Acknowledgment after positions fully processed
Before positions were acknowledged immediately before processing them. For example, if you had an error writing into the database, you will lose the position because the device already thinks that it is acknowledged. Now they are only acknowledged after, so the device will retry if there is a failure anywhere in the pipeline.
Advanced computed attributes configuration
You can check the documentation and also the pull request for more info:
Excellent work.
I think Advanced Computed Attributes is very good news and I will use it.
However, I have a question about Acknowledgment.
I am now using a type of device that spams every 2 seconds, measures the deformation of the vessel hull.
To be able to do the calculations in time I have configured the[protocol].timeout
In the new version, if for example there is a problem with some notification when parsing the template, the position will not store , will not return a response and the connection will break.
Am I reasoning correctly?
Is it possible to close the connection before Acknowledgment is returned?
Regards
I did a test with Osmand,
there's a notification using a broken template.
I ran position through the browser and it waited the whole timeout and returned Not Responding.
For applications that use HTTP I will need to evaluate the impact.
There's a configuration parameter to disable it.
Hello everyone,
I have a few questions about version 5.7 (playing around with version 6.4 preview),
where there are some new things I don't fully understand.
Thanks.