Great to hear.
On Wed, Feb 11, 2015 at 12:29 PM, Chesnay Schepler
wrote:
> Works now :) Thank you for your help.
>
>
> On 11.02.2015 11:39, Till Rohrmann wrote:
>
>> I found the error. Due to some refactoring, a wrong message was sent to
>> the
>> JobManager in the JobManagerInfoServlet.java. I
Works now :) Thank you for your help.
On 11.02.2015 11:39, Till Rohrmann wrote:
I found the error. Due to some refactoring, a wrong message was sent to the
JobManager in the JobManagerInfoServlet.java. I pushed a fix. Could you try
it out again?
On Wed, Feb 11, 2015 at 11:34 AM, Till Rohrmann
I found the error. Due to some refactoring, a wrong message was sent to the
JobManager in the JobManagerInfoServlet.java. I pushed a fix. Could you try
it out again?
On Wed, Feb 11, 2015 at 11:34 AM, Till Rohrmann
wrote:
> Could you check the rebasing because it seems as if the web server is now
Could you check the rebasing because it seems as if the web server is now
sending RequestArchivedJobs messages to the JobManager which should not
happen. These messages should go directly to the MemoryArchivist. The
corresponding file is JobManagerInfoServlet.java, I think.
On Wed, Feb 11, 2015 at
I just tried Till's fix, rebased to the latest master and got a whole
lot of these exceptions right away:
java.lang.Exception: The slot in which the task was scheduled has been
killed (probably loss of TaskManager).
at
org.apache.flink.runtime.instance.SimpleSlot.cancel(SimpleSlot.java:98)
Chesnay, could you try this again with Till's fix:
https://github.com/apache/flink/pull/378
The changes look good and I would like to merge it asap, but it would be nice
to double check with your problem. I will also run some tests.
– Ufuk
On 05 Feb 2015, at 10:42, Stephan Ewen wrote:
> I su
I checked and indeed the scheduleOrUpdateConsumers method can throw an
IllegalStateException without properly handling such an exception on the
JobManager level.
It is a design decision of Scala not to complain about unhandled exceptions
which are otherwise properly annotated in Java code. We shou
I suspect that this is one of the cases where an exception in an actor
causes the actor to die (here the job manager)
On Thu, Feb 5, 2015 at 10:40 AM, Till Rohrmann wrote:
> It looks to me that the TaskManager does not receive a
> ConsumerNotificationResult after having send the ScheduleOrUpdate
It looks to me that the TaskManager does not receive a
ConsumerNotificationResult after having send the ScheduleOrUpdateConsumers
message. This can either mean that something went wrong in
ExecutionGraph.scheduleOrUpdateConsumers method or the connection was
disassociated for some reasons. The logs
Hey Chesnay,
I will look into it. Can you share the complete LOGs?
– Ufuk
On 04 Feb 2015, at 14:49, Chesnay Schepler
wrote:
> Hello,
>
> I'm trying to run python jobs with the latest master on a cluster and get the
> following exception:
>
> Error: The program execution failed: JobManager
Hey!
The akka communication is not fully stable in the current snapshot master.
We are working on this. The Buffer recycled exception is probably an
artifact of the cancelling.
Stephan
On Wed, Feb 4, 2015 at 2:49 PM, Chesnay Schepler <
chesnay.schep...@fu-berlin.de> wrote:
> Hello,
>
> I'm try
Hello,
I'm trying to run python jobs with the latest master on a cluster and
get the following exception:
Error: The program execution failed: JobManager not reachable anymore.
Terminate waiting for job answer.
org.apache.flink.client.program.ProgramInvocationException: The program
execution
12 matches
Mail list logo
