Here is a question that we answered in the interrupt slack channel afterwards OTA updates panel. We share the answer in our blog to make it available to a wider audience.
As you mentioned, you think OTA The update should be developed early in the development phase. It is also obvious that you have to think about many different things. Make sure you implement all of these things at the beginning or what do you develop in this early project phase?
For the context, everyone in the committee agreed that OTA updates should be implemented in the product development life cycle as early as possible – ideally first. In fact, the discussion participants agreed that this should be the most important snack from the discussion.
Here is the core of the reason behind this council, which is described in more detail in conversation and the conversation Over-the-air update Entry:
- OTA updates have significant architectural effects. These are best addressed in advance.
- The support of OTA update is more complex than you expect. It is not only “I can send an update to the device from a distance”, but it is Fleet management skills How staged rollout Support.
- You want to get as much mileage out of your OTA update solution as possible so that you can be sure that it is reliable enough not to break devices in the field.
Back to the question: “Make sure you implement all of these things at the beginning or what do you develop in this early project phase?”
Ideally, if you shop, you will receive all essential OTA update and Fleet management skills That we discussed outside the box. You would integrate the update mechanism into your device and use the associated tools for the life of the project. In this case, the parts that should be implemented at an early stage do not have to be recorded and lined up.
If you create a custom OTA and fleet management solution, avoid the trap of “I can do a script to send a build with an OTA mechanism to a device.
It is a trap because it is indeed a reasonable approach that treats some of the main concerns. You will consider the architectural effects on the device side. You also have the option of exercising the update mechanism to ensure that the support of the device pages is reliable.
The problem is exactly the same as we discussed in the live conversation: The temptation is to cope with the simple part (OTA updates on the device) and the complex parts (fleet management, controlled version rollbacks, staged rollouts) to “Later” In the project. If your “later” planning represents the planning when, how and who these functions are implemented, this is okay and the reality of how this suite is implemented by functions. But in our experience, “later” the end of the project is almost always before they hand over to customers.
We cannot emphasize this enough: If you build a custom OTA and fleet management solution, you must give it dedicated technical time during the entire project life cycle. The fleet management functions described by us are essentialSince you significantly reduce the risk of catastrophic failure by sending a bad update. Do not leave this stuff for “later”.