Maximo

Maximo

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

 View Only
  • 1.  Automatically text & email me if Maximo is down

    Posted Mon May 24, 2021 10:16 AM
    Edited by System Admin Wed March 22, 2023 11:48 AM

    My organization had an incident recently where Maximo went down (after hours) due to Windows updates on a server. We didn't know about it until Monday morning - when users started emailing us about the issue.

    Is there an industry-standard product that can be used to notify IT if Maximo goes down?

    For example, is there a product you can recommend that would check every hour if Maximo us up, and if it's down for > 30 mins, then it would text & email a list of IT staff?

    Thanks.


    #Maximo
    #AssetandFacilitiesManagement


  • 2.  RE: Automatically text & email me if Maximo is down

    Posted Mon May 24, 2021 10:31 AM
    We offer a tool called MaxAware that monitors a number of things about your Maximo environment including system availability (even down to individual JVMs). You can have a look here or visit our YouTube channel for some demo videos.

    ------------------------------
    Tim Ferrill
    Solutions Consultant
    Intelligent Technology Solutions
    tferrill@webuildits.com
    www.webuildits.com
    @tferrill/@webuildits
    ------------------------------



  • 3.  RE: Automatically text & email me if Maximo is down

    Posted Mon May 24, 2021 11:48 PM
    First of all not great that the system went down.  Hope there was no damage caused by the outage.  

    This isn't really an issue that Maximo can address, once Maximo is down it's essentially out of action and wont tell you anything.  Think of it like an unconcious patient in a hospital.  They have an EKG hooked up and it will alarm if the patients heart stops....  Most organisations will have systems that monitor the health other systems using a variety of techniques.  Your IT dept should have something along theses lines and if they don't they should have one.  Email is usually a given option and there are are online services to convert email to SMS if direect SMS is not supported by the tool.   Also don't forget Maximo can be set up in a Highly Availble configuraton so if one server/JVM goes down the other can carry the load.  This also needs to be factored into patching regiemes - if all servers get patched at the same time you will still have an outage.         
    Also consider what you startup process is.  Does Maximo automatically start when the OS starts, how do you ensure the DB is active before the Application starts etc.

    ------------------------------
    Michael Kasteel
    Director
    ISW
    ------------------------------



  • 4.  RE: Automatically text & email me if Maximo is down

    Posted Tue May 25, 2021 08:11 AM
    There are a lot of products that are used for monitoring applications. So many that I wouldn't feel comfortable making a suggestion of what to utilize. 

    The product I would suggest for alerting, however, is PagerDuty (https://www.pagerduty.com/). We have multiple on call rotations (infrastructure, networking, application support, development, etc.) and when any of our monitoring systems alerts to an issue, the appropriate team is notified to address the issue. This can send push notifications to a device, call, text, email, etc. and ensures that events are acknowledged and resolved. It allows for setting up rules to escalate when someone doesn't acknowledge and even automatically unacknowledging after a period of time to help avoid issues where someone acknowledges and falls back asleep. As far as alerting goes, it pretty much is the industry standard.

    ------------------------------
    Steven Shull
    Director of Development
    Projetech Inc
    ------------------------------



  • 5.  RE: Automatically text & email me if Maximo is down

    Posted Tue May 25, 2021 10:58 AM
    I have seen organizations use Microsoft's System Center Operations Manager (SCOM) and Splunk for log and process monitoring and alerting. Those tools seem to work well.

    As an interim solution, you may also be able to set up a database job that checks the SERVERSESSION table every so often and lets you know when SERVERTIMESTAMPs are too old.

    ------------------------------
    Blessings,
    Jason Uppenborn
    Sr. Technical Maximo Consultant
    Ontracks Consulting
    ------------------------------



  • 6.  RE: Automatically text & email me if Maximo is down

    Posted Tue May 25, 2021 04:14 PM
    Hi,

    It starts with thinking about what is up what is down; the web server will either return something or nothing; if it returns something the most common error would be error code="500". Most monitoring solutions consider a healthy resource to be alive and the responds code = 200 or 302 (redirect).

    Now from your IT perspective that's nice but does not cover everything; one example is administrative mode; if you leave that on by accident; your monitor won't pick up that the end customer can't login :). If you use AD/OAUTH and the end customer can't authenticate you get where I'm going...

    Additionally you could be behind a strict firewall; so cloud based solutions might be able be used to orchestrate responds but actual identification should be done from within.

    It all depends on your use case; plan accordingly. We use a docker image to do this for us for example, it is a script that actually does a login every 15 minutes from within our network it sends the results to our telemetry platform. Our Public Edge is not a FW > Maximo web server configuration so we monitor health of the edge components, network routes and infrastructure and other dependent services individually in a continuous mode.

    The telemetry platform then makes decisions based on the metrics and notifies the appropriate persons via a paging service. Now when I say this please keep in mind that we monitor health of service and micro services not just up/down our selves so our use case might not apply.

    So if your Maximo is deployed internet facing and you allow ICMP (ping) a simple tool will do part of the job, if you have a strict posture in your approach to security or your use case is more complex you probably will need more than one tool ;) and one source of data to verify your service is available.

    If you need any help just PM me ;)

    Christiaan
    Maxlogic


    ------------------------------
    Christiaan Lok
    ------------------------------