Once we get this error, then hudson stops polling perforce completely. i.e. we
have to restart to get things working again.

Are you using this in a master/slave hudson configuration?

The strange thing about this is a few commands get through successfully and then
the failure hits. Can you give a brief overview of the network setup between
Hudson and Perforce? (are you going over a WAN or is everything local?)

We are not using master/slave. We have a single hudson server with 2 build queues.

I believe all of our projects use the same server, but obviously with separate
client specs and I think some different p4 users.

Everything is "fairly" local. i.e. it may be a different subnet but it is
located in our building in London.

We have had other errors with logins failing before but this looks like a
network problem.

Just noticed this exception in the log, but I'm not sure if it is related.

Problem: Failed to communicate with p4
com.tek42.perforce.PerforceException: Failed to communicate with p4
at
com.tek42.perforce.parse.AbstractPerforceTemplate.getPerforceResponse(AbstractPerforceTemplate.java:218)
at com.tek42.perforce.parse.Workspaces.getWorkspace(Workspaces.java:26)
at hudson.plugins.perforce.PerforceSCM.pollChanges(PerforceSCM.java:395)
at hudson.model.AbstractProject.pollSCMChanges(AbstractProject.java:621)
at hudson.triggers.SCMTrigger$Runner.runPolling(SCMTrigger.java:253)
at hudson.triggers.SCMTrigger$Runner.run(SCMTrigger.java:285)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:417)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:269)
at java.util.concurrent.FutureTask.run(FutureTask.java:123)
at
java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:650)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:675)
at java.lang.Thread.run(Thread.java:595)
Caused by: java.io.IOException: Write end dead
at java.io.PipedInputStream.read(PipedInputStream.java:244)
at java.io.PipedInputStream.read(PipedInputStream.java:305)
at sun.nio.cs.StreamDecoder$CharsetSD.readBytes(StreamDecoder.java:411)
at sun.nio.cs.StreamDecoder$CharsetSD.implRead(StreamDecoder.java:453)
at sun.nio.cs.StreamDecoder.read(StreamDecoder.java:183)
at java.io.InputStreamReader.read(InputStreamReader.java:167)
at java.io.BufferedReader.fill(BufferedReader.java:136)
at java.io.BufferedReader.readLine(BufferedReader.java:299)
at java.io.BufferedReader.readLine(BufferedReader.java:362)
at
com.tek42.perforce.parse.AbstractPerforceTemplate.getPerforceResponse(AbstractPerforceTemplate.java:188)
... 11 more

Is there any issue with having multiple build threads on a single hudson server?

Is the usage of the p4 command line concurrent safe?

Wow, I've never seen an error like this:

"Caused by: java.io.IOException: Write end dead"

It is thread safe as long as a single P4Depot object is used for each thread. I
believe, but am not quite sure, that no matter how many build executors are
running, only one thread is executing a project's build.

We are still getting this a lot.

It seems to be generally caused by the usage of Piped(In|Out)putStream.
Googling the message finds a lot of JDK issues closed as won't fix. It seems
part of the design of the class.

Basically if the thread of the writing side dies, then the next read will fail.

• If a thread was providing data bytes
• to the connected piped output stream, but
• the thread is no longer alive, then an
• <code>IOException</code> is thrown.

http://java.sun.com/j2se/1.4.2/docs/api/java/io/PipedInputStream.html#read()

Piped in/out was added for master/slave scenarios. I believe we can make this
conditional and resort to the original way of doing things if you are local.
Another thought I had was to retry the command for this type of exception and
make retries configurable. Let me take a look and see.

OK, that explains why we are seeing this now.

But I am suprised that a Hudson PipedInputStream gets eventually by the p4java
api. I was assuming that they would be internal to the p4java API.

let me know if you need me to test a fix.