Topic
  • 10 replies
  • Latest Post - ‏2013-12-30T13:38:04Z by kenhygh
DPUser77
DPUser77
70 Posts

Pinned topic Error in setting gateway node for api managment v2.0

‏2013-12-24T14:53:38Z |

Hi,

I am not able to set xi52 virtual box as gateway node in api management console. On giving dp admin credentials on gatewaynode_1.png (attached) i get a time out exception but a separate domain gets created in xi52 box as api_xxxxx.

After some googling i tried below things :

1. Tried to configure a virtual ip (which is entered while configuring load balancing (AO feature)) in 'Stand by control' for eth0 , but as soon as i click on apply buttong etho goes down and 'show int' command does not shows any ip which leads to  appliance restart. so not able to configure AO.

2. Tried to configure two services by referring this doc http://www.redbooks.ibm.com/15E39609-5DEF-4044-B5BB-D6FF8ED2D86E/FinalDownload/DownloadId-96E03A0E4571838433FBB007C992AF8E/15E39609-5DEF-4044-B5BB-D6FF8ED2D86E/redpieces/pdfs/sg248193.pdf  in api domain with port # 443 and 2443 for gate way traffic  to enable traffic on these ports .  but redployment gave another exception "
APIGW_REDEPLOY_DELETE_ERROR (com.ibm.wsci.dynamic.configuration)  . So please confirm if i need to create services in default domain  ??

Many a times i deleted the api domain for retry. I found that some time newly created domain goes into quiescence mode

 

Please share your valuable thoughts.

Attachments

  • kenhygh
    kenhygh
    2086 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-24T15:13:28Z  

    I haven't done a lot with API management, but...

    First you'll need to troubleshoot the AO config. Does your DP have AO? Status->System->Device Features.

    For your API services, you'll never configure them in the DataPower console, only through the API Management console. Which, as you noticed, creates its own domain. 

    You can configure any other services as you would in any other datapower device, in your own domain.

  • DPUser77
    DPUser77
    70 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-24T15:18:29Z  
    • kenhygh
    • ‏2013-12-24T15:13:28Z

    I haven't done a lot with API management, but...

    First you'll need to troubleshoot the AO config. Does your DP have AO? Status->System->Device Features.

    For your API services, you'll never configure them in the DataPower console, only through the API Management console. Which, as you noticed, creates its own domain. 

    You can configure any other services as you would in any other datapower device, in your own domain.

    Hi Ken,

    AO is enabled in my  virtual box. PFA the license command results in  XI52# console.

     

    Attachments

    Updated on 2013-12-24T15:19:36Z at 2013-12-24T15:19:36Z by DPUser77
  • DPUser77
    DPUser77
    70 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-25T20:11:36Z  
    • DPUser77
    • ‏2013-12-24T15:18:29Z

    Hi Ken,

    AO is enabled in my  virtual box. PFA the license command results in  XI52# console.

     

    Please share your thoughts on this.

  • kenhygh
    kenhygh
    2086 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-26T14:56:53Z  
    • DPUser77
    • ‏2013-12-25T20:11:36Z

    Please share your thoughts on this.

    What's the error you receive when you try to configure AO?

  • DPUser77
    DPUser77
    70 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-27T12:41:06Z  
    • kenhygh
    • ‏2013-12-26T14:56:53Z

    What's the error you receive when you try to configure AO?

    Hi Ken,

    Thanks !! for replying


    For AO features i tried following things,please provide your inputs on below points and correct me if am wrong :

     

    1. No Stand by control was getting created by a hit from API console,so i tried creating one manually to check AO.But getting this in the status logs "5:05:50 PM mgmt notice    141761      0x00350015 interface (eth0): Operational state down ".

      After referrring to an AO article i found that "Specify a unique Group Number that is not currently in use within your network and with a value higher than 20 to prevent conflict with network routers and switches. This information is generally available from your network administrator"

     earlier we were giving the last # of ip for a reference .So i will try to get the appropriate group #  

     Could you please confirm if this has to created manually or by api call and if we are following correct process.

