Hi Dulaj!

Okay, the logs give us some insight. Both setups seem to look good in terms
of TaskManager and JobManager startup.

In one of the logs (127.0.0.1) you submit a job. The job fails because the
TaskManager cannot grab the JAR file from the JobManager.
I think the problem is that the BLOB server binds to 0.0.0.0 - it should
bind to the same address as the JobManager actor system.

That should definitely be changed...

On Thu, Mar 5, 2015 at 10:08 AM, Dulaj Viduranga <vidura...@icloud.com>
wrote:

> Hi,
> This is the log with setting “localhost”
> flink-Vidura-jobmanager-localhost.log <
> https://gist.github.com/viduranga/e9d43521587697de3eb5#file-flink-vidura-jobmanager-localhost-log
> >
>
> And this is the log with setting “127.0.0.1”
> flink-Vidura-jobmanager-localhost.log <
> https://gist.github.com/viduranga/5af6b05f204e1f4b344f#file-flink-vidura-jobmanager-localhost-log
> >
>
> > On Mar 5, 2015, at 2:23 PM, Till Rohrmann <trohrm...@apache.org> wrote:
> >
> > What does the jobmanager log says? I think Stephan added some more
> logging
> > output which helps us to debug this problem.
> >
> > On Thu, Mar 5, 2015 at 9:36 AM, Dulaj Viduranga <vidura...@icloud.com>
> > wrote:
> >
> >> Using start-locat.sh.
> >> I’m using the original config yaml. I also tried changing jobmanager
> >> address in config to “127.0.0.1 but no luck. With my changes it works
> ok.
> >> The conf file follows.
> >>
> >>
> >>
> ################################################################################
> >> #  Licensed to the Apache Software Foundation (ASF) under one
> >> #  or more contributor license agreements.  See the NOTICE file
> >> #  distributed with this work for additional information
> >> #  regarding copyright ownership.  The ASF licenses this file
> >> #  to you under the Apache License, Version 2.0 (the
> >> #  "License"); you may not use this file except in compliance
> >> #  with the License.  You may obtain a copy of the License at
> >> #
> >> #      http://www.apache.org/licenses/LICENSE-2.0
> >> #
> >> #  Unless required by applicable law or agreed to in writing, software
> >> #  distributed under the License is distributed on an "AS IS" BASIS,
> >> #  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
> implied.
> >> #  See the License for the specific language governing permissions and
> >> # limitations under the License.
> >>
> >>
> ################################################################################
> >>
> >>
> >>
> >>
> #==============================================================================
> >> # Common
> >>
> >>
> #==============================================================================
> >>
> >> jobmanager.rpc.address: 127.0.0.1
> >>
> >> jobmanager.rpc.port: 6123
> >>
> >> jobmanager.heap.mb: 256
> >>
> >> taskmanager.heap.mb: 512
> >>
> >> taskmanager.numberOfTaskSlots: 1
> >>
> >> parallelization.degree.default: 1
> >>
> >>
> >>
> #==============================================================================
> >> # Web Frontend
> >>
> >>
> #==============================================================================
> >>
> >> # The port under which the web-based runtime monitor listens.
> >> # A value of -1 deactivates the web server.
> >>
> >> jobmanager.web.port: 8081
> >>
> >> # The port uder which the standalone web client
> >> # (for job upload and submit) listens.
> >>
> >> webclient.port: 8080
> >>
> >>
> >>
> #==============================================================================
> >> # Advanced
> >>
> >>
> #==============================================================================
> >>
> >> # The number of buffers for the network stack.
> >> #
> >> # taskmanager.network.numberOfBuffers: 2048
> >>
> >> # Directories for temporary files.
> >> #
> >> # Add a delimited list for multiple directories, using the system
> directory
> >> # delimiter (colon ':' on unix) or a comma, e.g.:
> >> #     /data1/tmp:/data2/tmp:/data3/tmp
> >> #
> >> # Note: Each directory entry is read from and written to by a different
> I/O
> >> # thread. You can include the same directory multiple times in order to
> >> create
