You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe
This is somewhat related to #644 - the SPI defines transient errors vs non transient errors such that a non-transient error cannot be recovered. It seems like most or all connection failures might benefit from a retry. In order to express this logic today, we need to look into the sqlState of the exception and check its error code prefix.
In the current state, we need to need to propagate our own definitions of SQL error codes across multiple applications using this driver in order to express the retry logic we'd like to have.
Describe the solution you'd like
Code like this already exists inside this driver. But it returns a generic non-transient resource exception. Based on the definition in the SPI, I'm not sure that all connection failures should be considered non-transient, but even so,
Would it be possible to either define (and expose) an error specific to connection failures and or introduce some utility method like what is done in the JDBC driver.
Describe alternatives you've considered
The alternative is to continue copying this code across our projects or writing our own small utility library for it.
Introducing a new exception type might be considered a breaking change, but I think introducing the utility method like in the JDBC driver could be explained through JavaDoc and the readme for this project might benefit from an example that talks specifically about retry logic.
The text was updated successfully, but these errors were encountered:
Feature Request
Is your feature request related to a problem? Please describe
This is somewhat related to #644 - the SPI defines transient errors vs non transient errors such that a non-transient error cannot be recovered. It seems like most or all connection failures might benefit from a retry. In order to express this logic today, we need to look into the sqlState of the exception and check its error code prefix.
In the current state, we need to need to propagate our own definitions of SQL error codes across multiple applications using this driver in order to express the retry logic we'd like to have.
Describe the solution you'd like
Code like this already exists inside this driver. But it returns a generic non-transient resource exception. Based on the definition in the SPI, I'm not sure that all connection failures should be considered non-transient, but even so,
Would it be possible to either define (and expose) an error specific to connection failures and or introduce some utility method like what is done in the JDBC driver.
Describe alternatives you've considered
The alternative is to continue copying this code across our projects or writing our own small utility library for it.
Teachability, Documentation, Adoption, Migration Strategy
Introducing a new exception type might be considered a breaking change, but I think introducing the utility method like in the JDBC driver could be explained through JavaDoc and the readme for this project might benefit from an example that talks specifically about retry logic.
The text was updated successfully, but these errors were encountered: