Log messages sent when Golioth isn’t connected generate errors and are lost. Is there a way they can they be queued up and sent later when Golioth is connected, or is there another workaround?
If so, how do I enable it?
Hey @john.stuewe, I’m unaware of anything that would work out of the box. There are log backends (CONFIG_LOG_BACKEND_FS) that save everything to persistent memory (flash) in a circular buffer fashion. The application (or SDK) would need to read those circular buffers (just files on the filesystem) and send those logs to the cloud. The filesystem log backend works in a similar fashion to the Golioth log backend, but logs go into a filesystem on flash instead of being sent to the cloud. Afterward, we would need to read those files and send them to the Golioth cloud. This is something to be implemented. Overall it makes sense for us to provide such functionality (aggregating logs and sending them periodically or on-demand) in the SDK.
Understood. Thanks Marko.