Problem starts from slave.jar 2.38+. With 2.37 - everything is fine.

This slave.jar version is too old to investigate unfortunately. I'd guess there is something being printed to STDERR/STDOUT, which corrupts the remoting protocol over SSH

Please provide...

• Version of the SSH server on the agent side
• Welcome message you receive after the connection. Maybe this welcome message interferes somehow

I`ve added 3 logs from different slave.jar`s

2.36 - OK

2.38 - problem

3.7 - problem

I see, that no-problem log doesn`t have "hudson.remoting.RemoteInvocationHandler$RPCRequest"

2 files added.

3.7 SGE and nonSGE logs. Difference from sting 241

Some more information

SEVERE: I/O error in channel channel java.io.UTFDataFormatException at java.io.ObjectInputStream$BlockDataInputStream.readUTFSpan(ObjectInputStream.java:3432) at java.io.ObjectInputStream$BlockDataInputStream.readUTFBody(ObjectInputStream.java:3357) at java.io.ObjectInputStream$BlockDataInputStream.readUTF(ObjectInputStream.java:3169) at java.io.ObjectInputStream.readString(ObjectInputStream.java:1849) at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1526) at java.io.ObjectInputStream.readObject(ObjectInputStream.java:422) at java.util.HashMap.readObject(HashMap.java:1402) at sun.reflect.GeneratedMethodAccessor10.invoke(Unknown Source) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at java.io.ObjectStreamClass.invokeReadObject(ObjectStreamClass.java:1058) at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:2122) at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:2013) at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1535) at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java:2231) at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:2155) at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:2013) at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1535) at java.io.ObjectInputStream.readObject(ObjectInputStream.java:422) at hudson.remoting.Command.readFrom(Command.java:97) at hudson.remoting.AbstractSynchronousByteArrayCommandTransport.read(AbstractSynchronousByteArrayCommandTransport.java:34) at hudson.remoting.SynchronousCommandTransport$ReaderThread.run(SynchronousCommandTransport.java:59)

Please run the connection with FINER logging for ChannelBuilder, let's see what happens with the channel mode.

Here is

slave_3.14.log

Now the error is more interesting.

ERROR: ERROR: Unexpected error in launching a slave. This is probably a bug in Jenkins.
ERROR: Failed to install restarter
Connection terminated
hudson.remoting.RequestAbortedException: java.io.IOException: Unexpected termination of the channel
	at hudson.remoting.Request.abort(Request.java:329)
	at hudson.remoting.Channel.terminate(Channel.java:992)
	at hudson.remoting.SynchronousCommandTransport$ReaderThread.run(SynchronousCommandTransport.java:96)
	Suppressed: hudson.remoting.Channel$CallSiteStackTrace: Remote call to de02-sgeslot-linux26
		at hudson.remoting.Channel.attachCallSiteStackTrace(Channel.java:1693)
		at hudson.remoting.Request.call(Request.java:192)
java.io.EOFException
	at java.io.ObjectInputStream$PeekInputStream.readFully(ObjectInputStream.java:2624)
	at java.io.ObjectInputStream$BlockDataInputStream.readShort(ObjectInputStream.java:3099)
	at java.io.ObjectInputStream.readStreamHeader(ObjectInputStream.java:853)
	at java.io.ObjectInputStream.<init>(ObjectInputStream.java:349)
	at hudson.remoting.ObjectInputStreamEx.<init>(ObjectInputStreamEx.java:48)
	at hudson.remoting.AbstractSynchronousByteArrayCommandTransport.read(AbstractSynchronousByteArrayCommandTransport.java:35)
	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)
		at hudson.remoting.Channel.call(Channel.java:907)
		at org.jenkinsci.modules.slave_installer.impl.ComputerListenerImpl.onOnline(ComputerListenerImpl.java:32)
		at hudson.slaves.SlaveComputer.setChannel(SlaveComputer.java:611)
		at hudson.slaves.SlaveComputer.setChannel(SlaveComputer.java:413)
		at hudson.plugins.sshslaves.SSHLauncher.startSlave(SSHLauncher.java:1074)
		at hudson.plugins.sshslaves.SSHLauncher.access$500(SSHLauncher.java:145)
		at hudson.plugins.sshslaves.SSHLauncher$2.call(SSHLauncher.java:818)
		at hudson.plugins.sshslaves.SSHLauncher$2.call(SSHLauncher.java:793)

Master tries to propagate installers to the SSH slaves. It should not do that (I knew about this issue, created JENKINS-48574 just now). We could fix that, but it should not be a root cause on the issue.

OTOH I see that:

FINER: fetch3(org.jenkinsci.modules.slave_installer.impl.InstallerGui)
Dec 15, 2017 8:32:15 AM hudson.remoting.Channel send
FINE: Send RPCRequest(2,fetch3)
Dec 15, 2017 8:32:50 AM hudson.remoting.SynchronousCommandTransport$ReaderThread run
SEVERE: I/O error in channel channel

So the installer GUI call takes 35 seconds before it gets interrupted.... needs to be fixed

I have marked the plugin for adoption recently, see https://groups.google.com/d/msg/jenkinsci-dev/uc6NsMoCFQI/AIO4WG1UCwAJ for the context. Unassigning so that anybody else can work on it.