Hi Sam,
We are still trying to come up with the best work-around. Currently we are
kicking off a the build/job and returning immediately. Then we have another
job that monitors the status of the first job. Definitely not an ideal
situation, but it works for now.
On Wed, Oct 18, 2017 at 12:32 AM,
Howdy, Cooper99!
I am encountering this as well, sir. If I may ask, how did you resolved
this?
You understand this but others may not: This is a tricky situation. Without
the functionality to fail the job on disconnection from the node in this
context, we end up in an endlessly suspended/wait
Hi Carlos,
Thanks for the prompt reply.
What I have seen is that when the node is deleted the slave/pod doesn't
crash, it is just deleted. Then the Jenkins master just sits there waiting
for the slave to return with the following output:
Cannot contact default-6b0e4a2d33a: java.io.IOException:
It doesn't restart the agents because as soon as the agent crashes the
build will fail. So there is no point in restarting them
On Tue, Aug 29, 2017 at 5:30 PM, Cooper99 wrote:
> I am new to Jenkins so this may be a simple question. I am using the
> kubernetes-plugin to dynamically create Jenki
I am new to Jenkins so this may be a simple question. I am using the
kubernetes-plugin to dynamically create Jenkins slaves. The one thing I
have noticed is that when using the plugin to create the slaves is if a
node gets deleted the slave pod is running on, the slave pod is not
restarted. I
