDS are configured as storage CF?
    How vNIC, vSAN (vSAN how do I need), vHBA & Service profile works to each other? can any advise please?
    What steps do we need to set up in FI & MDS so that we can use SAN storage for hyperV/Windows installation?

    It's my first deployment of the UCS 5108, to apologize to ask fundamental questions.

    Kind regards

    1. what would be the impact of multiple VSANS vs default vSAN, I plan to use default vSan in the two FI-6248 & I can't find any problem with this, please advise.

    Please do not use default vsan (vsan nr or 1); best practices: create different VSAN fabric A and B; e.g. vsan 10A, vsan 20 b

    2 - pool WWPN, should I create two different WWPN pool 1 for FI-6248-A & others for FI-6248-B? I have only a UCS 5108 chasis, we cannot set an unique WWPN for the two FI-6248 Pool? What is the best practice? I can't find any problem with single WWPN pool. Please advise.

    Best practices: use pools of different pwwn to fabric A and B; useful for troubleshooting; in General: all pools must also encode a field id UCS (your starting with 1). just in case in the future, you have another! See attachment.

    3 - incase, if I will configure 1 WWPN pool would I face any problem in mapping the area in MDS? According to my understanding there will be no problem because the area is just the initiator (Server Blade) mapping & target (IBM Storage). Keep in mind, I only Cisco MDS SAN Switch NOT redundant. Please advise.

    Even with an SDM, you create a virtual double fabric with 10 and 20 VSANs. zoning is done by VSANs, zones and activation as well!

    what adapter policy 4 - Creating Service profile should I choose in the menu Dynamics vNIC? I want to install Windows on Hyper V environment, but I'm not hyper V option. It has linux, VMWare, VMWare passthrough, window, default & some HyperV more but not?

    You do not create dynamic vNIC! This only applies to the VM - FEX, resp. SR - IOV. Hyper-V means political adapter = window

    Good luck

    Walter.

  • Cannot select the Panel of configuraiton of 'Security' with Down and Up the keys on the keyboard

    After you install BlackBerry10Simulator-Installer-BB10_0_06-384-Win-201207130120.exe on Windows XP (Vmware player 3.1.3), I can't select the Panel of configuraiton of 'Security' with Down and Up the keys on the keyboard, because "Screen" is the last item on the screen.

    It works with a mouse. Thank you.

    I use a computer thinkpad notebook and always use the navigation button (red button), so I now have to use a mouse.

  • 6248 fabric interconnects FC interfaces and MDS ports license

    Hello

    I need help. I need to configure port-channel of CF between 2 x 9148 MDS and 2 x Cisco fabric interconnects. My main problem here is interfaces CF appear not on the fabric of interconnections (UCS Manager). I don't see ethernet ports. I have to turn on the unified ports or do I license for CF?

    I also connected the ds-sfp-fc8g-sw FPS but I get the error "operation status failed, reason: SFP Validation failed.

    Another thing, all 48 FC ports on switches DMS 9148 acquire the license. Aren't they supposed to come with 16 ports enabled by default?

    Any ideas?

    Hi Selby

    6248 FI unified ports, by default, all are Ethernet; so as a first step, you must configure the FC ports.

    You just need Ethernet port licenses, which are also universal.

    See, for example.

    https://supportforums.Cisco.com/message/3763853#3763853

    For SDM 9148 licenses take a look on

    http://www.Cisco.com/c/en/us/products/collateral/storage-networking/MDS-9148-multilayer-fabric-switch/data_sheet_c78-571411.html

    There are packages with

    Cisco MDS 9148 48 - Port Multilayer Fabric Switch with 16, 32 or 48 active ports of 8 Gbps

    Good luck

  • 8900 comm http hangs after the second connection (do not view with the 83XX devices or 8900 Simulator)?

    Hello

    I currently have a j2me application that is deployed in production on different devices of BB. The application is pretty simple and mainly collects the data of the user (via text fields, drop-down list boxes etc.) and send these data to our server via an http connection

    The application was deployed in production for more than 6 months and work with end users using various BB devices, like the 8310, 8320 and 9000.

    Our end users now upgrade to the 8900 and we assumed that our application would work in the same way as the 8320-I even tested my app on the Simulator to 8900 and everything works fine.

    However, using the app on a real 8900 device connected to the network Vodafone here in the United Kingdom has cast some weird problems. The app installs fine OTA, communicates initially very well (as we connect via an http post), but then any attempt further communication over http hang everything simply no error thrown?

    Closing the application does not have any change, and blocking any attempt at communication. If I restart the device (battery etc) and start the application then I can connect once more, and then the app just crashes again.

    I can't find the cause, and the fact that other physical devices and even the Simulator to 8900 work very well is weird!

    Any help or pointer to start looking in my code/settings would be appreciated?

    Daniel

    Please make sure you close your connection. or check if the current connection is close before opening another.

  • The network connection issues of comments with EA6300 (Linksys Smart wifi router AC1200)

    I have a friend who runs a Bed and Breakfast, and it is problems with guests to be able to connect to the Guest network.  She bought this router on the recommendation of Best Buy Geek Squad, after that she had a few problems with his predecessor.

    Everything is configured correctly, the router seems to work very well, especially with the normal network.  Network comments can be seen, with a decent amount of signal strength, but when people try and connect to the network and try to use it, they are never invited for comments login screen.

    I was able to verify that it works in fact - I helped set it up, but apparently there are several people who use Android phones and tablets and iPads that do not receive a prompt to enter the password of comments and continue.

    Is there any manual addresses can be entered in a browser to access the web connection screen invited?

    OR even better, is there a way to assign the password to the guest SSID as a normal wireless access point rather than allowing customers to connect to a password-promptless guest SSID and must try to finagle screen comments for pop up password?

    It's very frustrating for both my friend who runs a business and myself who are in favour.  The router is defective?  Y at - it a new firmware?  Any help would be appreciated.

    Hi troybenedict. For Android tablets and phones, there is an option there to forget the network.  Let them do first and connect again.

    In addition, they should asked for the password only when they begin to access Web sites.  They do not have that?

    You can also check the number of comments allowed Guest Access tab. There are limits the number of authorized clients.

  • Disable the automatic connection for Windows 7 with 1 single active account.

    I've searched high and low on Google but all I could find are methods to allow the automatic connection.

    Is it possible to have Windows 7 load to the top of the selection of Auditors even screen with 1 single active account?

    I remember what is a function available in XP.

    There are probably more simple methods, but it certainly works: create a fake account, then bring it to be hidden from the logon screen. The following batch file will do it for you when it is run under elevated privileges. Of course, you must disable the usual automatic connection setting via this command:

    Control userpasswords2

    And here's the batch file. Use the copy & paste - do not retype! Note that the line starting with
    "echo > % RegFile" is a very long line that ends in 'UserList]. " Not to break upward.

    off @echo
    Set RegFile="%temp%\reg.reg".
    echo > RegFile % REGEDIT4
    echo > RegFile % [HKey_Local_Machine\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon\SpecialAccounts\UserList]
    echo > RegFile % "DummyUser" = DWORD: 00000000
    Regedit RegFile %
    NET user DummyUser supercalifrag / add

  • The db connection takes too long with connection of shared servers

    Hi all

    Operating system: Solaris SPARC 64-bit 11
    DB: 2 node RAC Oracle 11.2.0.3.5

    After a migration from a database of 40GBs of a 9i database to an 11g database, he is takin too long to connect to the db, which uses the shared server. At present, it is not all applications using the Pb, just me. Do a few tests, I saw that when I use a dedicated connection, I can connect almost instantly, but by using the shared connection, it takes almost 4 seconds to connect via sqlplus. Here are a few shared server settings:
    shared_server_sessions               integer
    shared_servers                          integer     1
    max_shared_servers                   integer     20
    dispatchers                               string      (protocol=TCP)(disp=5)
    max_dispatchers                        integer     5
    
    lsnrctl service
    
    
      Instance "INSTANCE", status READY, has 6 handler(s) for this service...
        Handler(s):
          "DEDICATED" established:5 refused:0 state:ready
             LOCAL SERVER
          "D004" established:93 refused:0 current:0 max:1022 state:ready
             DISPATCHER <machine: orahml01, pid: 4073>
             (ADDRESS=(PROTOCOL=tcp)(HOST=hostname)(PORT=65402))
          "D003" established:85 refused:0 current:0 max:1022 state:ready
             DISPATCHER <machine: orahml01, pid: 4070>
             (ADDRESS=(PROTOCOL=tcp)(HOST=hostname)(PORT=44055))
          "D002" established:90 refused:0 current:2 max:1022 state:ready
             DISPATCHER <machine: orahml01, pid: 4066>
             (ADDRESS=(PROTOCOL=tcp)(HOST=hostname)(PORT=41277))
          "D001" established:89 refused:0 current:1 max:1022 state:ready
             DISPATCHER <machine: orahml01, pid: 4062>
             (ADDRESS=(PROTOCOL=tcp)(HOST=hostname)(PORT=61887))
          "D000" established:91 refused:0 current:0 max:1022 state:ready
             DISPATCHER <machine: orahml01, pid: 4060>
             (ADDRESS=(PROTOCOL=tcp)(HOST=hostname)(PORT=44222))
    
      INSTANCE =
        (DESCRIPTION =
          (ADDRESS = (PROTOCOL = TCP)(HOST = hostname-scan)(PORT = 1521))
          (CONNECT_DATA =
            (SERVER = SHARED)
            (SERVICE_NAME = INSTANCE)
          )
      )
      
    INSTANCE1 =
      (DESCRIPTION =
        (ADDRESS = (PROTOCOL = TCP)(HOST = hostname-vip)(PORT = 1521))
        (CONNECT_DATA =
          (SERVER = DEDICATED)
          (SERVICE_NAME = INSTANCE)
          (INSTANCE_NAME = INSTANCE1)
        )
      )  
      
    
    INSTANCE2 =
      (DESCRIPTION =
        (ADDRESS = (PROTOCOL = TCP)(HOST = hostname-vip)(PORT = 1521))
        (CONNECT_DATA =
          (SERVER = DEDICATED)
          (SERVICE_NAME = INSTANCE)
          (INSTANCE_NAME = INSTANCE2)
        )
      ) 
    The alert.log doesn't show anything wrong.

    It can be the cause of this problem with shared server?

    Thanks in advance.

    Published by: Fabricio_Jorge on 03/27/2013 06:14

    You have five dispatchers/dispatchers and a shared server. A Dispatcher and five servers shared would be a more normal setting. Try to launch a few servers more shared, perhaps

    alter system set shared_servers=20;
    

    --
    John Watson
    Oracle Certified Master s/n
    http://skillbuilders.com

  • Just as a reference: how to FIX the annoying 'White Screen Of Death"with Silverlight and"Access denied"error with computer management.

    Well, I just hope that someone who has met these two mistakes will find my dose. It might save a lot of pointless research on the net.

    SYMPTOMS:
    -When you try to open a Silverlight application in your browser (Firefox, Chrome, IE, etc), your browser displays only a white area. When you right click on the white area, you can click on the Silverlight properties.
    -If you try to start "Computer management" by clicking on the 'My Computer' icon and select 'Manage', you get the following error: 'access denied '.

    CAUSE:
    The following errors are usually the result of a bad environment variable. In this case the "windir" environment variable does not work correctly. You can try running the management of the computer to start > run and entering:
    %windir%\system32\mmc.exe/s %windir%\system32\compmgmt.msc

    Notice the between the characters of percentage windir environment variable. If you run the above code, you will probably get an error saying that MMC cannot open the file. Look closely at the path, it's back!
    You will probably find any % somewhere.

    What actually happened here is the variable windir made reference to another variable: systemroot
    Why this creates problems, I'm not sure. Also, I don't understand what made then variable windir change.

    SOLUTION:
    -From your desktop or Explorer: right-click on "My computer" and select "Properties".
    -Click on the "Advanced" tab and then click "Environment Variables".

    http://pcauthorities.com/images/advanced-system-properties-XP.jpg

    -Did the windir variable. Together to "C:\Windows" do not think that it is already set to "C:\Windows", because windows then tells you. Double-click it to know for sure.

    http://www.adminxp.com/windowsvista/envvariabl.PNG

    -See, I told you! Change this to "C:\Windows".

    http://www.adminxp.com/windowsvista/editenvvar.PNG

    Hi Jooch,
     
    Thanks for posting such a detailed response, including symptoms, Cause, and resolution to "White Screen Of Death" with Silverlight and "Access denied" error with computer management"issues.
     
    I appreciate your time and effort to do the same. This could help other users of the forum as well.
  • (Redirected) Cannot boot from the HARD drive on dell n5010 with SSD and HARD drive installed

    I have my old dell n5010 (15R Inspiron) with SSD (Samsung EVO 840 120 G) installed on the main drive Bay and the HDD (Samsung HJJ 500 G) on the optical drive Bay.

    It is a typical way to add an SSD on such an old laptop (bought from dell.com.cn in 2011).

    My Windows 10 operating system has been installed on the SSD using Samsung Data Migration software, so the previous C drive was pointless. So I formatted and installed an operating system Ubuntu 14.04 on it, hoping the grub OSs double beginning located on the HARD disk.

