Environment patch testing

 2 Replies
 0 Subscribed to this topic
 27 Subscribed to this forum
Sort:
Author
Messages
Steve Weary
New Member
Posts: 4
New Member

    How do you determine what to test and how extensive testing should be for an env patch?  Many patches have 80 or more files in the patch and it is difficult to determine what changed (or what may have broken) in the patch.  We are having an internal debate on what constitutes a 'full regression test' and how much time and effort is involved.  What is your approach to applying patches and how often do you install ESPs? 

    k-rock
    Veteran Member
    Posts: 142
    Veteran Member
      Test everything everytime! If you take this approach you never miss something you thought was unimportant, you never have this debate again, everyone knows what they are to do. I used to have a robust set of tests that covered every task in GL and AP that 2 business users could complete in 2 days.
      John Henley
      Posts: 3353
        Unfortunately, that's not a real option for a lot of organizations, with financials, HR, Payroll, plus inventory, procurement, req self service, LBI, Mobile SCM, PSA/T&E. Multiple servers, etc.

        It seems that every week Lawson has a new bundle of patches. And every response to calls for support is "install the latest patches and call is back if you still have the problem".

        So. A lot depends on what the patch is touching. For something like SSO or database, you need to do more intense testing than for a patch to fix a specific Portal issue.

        I usually recommend--when installing a large patch bundle--that you simulate everything you do for a month. So, put in some invoices, timecards, run posting and closing jobs, print reports, etc.

        For the most part, the way I view it, env patches need to be tested to make sure jobs still run to completion with "reasonable" results, whereas app patches require more attention to "accurate" results.
        Thanks for using the LawsonGuru.com forums!
        John