Limitations
Limitations
While CIB ins7ght provides powerful insights into your CIB seven processes, there are some limitations that users should be aware of. Understanding these constraints helps set expectations and interpret the data correctly.
Filters
-
If you deploy a process definition, delete it, and then deploy it again, you may see duplicate versions in CIB ins7ght. This happens because historical deployments are not deleted in the ins7ght layer.
-
Some filters may not fully reflect concurrent or highly dynamic process changes if they occur during ongoing data extraction.
Waiting Time Between Two Activities
- When using the feature to calculate average waiting time between two activities, if the selected activities are part of a loop, CIB ins7ght will only consider the first instance.
For example, if a task repeats 10 times in a single process instance, only the first iteration is used in the calculation.
Data Completeness
- Real-time data is subject to the polling interval of the client fetching CIB seven data.
- Very recent changes in process instances or deployments may not immediately appear until the next polling cycle.
Performance
- The performance of CIB ins7ght is tied to two main topics: database specifications and data size.
- Internal testing shows calculations work properly (without timeouts) in a a single non-autoscaling db.m7g.xlarge with 4 vCPUs and 16 GB of RAM, also used with other apps running at the same time.
- Data quantity testing results show that calculations work fine with 766200 rows in act_hi_procinst table and 11240560 in act_hi_actinst table. If performance is downgraded a solution is to execute SQL DELETE statements on the CIB ins7ght database in order to remove data from it.