Hi folks,
I've pushed what I believe is a reasonable starting point for how node
tasks (i.e. tests to perform against external hosts) should be
structured, and it's available at:
https://apache-warble.readthedocs.io/en/latest/design-nodedb.html
I have not gone into any details about the actual payload design just
yet, but rather just the overall structure of a task, categories and
alerting settings. I also haven't gone into any detail about alerting
and reporting settings yet, so ideas on what sort of platforms we should
support for alerting are very much welcome!
One thing I haven't properly considered yet is chaining - that is, stuff
like "don't report on test A failing unless test B also fails" and so
on. Ideas here are also welcome :)
Let me know what you think, and if you have suggestions or fixes, you
can either just correct the source RST file(s), or suggest things here
on the mailing list. I expect we'll do a recap of this before any actual
coding on tasks begin, and come to an agreement on a v1 task design that
we can develop the software around.
With regards,
Daniel.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@warble.apache.org
For additional commands, e-mail: dev-h...@warble.apache.org