    Note : we are not using virtual ip VMnet8 created by vm ware player so that team members can access this machine over a n/w


    2. Inorder to run .vmx file we are using a virtual player with limited # of cpu's, which is contributing to throttling log entries in default logs.
    I disabled/changed throttling settings for testing purpose, but same result. Can this be an issue in setting the gateway node ??

    "0x80400036][system][critic] throttle(Throttler): tid(1199): CPUs missing, expected 4, have 2."


    Please share your valuable inputs on above points.

    Regards,
    Simmer

     

  • HermannSW
    HermannSW
    6201 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-27T22:25:11Z  
    • DPUser77
    • ‏2013-12-27T12:41:06Z

    Hi Ken,

    Thanks !! for replying


    For AO features i tried following things,please provide your inputs on below points and correct me if am wrong :

     

    1. No Stand by control was getting created by a hit from API console,so i tried creating one manually to check AO.But getting this in the status logs "5:05:50 PM mgmt notice    141761      0x00350015 interface (eth0): Operational state down ".

      After referrring to an AO article i found that "Specify a unique Group Number that is not currently in use within your network and with a value higher than 20 to prevent conflict with network routers and switches. This information is generally available from your network administrator"

     earlier we were giving the last # of ip for a reference .So i will try to get the appropriate group #  

     Could you please confirm if this has to created manually or by api call and if we are following correct process.

    Note : we are not using virtual ip VMnet8 created by vm ware player so that team members can access this machine over a n/w


    2. Inorder to run .vmx file we are using a virtual player with limited # of cpu's, which is contributing to throttling log entries in default logs.
    I disabled/changed throttling settings for testing purpose, but same result. Can this be an issue in setting the gateway node ??

    "0x80400036][system][critic] throttle(Throttler): tid(1199): CPUs missing, expected 4, have 2."


    Please share your valuable inputs on above points.

    Regards,
    Simmer

     

    I can provide answer on (2.):

    This message can only be seen on a physical appliance in case some hardware (CPU) is broken.

    You see it because your DataPower installation violates the minimal VE requirements:
    4GB main memory, 4 vCPUs



    Only as a DataPower developer I do see similar (funny) messages when working on my Thinkpad:

    ...

    20131227T221403Z [0x80400036][system][critic] throttle(Throttler): tid(1199): CPUs missing, expected 24, have 8.
    20131227T221403Z [0x80400037][system][critic] throttle(Throttler): tid(1199): Memory missing, expected 100663296 KB, have 16286568 KB.
    ...

     

    These two event-suppressions for default log prevent default log gettting polluted with these messages:

    0x080400036 (CPUs missing, expected %1$d, have %2$d.)
    0x080400037 (Memory missing, expected %1$d KB, have %2$d KB.)

     

  • DPUser77
    DPUser77
    70 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-28T04:25:34Z  
    • HermannSW
    • ‏2013-12-27T22:25:11Z

    I can provide answer on (2.):

    This message can only be seen on a physical appliance in case some hardware (CPU) is broken.

    You see it because your DataPower installation violates the minimal VE requirements:
    4GB main memory, 4 vCPUs



    Only as a DataPower developer I do see similar (funny) messages when working on my Thinkpad:

    ...

    20131227T221403Z [0x80400036][system][critic] throttle(Throttler): tid(1199): CPUs missing, expected 24, have 8.
    20131227T221403Z [0x80400037][system][critic] throttle(Throttler): tid(1199): Memory missing, expected 100663296 KB, have 16286568 KB.
    ...

     

    These two event-suppressions for default log prevent default log gettting polluted with these messages:

    0x080400036 (CPUs missing, expected %1$d, have %2$d.)
    0x080400037 (Memory missing, expected %1$d KB, have %2$d KB.)

     

    Yes Hermann !!

    But, can this be an issue in setting the gateway node ??

    we are trying to install  a vmware workstation with more # of cpu's.