    However, I find no the Dell bios boot option to boot from the HARD drive installed in my optical drive Bay. All the choices are 'Hard Drive', 'Network' and 'Diagnostic '.

    But I was definitely sure that my bios program has detected the SSD and HDD with information on the main menu of the bios.

    And the LED on the optical drive Bay was immediately once I pressed the Start button on my laptop.

    As a general rule, all things were right except for options of the bios boot.

    Is there a way I can boot from HDD (with installed Ubuntu) while remaining has my SSD on the main drive Bay? I don't want to swap their positions, because the SSD is super fast on the main drive Bay.

    I have alse won't migrate 10 Windows to the HARD drive and reinstall Ubuntu on the SSD, because I'd like my windows be loaded quickly.

    Someone at - it the same problem with me?

    Any suggestion is appreciated!

    As you have a laptop, better make this laptop in the computer Forum post here:

    http://en.community.Dell.com/support-forums/laptop/default.aspx

    Bev.

  • I've updated the plan of the photographer and since the download, I had problems nonstop with fluidity and now the bridge is not displayed large previews, a few thumbnails. I restarted and reset the tool, is there anything else I can do or should I cancel

    I've updated the plan of the photographer and since the download, I had problems with non stop with fluidity and now the bridge is not displayed large previews, a few thumbnails. I restarted and reset the tool, is there anything else I can do or do I cancel my subscription to the CC and return to the CS6?

    The problem of fluidity can be the result of a graphics card problem:

    Try to uncheck the box in the preferences > performance > graphics processor use, then by restarting Photoshop. If this solves the problem, check with the manufacturer about a possible update.

    On the bridge: when you select an image and press the space bar, what happens?

  • Integration of the Probs in Jdev 11 g with Struts. Help, please

    Hi all

    I'm doing a trial using 10.3 Struts application in Jdev 11 g.
    I created the web application wizard Jdev using but don't select Struts as technology. Then, I downloaded the Jar of Struts files and them copied in the directory WEB - INF lib in my sample application.
    But when I am trying to importorg.apache.struts.action.ActionForm and expand ActionForm in my Bean I get below error:


    Error (2.32): package org.apache.struts.action does not exist
    Error (3.36): cannot find symbol


    My Bean is simple as below:
    package projectname.nano.common;
    org.apache.struts.action.ActionForm import;
    SerializableAttribute public class InputFormBean extends ActionForm
    {

    }

    Then I thought that this compiler may not be able to integrate the copied jars. I went to Applications-> the Defalut project properties and include the my app lib directory.
    Probs still persists. Please help as I m stuck here.

    Check out this thread where I described how to setup a struts application in 11g
    Re: Error when deploying JDeveloper 11.1.1.3.0 on WebLogic 10.3
    Timo

  • I want to USB Hub with sync Ipad pro please suggest compitable for this device

    I want to buy power USB Hub with Sync to IPad Pro please suggest me the device appropriate for this work

    A powered hub will do, choose one.

  • Physical connections using 4 counters with NI 9411 cDAQ chassis 9178

    Hello

    Currently I have a NI 9411 module in a 9178 cDAQ chassis.  I use 2 meter to read the frequency measures.

    The NI 9411 pinout diagram (see pdf file attached to this post), here are the links that I did:

    CTR 0 SRC, pin 1 ground COM

    DOOR of the CTR 0, pin 2 to the first signal of frequency

    CTR 1 CBC, PIN 6 COM/ground

    CTR 1 DOOR, pin 01:53 frequency signal

    The problem now is that I had like to use meters of 3rd and 4th in my 9178 cDAQ chassis, but I am confused how to wire

    frequency signals in 3rd and 4th, because CTR2 GATE is pin 1 and GATE CTR3 is pin 6.  Can I change the assignment of pins

    to use the other pins in the NI 9411 module for frequency counter measures?  I didn't understand how to do this in MAX

    Configuration.

    Thank you

    Anna

    Hi EETDer,

    Using 'C-Series Signal connections for counters that you found is a good resource.  One thing I want to point out, that's a measure of the frequency on a single meter, simply connect the source of the signal up to the door of the RTC.  Internally, the module will forward a database internal time appropriate to the source of counters, so DO NOT want to connect the Source CTR to the mass.  On your device, you will be just wire each signal to the source of the meter, so only utilizting pines 1,6,3,8 for your 4 frequency measurements.

Maybe you are looking for