Jenkins build faild on slave node with `java.io.EOFException`

Question

The stacktrace is below:

    Evacuated stdout
Starting Selenium nodes on ci2
March 18, 2019 11:04:00 AM org.jenkinsci.remoting.util.AnonymousClassWarnings warn
WARN: Attempt to (de-)serialize anonymous class org.jenkinsci.plugins.gitclient.Git$1; see: https://jenkins.io/redirect/serialization-of-anonymous-classes/
March 18, 2019 11:04:03 AM org.jenkinsci.remoting.util.AnonymousClassWarnings warn
WARN: Attempt to (de-)serialize anonymous class org.jenkinsci.plugins.gitclient.RemoteGitImpl$CommandInvocationHandler$1; see: https://jenkins.io/redirect/serialization-of-anonymous-classes/
Slave JVM has not reported exit code. Is it still running?
[03/18/19 11:04:06] Launch failed - cleaning up connection
ERROR: Connection terminated
java.io.EOFException
    at java.io.ObjectInputStream$PeekInputStream.readFully(ObjectInputStream.java:2681)
    at java.io.ObjectInputStream$BlockDataInputStream.readShort(ObjectInputStream.java:3156)
    at java.io.ObjectInputStream.readStreamHeader(ObjectInputStream.java:862)
    at java.io.ObjectInputStream.<init>(ObjectInputStream.java:358)
    at hudson.remoting.ObjectInputStreamEx.<init>(ObjectInputStreamEx.java:49)
    at hudson.remoting.Command.readFrom(Command.java:140)
    at hudson.remoting.Command.readFrom(Command.java:126)
    at hudson.remoting.AbstractSynchronousByteArrayCommandTransport.read(AbstractSynchronousByteArrayCommandTransport.java:36)
    at hudson.remoting.SynchronousCommandTransport$ReaderThread.run(SynchronousCommandTransport.java:63)
Caused: java.io.IOException: Unexpected termination of the channel
    at hudson.remoting.SynchronousCommandTransport$ReaderThread.run(SynchronousCommandTransport.java:77)
[03/18/19 11:04:06] [SSH] Connection closed。

I use JDK8 and jenkins 2.164.1 on Ubuntu 16.04.6

How to fix this?

Do you have some code to show? Which execution causes this problem? — Michael Kemmerzell, Mar 18 '19 at 07:30
This occurs always on any project on my slave node, but works localy on the master node — Yu Jiaao, Mar 19 '19 at 04:08
My work aroud is [Configure Security] enable [Enable Agent → Master Access Control] option, and it works now — Yu Jiaao, Apr 24 '19 at 01:41

score 1 · Answer 1 · edited Apr 15 '19 at 14:54

1

This issue has been bugging us as well. There seems to be a temporary work around, not a solution. Just set the remote root directory of the slave node configuration on jenkins server to a new path, and remoting will take care of the rest.. but, the issue seem to reoccur now and then. We don't know the root cause yet. Any word is more than welcome.

edited Apr 15 '19 at 14:54

ざわざわ ZAWA

53
1
1
5

answered Apr 11 '19 at 11:29

kincki

11
1

Simpler to just delete the contents of the remote root directory. Jenkins will push the remoting stuff to it on connect, and assuming there's not anything in the workspace needing keeping. My guess is there's corrupted file of some sort in the workspace, and when Jenkins connects, it finds that file and barfs. – Endareth Apr 23 '19 at 00:57

Jenkins build faild on slave node with `java.io.EOFException`

1 Answers1