    Please share your valuable thoughts

  • kenhygh
    kenhygh
    2086 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-30T12:55:26Z  
    • DPUser77
    • ‏2013-12-27T12:41:06Z

    Hi Ken,

    Thanks !! for replying


    For AO features i tried following things,please provide your inputs on below points and correct me if am wrong :

     

    1. No Stand by control was getting created by a hit from API console,so i tried creating one manually to check AO.But getting this in the status logs "5:05:50 PM mgmt notice    141761      0x00350015 interface (eth0): Operational state down ".

      After referrring to an AO article i found that "Specify a unique Group Number that is not currently in use within your network and with a value higher than 20 to prevent conflict with network routers and switches. This information is generally available from your network administrator"

     earlier we were giving the last # of ip for a reference .So i will try to get the appropriate group #  

     Could you please confirm if this has to created manually or by api call and if we are following correct process.

    Note : we are not using virtual ip VMnet8 created by vm ware player so that team members can access this machine over a n/w


    2. Inorder to run .vmx file we are using a virtual player with limited # of cpu's, which is contributing to throttling log entries in default logs.
    I disabled/changed throttling settings for testing purpose, but same result. Can this be an issue in setting the gateway node ??

    "0x80400036][system][critic] throttle(Throttler): tid(1199): CPUs missing, expected 4, have 2."


    Please share your valuable inputs on above points.

    Regards,
    Simmer

     

    Simmer,

    Sorry I didn't respond earlier, I've been out sick a few days.

    1. I wouldn't expect this to be done from the API console, but from the Environment console. I have never installed API Management myself, so I'm not sure if the install directions have you configure standby control from within the DataPower console directly or through the Environment console.

    2. Group numbers are not the same thing as IP addresses, and I have no idea how this should run in VMWare Player (which isn't supported). Try setting it to something around 21 or something (single-digit numbers are often used by Cisco network devices).

    3. Like Hermann pointed out, you need the minimum hardware (and software) to expect this stuff to run. No telling what odd errors you might get without it.

  • DPUser77
    DPUser77
    70 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-30T13:21:12Z  
    • kenhygh
    • ‏2013-12-30T12:55:26Z

    Simmer,

    Sorry I didn't respond earlier, I've been out sick a few days.

    1. I wouldn't expect this to be done from the API console, but from the Environment console. I have never installed API Management myself, so I'm not sure if the install directions have you configure standby control from within the DataPower console directly or through the Environment console.

    2. Group numbers are not the same thing as IP addresses, and I have no idea how this should run in VMWare Player (which isn't supported). Try setting it to something around 21 or something (single-digit numbers are often used by Cisco network devices).

    3. Like Hermann pointed out, you need the minimum hardware (and software) to expect this stuff to run. No telling what odd errors you might get without it.

    Hi Ken,

    No Problem . I hope you are doing good now.

    I have asked our network team to provide an ip segment dedicated for our PoC's only so that every node will be on the same subnet.

    And they  referrred group # to be equivalent to v-lan # .

    I will try configuring again and post my obseravations here.

    On the other note, does IBM provide any detailed documentation on this API management apart from infocenter. ??

    Thanks for your valuable thoughts

    Updated on 2014-01-01T08:34:33Z at 2014-01-01T08:34:33Z by DPUser77
  • kenhygh
    kenhygh
    2086 Posts

    Re: Error in setting gateway node for api managment v2.0

    ‏2013-12-30T13:38:04Z  
    • DPUser77
    • ‏2013-12-30T13:21:12Z

    Hi Ken,

    No Problem . I hope you are doing good now.

    I have asked our network team to provide an ip segment dedicated for our PoC's only so that every node will be on the same subnet.

    And they  referrred group # to be equivalent to v-lan # .

    I will try configuring again and post my obseravations here.

    On the other note, does IBM provide any detailed documentation on this API management apart from infocenter. ??

    Thanks for your valuable thoughts

    I don't know of any docs other than the infocenter, but I'd expect a redbook to become available sometime.