Maximo

Maximo

Come for answers, stay for best practices. All we're missing is you.

 View Only
  • 1.  MAS Manage & JMS Queues

    Posted Thu July 27, 2023 10:16 AM

    I used Ansible MASDEVOPS to install MAS Manage on SNO, so far so good even though resources are limited.

    When I got to testing integrations get errors about the JMS queues.

    From what I read, the bundle type "all" runs a single Manage pod, and I have one of the 2 following options

    1) Start a standalonejms bundle from the Workspace configuration

    2) Configure the queues in Maximo to use some other queue system, whether in or outside the OpenShift cluster

    I guess option 1 is the easiest, but I wonder if any of you could tell me if those are the 2 only available options



    ------------------------------
    Francois Beliveau
    ------------------------------


  • 2.  RE: MAS Manage & JMS Queues

    Posted Fri July 28, 2023 08:37 AM
    Edited by Arif Ali Fri July 28, 2023 08:39 AM

    You have two variables available.

    # export MAS_APP_SETTINGS_SERVER_BUNDLES_SIZE=jms
    export MAS_APP_SETTINGS_SERVER_BUNDLES_SIZE=dev

    jms var will get you multiple/all the bundle pods. It will also configure the JMS via xml config so you don't have to. The dev var will get you a single (all) bundle pod. On SNO, you likely should use dev.



    ------------------------------
    Arif Ali
    ------------------------------



  • 3.  RE: MAS Manage & JMS Queues

    Posted Fri July 28, 2023 04:41 PM
    Edited by Dianne Woodley Fri July 28, 2023 04:41 PM

    Hi Francois,

    Could you share me the ansible GitHub(?) link that you use for for the SNO?  Thanks!



    ------------------------------
    Dianne Woodley
    ------------------------------



  • 4.  RE: MAS Manage & JMS Queues

    Posted Fri July 28, 2023 04:56 PM

    Dianne,

    https://ibm-mas.github.io/ansible-devops/

    is where I got most of my information from (I did a LOT of try and errors before that trying to install the "core" portion with the MAS Automation operator, also with some issues).

    I am using a 16 vCPU 64GB memory single-node openshift.

    Things you may want to consider : review all the sizing parameters from the ansible playbook. Default values will kill you on a SNO ; first and foremost the cpu limits. If you think about running DB2 inside the cluster, you will need to tweak the cpu limits of mongodb and db2 ; other wise your SNO will not have enough CPU. You need to tweak the memory of DB2 too. Also check and override default disk space settings for DB2, prometheus, mongodb components, unless you have unlimited NFS storage. I am using additional drives on my VPS so I limited it to the max

    But all in all, if you make sure to adjuste CPU limits, you can run MAS Code, Manage app with 1 pod, along with it's JMS server friend pod in that SNO and IBM says it supports up to 70 concurrent users



    ------------------------------
    Francois Beliveau
    ------------------------------