> >> # multiple I/O threads against that directory. This is for example
> >> relevant for
> >> # high-throughput RAIDs.
> >> #
> >> # If not specified, the system-specific Java temporary directory
> >> (java.io.tmpdir
> >> # property) is taken.
> >> #
> >> # taskmanager.tmp.dirs: /tmp
> >>
> >> # Path to the Hadoop configuration directory.
> >> #
> >> # This configuration is used when writing into HDFS. Unless specified
> >> otherwise,
> >> # HDFS file creation will use HDFS default settings with respect to
> >> block-size,
> >> # replication factor, etc.
> >> #
> >> # You can also directly specify the paths to hdfs-default.xml and
> >> hdfs-site.xml
> >> # via keys 'fs.hdfs.hdfsdefault' and 'fs.hdfs.hdfssite'.
> >> #
> >> # fs.hdfs.hadoopconf: /path/to/hadoop/conf/
> >>
> >>
> >>> On Mar 5, 2015, at 2:03 PM, Till Rohrmann <trohrm...@apache.org>
> wrote:
> >>>
> >>> How did you start the flink cluster? Using the start-local.sh, the
> >>> start-cluster.sh or starting the job manager and task managers
> >> individually
> >>> using taskmanager.sh/jobmanager.sh. Could you maybe post the
> >>> flink-conf.yaml file, you're using?
> >>>
> >>> With your changes, everything works, right?
> >>>
> >>> On Thu, Mar 5, 2015 at 8:55 AM, Dulaj Viduranga <vidura...@icloud.com>
> >>> wrote:
> >>>
> >>>> Hi Till,
> >>>> I’m sorry. It doesn’t seem to solve the problem. The taskmanager still
> >>>> tries a 10.0.0.0/8 IP.
> >>>>
> >>>> Best regards.
> >>>>
> >>>>> On Mar 5, 2015, at 1:00 PM, Till Rohrmann <till.rohrm...@gmail.com>
> >>>> wrote:
> >>>>>
> >>>>> Hi Dulaj,
> >>>>>
> >>>>> I looked through your commit and noticed that the JobClient might not
> >> be
> >>>>> listening on the right network interface. Your commit seems to fix
> it.
> >> I
> >>>>> just want to understand the problem properly and therefore I opened a
> >>>>> branch with a small change. Could you try out whether this change
> would
> >>>>> also fix your problem? You can find the code here [1]. Would be
> awesome
> >>>> if
> >>>>> you checked it out and let it run on your cluster setting. Thanks a
> lot
> >>>>> Dulaj!
> >>>>>
> >>>>> [1]
> >>>>>
> >>>>
> >>
> https://github.com/tillrohrmann/flink/tree/fixLocalFlinkMiniClusterJobClient
> >>>>>
> >>>>> On Thu, Mar 5, 2015 at 4:21 AM, Dulaj Viduranga <
> vidura...@icloud.com>
> >>>>> wrote:
> >>>>>
> >>>>>> The every change in the commit b7da22a is not required but I thought
> >>>> they
> >>>>>> are appropriate.
> >>>>>>
> >>>>>>> On Mar 5, 2015, at 8:11 AM, Dulaj Viduranga <vidura...@icloud.com>
> >>>>>> wrote:
> >>>>>>>
> >>>>>>> Hi,
> >>>>>>> I found many other places “localhost” is hard coded. I changed them
> >> in
> >>>> a
> >>>>>> better way I think. I made a pull request. Please review. b7da22a <
> >>>>>>
> >>>>
> >>
> https://github.com/viduranga/flink/commit/b7da22a562d3da5a9be2657308c0f82e4e2f80cd
> >>>>>>>
> >>>>>>>
> >>>>>>>> On Mar 4, 2015, at 8:17 PM, Stephan Ewen <se...@apache.org>
> wrote:
> >>>>>>>>
> >>>>>>>> If I recall correctly, we only hardcode "localhost" in the local
> >> mini
> >>>>>>>> cluster - do you think it is problematic there as well?
> >>>>>>>>
> >>>>>>>> Have you found any other places?
> >>>>>>>>
> >>>>>>>> On Mon, Mar 2, 2015 at 10:26 AM, Dulaj Viduranga <
> >>>> vidura...@icloud.com>
> >>>>>>>> wrote:
> >>>>>>>>
> >>>>>>>>> In some places of the code, "localhost" is hard coded. When it is
> >>>>>> resolved
> >>>>>>>>> by the DNS, it is posible to be directed  to a different IP other
> >>>> than
> >>>>>>>>> 127.0.0.1 (like private range 10.0.0.0/8). I changed those
> places
> >> to
> >>>>>>>>> 127.0.0.1 and it works like a charm.
> >>>>>>>>> But hard coding 127.0.0.1 is not a good option because when the
> >>>>>> jobmanager
> >>>>>>>>> ip is changed, this becomes an issue again. I'm thinking of
> setting
> >>>>>>>>> jobmanager ip from the config.yaml to these places.
> >>>>>>>>> If you have a better idea on doing this with your experience,
> >> please
> >>>>>> let
> >>>>>>>>> me know.
> >>>>>>>>>
> >>>>>>>>> Best.
> >>>>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>>
> >>>>
> >>>>
> >>
> >>
>
>

Reply via email to