Standard Operating Procedure
We often work on large & complex operations spanning across various technologies and supporting various regions like US UK Asia and many more depending upon your organization ongoing projects. It has been always hot cake in top management how to improve service delivery to the optimum level to avoid any penalties in ongoing projects in case of any failures due to any reason or due to human error or it may be due to negligence .So question comes how to streamline service delivery.
There is no such hard and fast rule which we can implement which makes the delivery of our services perfect and life very much easier. Various minds have various views to tackle service deliveries and there are many theories and standards also in practice with many overheads. It is something similar to the same problem can be solved using a different method. We can make it kind of simple approach to make service delivery efficient by keeping it simple stupid (KISS).
So, it is really important to ensure that we and our team follow standard delivery principles all the time to ensure that our clients and their business is least or not at all impacted due to simple human mistakes & improper kind of planning.
Standard Production Support Delivery Principles
According to my understanding, we need to adopt the below in our practice to minimize the possibilities of errors in service delivery.
- DO NOT perform any changes in the PRODUCTION environment during business hours. (Exception to this only is if approved by the client explicitly and have been told to implement. )
- DO NOT perform changes either to project or business as usual without a clear Implementation Plan & Recovery Plan or called Restore Plan or Backup Plan means how you can restore services if the proposed change did not go well.
- Always keep management updated about any deviation in the action plan, any issue in back-out if the change did not go well. If change is getting extended beyond the approved hour, go for extended window approval if the change exceeds the window as it is also considered as incomplete change or failed change.
- Any human error immediately inform your supervisor or manager and if describe the impact if any.
- Keep your focus on recovery /restoration of the service on top priority compared to investigate the root cause of the issue in case of any major incident.
- For all changes to servers, you must have either incident or request to do so. Please adhere to client methods without fail. Learn to say NO in absence of an incident or approved request.
- Try to develop an SOP for all frequent change requests which will help the entire support infrastructure. Make a checklist of activities and validate it during the process of implementations.
- Keep All Incidents, requests updated in a timely manner with required details in case of any dependency on why it is still open like H/W dependency or dependency due to some other team activity.
- Never sharing credentials/passwords as these represent individuality. Accountability is associated with it.
This is a matter of discussion and an endless kind of topic so it is better I will close this out but if you bring the above-mentioned items into your practice during your support hour you will definitely reduce the chances of human error and can improve your service delivery.
How to fix the failover service group in VCS
How to fix service group auto disabled in VCS
How to restart vxconfigd manually in VCS