You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hama.apache.org by Chris Hill <cm...@umd.edu> on 2010/10/27 07:26:15 UTC
Hangs on Pi example.
First of all, amazing work, however I am having some issues trying to get
the Pi example to work. I can run the test example:
$ bin/hama jar hama-0.2.0-core-examples.jar test
10/10/27 01:20:35 DEBUG bsp.BSPJobClient: BSPJobClient.submitJobDir:
hdfs://localhost:54310/tmp/hadoop-cmhill/bsp/system/job_201010270120_0001
10/10/27 01:20:38 INFO bsp.BSPJobClient: Running job: job_201010270120_0001
10/10/27 01:20:43 INFO bsp.BSPJobClient: Job complete: job_201010270120_0001
And then when I inspect the log file, I see:
2010-10-27 01:20:39,113 INFO
org.apache.hama.examples.SerializePrinting$HelloBSP: Hello BSP from 0 of 2:
192.168.1.2:61000
2010-10-27 01:20:39,314 DEBUG org.apache.hama.bsp.BSPPeer: [
192.168.1.2:61000] enter the enterbarrier
2010-10-27 01:20:43,208 DEBUG org.apache.hama.bsp.GroomServer: Removing task
from runningTasks: task_groomd_192.168.1.2_61000_0
2010-10-27 01:20:43,208 DEBUG org.apache.hama.bsp.GroomServer: Remote peer,
host:port is groomd_192.168.1.2_61000
2010-10-27 01:20:43,209 DEBUG org.apache.hama.bsp.GroomServer: Got
heartbeatResponse from BSPMaster with responseId: 10 and 0 actions
So it seems to be doing something. I can run the 'test' many times and it
will always print Job complete, however when I run the Pi example it hangs:
$ bin/hama jar hama-0.2.0-core-examples.jar pi
10/10/27 01:23:11 DEBUG bsp.BSPJobClient: BSPJobClient.submitJobDir:
hdfs://localhost:54310/tmp/hadoop-cmhill/bsp/system/job_201010270120_0002
10/10/27 01:23:12 INFO bsp.BSPJobClient: Running job: job_201010270120_0002
Inspecting the log file, it seems to have froze at:
2010-10-27 01:23:13,659 DEBUG org.apache.hama.bsp.GroomServer: Got
heartbeatResponse from BSPMaster with responseId: 60 and 1 actions
2010-10-27 01:23:14,831 DEBUG org.apache.hama.bsp.BSPPeer: Send bytes
([B@57922f46) to 192.168.1.2
2010-10-27 01:23:14,938 DEBUG org.apache.hama.bsp.BSPPeer: [
192.168.1.2:61000] enter the enterbarrier
Any suggestions? I am running the latest Hama packing on OSX snow leopard.
I don't believe the os is the issue, since I've been running hadoop on osx
with no issue.
Chris
cmhill@umd.edu
Re: Hangs on Pi example.
Posted by "Edward J. Yoon" <ed...@apache.org>.
Oh.. it looks like a bug. We'll look at it. Thanks.
If you want to run it now, you need to set the value of
'bsp.peers.num" in hama-default.xml, as number of groomserver.
On Wed, Oct 27, 2010 at 2:26 PM, Chris Hill <cm...@umd.edu> wrote:
> First of all, amazing work, however I am having some issues trying to get
> the Pi example to work. I can run the test example:
>
> $ bin/hama jar hama-0.2.0-core-examples.jar test
> 10/10/27 01:20:35 DEBUG bsp.BSPJobClient: BSPJobClient.submitJobDir:
> hdfs://localhost:54310/tmp/hadoop-cmhill/bsp/system/job_201010270120_0001
> 10/10/27 01:20:38 INFO bsp.BSPJobClient: Running job: job_201010270120_0001
> 10/10/27 01:20:43 INFO bsp.BSPJobClient: Job complete: job_201010270120_0001
>
> And then when I inspect the log file, I see:
> 2010-10-27 01:20:39,113 INFO
> org.apache.hama.examples.SerializePrinting$HelloBSP: Hello BSP from 0 of 2:
> 192.168.1.2:61000
> 2010-10-27 01:20:39,314 DEBUG org.apache.hama.bsp.BSPPeer: [
> 192.168.1.2:61000] enter the enterbarrier
> 2010-10-27 01:20:43,208 DEBUG org.apache.hama.bsp.GroomServer: Removing task
> from runningTasks: task_groomd_192.168.1.2_61000_0
> 2010-10-27 01:20:43,208 DEBUG org.apache.hama.bsp.GroomServer: Remote peer,
> host:port is groomd_192.168.1.2_61000
> 2010-10-27 01:20:43,209 DEBUG org.apache.hama.bsp.GroomServer: Got
> heartbeatResponse from BSPMaster with responseId: 10 and 0 actions
>
> So it seems to be doing something. I can run the 'test' many times and it
> will always print Job complete, however when I run the Pi example it hangs:
>
> $ bin/hama jar hama-0.2.0-core-examples.jar pi
> 10/10/27 01:23:11 DEBUG bsp.BSPJobClient: BSPJobClient.submitJobDir:
> hdfs://localhost:54310/tmp/hadoop-cmhill/bsp/system/job_201010270120_0002
> 10/10/27 01:23:12 INFO bsp.BSPJobClient: Running job: job_201010270120_0002
>
> Inspecting the log file, it seems to have froze at:
>
> 2010-10-27 01:23:13,659 DEBUG org.apache.hama.bsp.GroomServer: Got
> heartbeatResponse from BSPMaster with responseId: 60 and 1 actions
> 2010-10-27 01:23:14,831 DEBUG org.apache.hama.bsp.BSPPeer: Send bytes
> ([B@57922f46) to 192.168.1.2
> 2010-10-27 01:23:14,938 DEBUG org.apache.hama.bsp.BSPPeer: [
> 192.168.1.2:61000] enter the enterbarrier
>
>
> Any suggestions? I am running the latest Hama packing on OSX snow leopard.
