IBM Order Management: Holiday Readiness Series 2023 Kickoff - Retrospect

Q & A 

Q: Is there any recommendation for 3rd party payment capturing systems like Adyen (PayByLink)?
A: Yea, we do provide guidelines on it. Refer to Integration with external payment gateway and the community article on Integrate Adyen with IBM Sterling Order Management.

Q: In Self-Service Tool, apart from heap and GC what are the other key performance metric that we can monitor for application server JVM? 
A: You can use default executor threads, data source DB connection wait time, and details about the number of active and available connections in the database pool for each JVM. Read more →  
For OMS SaaS, the IBM Support team monitors these key metrics and works with customers by engaging with them through a proactive Case (ticket). Refer to slide 20 of the Webcast to get more details about notification flow. 

Q: Can we reschedule maintenance or upgrades by using the Self-Service Tool? 
A: As an organization administrator, you can view and update the calendar events to reschedule all the environment upgrades of their organization within the upgrade window. Read more →  

Q: How frequently do we run database maintenance such as REORG and RUNSTATS? 
A: We need to run them on a weekly cadence, specifically after running purges. 
For OMoC, we are running RUNSTATS daily, and in place REORG weekly. 

Q: How to handle reservations and demand updates for peak? 
A: Use the HotSku feature for handling real-time inventory calls. For more information, refer to Hot SKU properties tuning →  

Q: Could you provide details on the solution or update done on the IBM end when the eligible nodes provided for sourcing is high?
A: There are various tuning features available within the product to handle such scenarios. Read more on Using IBM Order Management for complex sourcing requirements →  

Q: How to avoid YFC0009 when running item feed (manageItem)? Error: 'YFC0009 - Update failed. The record to be updated might either not be found or it was already deleted or updated by another transaction
A: When multiple threads are working on the same record, the threads end up by using same LOCKID. One thread updates the LOCKID, but other threads do not know. Another thread assumes the LOCKID to be the same as set initially. The other thread tries to update the initial LOCKID (which it reads when it started processing), then the other thread fails with the error YFC0009.  To resolve this problem, consider reducing the transaction boundary by having an optimal number of items in a single manageItem API call. 

Q: In what scenarios would IV availability is correct, but the published event isn't?
A: Both published and actual availability should match all the time.  If you are observing differences in published events, then there might be some failed events. When IBM Sterling Inventory Visibility cannot publish events to configured endpoints (IBM® Cloud Object Storage or webhooks), the failed events are made available to you for download by using REST API. Retrieving failed events →  


Reading Material

The authors shared the following links to continue learning about the topic: