How is the legato application roll back triggered ? How to prevent it?


We are observing that the application developed and installed into the sierra wireless module with the developer studio is prone to rollback to its golden backup when there is any error with the current application. We are aware that it is by design it has been incorporated to prevent device being bricked due to random application error. But is there a way to avoid this roll back, is this roll back triggered externally or by the legato itself.

1.Is there possibility to initiate this rollback as a request via uart from an external controller?

2.This roll back is initiated when a certain error in the application where AT commands are processed ,when these errors are repeated for 3 times in our case, the roll back happens, is it something defined by the legato frame work ?

  1. So this roll back if controlled only by legato frame work, how is it triggered by the number of time a error is repeated (3time in our case) or any predefined timeout period before which the legato expects some acknowledgement from the application ?


" is there a way to avoid this roll back?" >>> You can mark the system to be good, or build your applications to the legato.cwe

  1. have you tried to mark to system to be bad or reboot during the probation period (30 mins)?
  1. “This roll back is initiated when a certain error in the application where AT commands are processed”, what AT command did you enter actually?

  2. you can see here:
    Updates - Legato Docs

thanks for your reponse.

  1. Could you please share documentation for building the application to legato.cwe
    and How to mark the application as good or bad ?

  2. AT+GMM , and other Network related commands whereafter three tries the application rolls back. (We are incorporating errors in these commands purposely to experiment on this behaviour)


to build application to legato.cwe, you can see here:

to mark system good or bad, you can see here: