With SAP Commerce application going live, many activities are performed by the implementation team. This includes go-live preparation, cutover, deployment, warranty support, and more. Alongside this, immediately after go-live new implementation scope is often introduced to accommodate use cases observed as part of customer feedback and other critical requirements.
While ensuring the continued success of the go-live steps and post-go-live tasks, it’s also important to ensure application quality. It is critical to establish process for system health monitoring and housekeeping. The importance of defined system monitoring and housekeeping process cannot be overstated enough.
To take an example from real life – it’s not enough to build a hospital with good doctors, it’s equally vital that the establishment is kept clean and hygienic from day one, with patients managing to easily find help and support.
The above example also applies to all IT applications, especially eCommerce.
System monitoring and maintenance can be a diverse topic with varying levels of complexity. But many standard practices are a must to start with. Listed below are some of the major activities that must be performed to ensure system stability.
Log Management and Monitoring
Log Management and monitoring is an important step in managing application health. Effective logging can help the application team to analyze and resolve issues proactively. Special attention must be paid to ensure quality of log itself; the application team must ensure unnecessary logs are not printed – this impacts application performance, security and hinders analysis of real issues.
Along with manual monitoring of logs, application management team must set up alerts to proactively monitor critical processes and thereby ensure speedy resolution.
SAP Commerce cloud allows creation of logging dashboard in Kibana. This is highly recommended – dashboard enables easy segregation and analysis of logs.
Database health monitoring helps check load on database and query performance. This in turn helps monitor and improve application performance. Two major areas to regularly monitor are –
- DTU Usage – this can be checked from HAC (Under Azure à Database). DTU usage gives a measure of CPU + Memory + I/O. A regular check of DTU usage ensures the database in performance is optimal and database cores are not overloaded. DTU usage monitoring help monitor and scaleup database resources whenever necessary
- Dynatrace monitoring helps deep dive into database transactions to review and optimize database latency, failure rate, and throughput. The application management team can view database performance at query level. This allows analysis of slow-running queries and proactive issue resolution. Dynatrace also helps review queries that are being executed and streamline applications by removing unnecessary database calls. Analysis of Dynatrace queries can also help define new database index to improve database and overall application performance
Item Type and Media Monitoring and Management
Item Type monitoring and maintenance is an integral part of application health management. Regular checks on data quality and ensuing cleanup activities help benchmark and uphold application quality. Few objects that require special mention are listed below –
Cron Jobs – Execution history and status of Cronjobs are to be cleaned at regular intervals. Already executed job losses relevance quite soon. Retaining this data for long serves no purpose. Very high load of historical information even makes analysis of current issues more difficult.
Application maintenance team should also ensure and disable auto-trigger of non-relevant jobs. Complete list of OOTB jobs must be reviewed to ensure jobs critical to application are only executed.
Log Files – Housekeeping of logfile is an absolute necessity, this unnecessarily consumes CPU and slows down Backoffice. High Volume of log files associated with Job makes it impossible to analyze the associated jobs. Maintaining Correct Log Level is also critical, as this would avoid creation of unnecessary log files.
Impex Media Files – These media files are created upon execution of Impex. The files are non-relevant and must be cleaned at regular intervals.
Other Media Files – Housekeeping of all other media files are equally important. This helps save CPU and allow effective analysis of issues.
Business Process – Already completed business process can also be cleaned up. This helps free up database streamline analysis of more recent business processes.
Patch Update and Management
SAP Commerce releases product patch upgrades frequently to address known product issues. These upgrades must be analyzed, and application solution updated at regular intervals. As part of patch upgrades SAP provided critical fixes related to security, performance, and major application issues. These fixes must be introduced in the application at the earliest.
Application developers have the option of implementing automatic upgrades, this is highly recommended.
As part of automatic upgraded application checks for availability of patches as part of any new build. If found, patches are then applied to the build.
The above list is by no means exhaustive, but this gives a good starting point. Application development team must review application landscape and prepare a detailed list of monitoring and housekeeping activities.
Teams must also ensure activation and execution of all housekeeping activities. This will reduce operational challenges
To know about Dynatrace, refer SAP Article: Dynatrace