RequestAbortedException (java.nio.channels.ClosedChannelException) with Slaves in openShift

30 views

Skip to first unread message

Torsten Reinhard

unread,

Jul 15, 2021, 5:54:54 AM7/15/21

to Jenkins Users

Hi,

in our setup we have Jenkins running in openStack, launching Jenkins slaves in openShift, using the Kubernetes plugin.

Some jobs randomly fail with

Caused: hudson.remoting.RequestAbortedException 19:06:32 at hudson.remoting.Request.abort(Request.java:340) 19:06:32 at hudson.remoting.Channel.terminate(Channel.java:1038) 19:06:32 at org.jenkinsci.remoting.protocol.impl.ChannelApplicationLayer.onReadClosed(ChannelApplicationLayer.java:209) 19:06:32 at org.jenkinsci.remoting.protocol.ApplicationLayer.onRecvClosed(ApplicationLayer.java:221) 19:06:32 at org.jenkinsci.remoting.protocol.ProtocolStack$Ptr.onRecvClosed(ProtocolStack.java:816) 19:06:32 at org.jenkinsci.remoting.protocol.FilterLayer.onRecvClosed(FilterLayer.java:287) 19:06:32 at org.jenkinsci.remoting.protocol.impl.SSLEngineFilterLayer.onRecvClosed(SSLEngineFilterLayer.java:172) 19:06:32 at org.jenkinsci.remoting.protocol.ProtocolStack$Ptr.onRecvClosed(ProtocolStack.java:816) 19:06:32 at org.jenkinsci.remoting.protocol.NetworkLayer.onRecvClosed(NetworkLayer.java:154) 19:06:32 at org.jenkinsci.remoting.protocol.impl.NIONetworkLayer.ready(NIONetworkLayer.java:142) 19:06:32 at org.jenkinsci.remoting.protocol.IOHub$OnReady.run(IOHub.java:795) 19:06:32 at jenkins.util.ContextResettingExecutorService$1.run(ContextResettingExecutorService.java:28) 19:06:32 at jenkins.security.ImpersonatingExecutorService$1.run(ImpersonatingExecutorService.java:59) 19:06:32 at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) 19:06:32 at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) 19:06:32 at java.lang.Thread.run(Thread.java:748)

In openShift the JNLP log says

INFO: Failed to send back a reply to the request hudson.remoting.Request$2@36478fd9: hudson.remoting.ChannelClosedException: Channel "hudson.remoting.Channel@bd8e4c3:JNLP4-connect connection to ec-ci.mycompany.net/nnn.x.y.z:50000": channel is already closed

This is happening while cloning huge repositories, our guess is that something in the network (LoadBAlancer, ...) is killing the connection between Jenkins and the Slave.

Is there any option to setup the PingThread (or whatever) so produce traffic and to workaround this timeout from the network ? What options are available and where do they have to be applied ?

Thanx for any help, I already googled hours and found others having the same issue, but the solution isn´t clear.