Andy Seaborne kirjoitti 30.10.2017 klo 16:01:
Hi Osma,
I think you've identified the "group id" gotcha.
I did a PR on this (JENA-1411 / PR #300).
I can't tell if the elasticsearch test failure is one of it's "usual
ones" (it's erratic on Jenkins).
I think the ES issue may be due to high load, as you say below. I have a
HDD on this machine and running the build makes everything painfully
slow. ES can be slow to start even in much better circumstances (on my
laptop with SSD it takes several seconds to boot up).
BaseESTest.POST is 9500, direct transport.
What should it be?
This is OK. It's different from the default ES ports (9200, 9300).
Yes - elasticsearch testing does fail intermittently sometimes. Theer
are variations on this happening occasionally on Jenkins when it is busy
(scheduling >1 build per slave machine).
Maybe simply upping the 30 second timeout would help. It can be
specified in the jena-text-es/pom.xml configuration for
elasticsearch-maven-plugin, the same place where transportPort and
httpPort are set, like this:
<timeout>60</timeout>
I can test this on my HDD machine. So far it's failed for me twice and
worked maybe four times, so with the current timeout of 30 second, it
seems to fail maybe one third of the time.
-Osma
--
Osma Suominen
D.Sc. (Tech), Information Systems Specialist
National Library of Finland
P.O. Box 26 (Kaikukatu 4)
00014 HELSINGIN YLIOPISTO
Tel. +358 50 3199529
osma.suomi...@helsinki.fi
http://www.nationallibrary.fi
