I presented this at the IBM Community Festival 2021

When ever I role off a project, I try to call out the good people working on the ground, those unsung heroes who are never noticed because they are too busy doing a good job.

When using OCP licenses provided with IBM Cloud Pak for Integration it is recommended that CP4I is deployed to assigned worker nodes.

THIS ARTICLE HAS BEEN REPLACED with https://chrisphillips-cminion.github.io/apiconnect/2022/08/02/APIConnectAnalyticsTuning-2.html

While installing IBM common services (a preq for CP4I) we hit an issue were some of the pods were terminating because of memory or CPU limits. This article will show how to increase these limits.

I have discovered today that an extra step is now required.

Apparently I recorded in API Masterclass on Getting the most of your API Strategy last year, https://developer.ibm.com/videos/get-the-most-out-of-your-api-strategy/.

I have no memory of this.

Recently I have had two customers come to me saying they are adding new API Gateway Service and they need a way to expose all of their existing APIs on these Gateways.

Recovery Time Objective and Recovery Point Objective are the most important requirements designing a HA or DR solution. High Availability (HA) will have a very low RTO and RPO, is automatic and usually achieved by having multiple systems running concurrently. Disaster Recovery is normally a longer manual process that often is triggered when a DataCenter has had a disaster, In this Article I will go through the time line that needs to be considered for an RTO. I will not go into RPO details here but I will probably do a follow up article if it is requested.

I spend a lot of time advising clients on what they need to keep their system operational. The topic of a RunBook / Standard Operating Procedure (SOP) is always interesting. In my experience only 50-60% of clients have one, 30% of clients know what should go into one but have not yet written it and 10% of clients don’t think they need it.