[kepler-dev] [Bug 5378] add rety logic to SpanTodt, attempt to recover if RBNB goes down

bugzilla-daemon at ecoinformatics.org bugzilla-daemon at ecoinformatics.org
Thu Jun 14 18:30:20 PDT 2012


http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5378

--- Comment #3 from Derik Barseghian <barseghian at nceas.ucsb.edu> 2012-06-14 18:30:20 PDT ---
Should be fixed with r29937, r29939, r29950, 29951, with the caveat I've only
tested using the simulator as SPAN. Leaving open for testing against real SPAN.

SpanToDT can now perserve when DataTurbine stops and then restarts, when SPAN
stops and then restarts, or when both go down and restart during the same
interval of time, in either order. It can also be started before DT or SPAN are
started.

If the connection to DataTurbine is determined to be down, SpanToDT attempts to
repair this first. When the connection is restored, all data gathered in the
interim is inserted.

If the connection to DT is up, but the connection to SPAN is determined to be
down, all threads are restarted. When the simulator is down there's of course
no data to be lost here, were it really SPAN instead, SPAN buffers the data to
disk.

-- 
Configure bugmail: http://bugzilla.ecoinformatics.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA Contact for the bug.


More information about the Kepler-dev mailing list