|Re: [cbi-dev] Something killing my build on clustered CI|
Looks like the behaviour I was seeing when the container memory limit was too low. I think the default on the new system is quite low, see https://github.com/eclipse-cdt/cdt-infra/blob/master/jenkins/pod-templates/cdt-full-pod-standard.yaml as an example of a template with higher limits set. I found I needed to experiment to find the correct limit for the build to work.
From: cbi-dev-bounces@xxxxxxxxxxx <cbi-dev-bounces@xxxxxxxxxxx>
On Behalf Of Mickael Istria
I'm trying to adopt a Docker image to build Eclipse aCute. Things seem fine on that topic, but I get the following strange behaviour in my build, after a few minutes (while timeout is configured to 1 hour)
The 2 [INFO] line are expected Maven output, but the "Killed" line is not.
The failure seems to always happen at the same place.
Here is the link to full log: https://ci.eclipse.org/acute/job/aCute/view/change-requests/job/PR-165/5/console
A very similar build without kubernetes agent work fine (it's on the master branch).
The Jenkinsfile is also mostly a clone of the Wild Web Developer one which doesn't show such issue.
Does anyone have a clue of what the issue could be?
Renesas Electronics Europe GmbH,Geschaeftsfuehrer/President : Olav Schulte, Sitz der Gesellschaft/Registered office: Duesseldorf, Arcadiastrasse 10, 40472 Duesseldorf, Germany,Handelsregister/Commercial Register: Duesseldorf, HRB 3708 USt-IDNr./Tax identification no.: DE 119353406 WEEE-Reg.-Nr./WEEE reg. no.: DE 14978647