> I don't believe the os is the issue, since I've been running hadoop on osx
> with no issue.
>
> Chris
> cmhill@umd.edu
>
--
Best Regards, Edward J. Yoon
edwardyoon@apache.org
http://blog.udanax.org
Re: Hangs on Pi example.
Posted by "Edward J. Yoon" <ed...@apache.org>.
Oh.. it looks like a bug. We'll look at it. Thanks.
If you want to run it now, you need to set the value of
'bsp.peers.num" in hama-default.xml, as number of groomserver.
On Wed, Oct 27, 2010 at 2:26 PM, Chris Hill <cm...@umd.edu> wrote:
> First of all, amazing work, however I am having some issues trying to get
> the Pi example to work. I can run the test example:
>
> $ bin/hama jar hama-0.2.0-core-examples.jar test
> 10/10/27 01:20:35 DEBUG bsp.BSPJobClient: BSPJobClient.submitJobDir:
> hdfs://localhost:54310/tmp/hadoop-cmhill/bsp/system/job_201010270120_0001
> 10/10/27 01:20:38 INFO bsp.BSPJobClient: Running job: job_201010270120_0001
> 10/10/27 01:20:43 INFO bsp.BSPJobClient: Job complete: job_201010270120_0001
>
> And then when I inspect the log file, I see:
> 2010-10-27 01:20:39,113 INFO
> org.apache.hama.examples.SerializePrinting$HelloBSP: Hello BSP from 0 of 2:
> 192.168.1.2:61000
> 2010-10-27 01:20:39,314 DEBUG org.apache.hama.bsp.BSPPeer: [
> 192.168.1.2:61000] enter the enterbarrier
> 2010-10-27 01:20:43,208 DEBUG org.apache.hama.bsp.GroomServer: Removing task
> from runningTasks: task_groomd_192.168.1.2_61000_0
> 2010-10-27 01:20:43,208 DEBUG org.apache.hama.bsp.GroomServer: Remote peer,
> host:port is groomd_192.168.1.2_61000
> 2010-10-27 01:20:43,209 DEBUG org.apache.hama.bsp.GroomServer: Got
> heartbeatResponse from BSPMaster with responseId: 10 and 0 actions
>
> So it seems to be doing something. I can run the 'test' many times and it
> will always print Job complete, however when I run the Pi example it hangs:
>
> $ bin/hama jar hama-0.2.0-core-examples.jar pi
> 10/10/27 01:23:11 DEBUG bsp.BSPJobClient: BSPJobClient.submitJobDir:
> hdfs://localhost:54310/tmp/hadoop-cmhill/bsp/system/job_201010270120_0002
> 10/10/27 01:23:12 INFO bsp.BSPJobClient: Running job: job_201010270120_0002
>
> Inspecting the log file, it seems to have froze at:
>
> 2010-10-27 01:23:13,659 DEBUG org.apache.hama.bsp.GroomServer: Got
> heartbeatResponse from BSPMaster with responseId: 60 and 1 actions
> 2010-10-27 01:23:14,831 DEBUG org.apache.hama.bsp.BSPPeer: Send bytes
> ([B@57922f46) to 192.168.1.2
> 2010-10-27 01:23:14,938 DEBUG org.apache.hama.bsp.BSPPeer: [
> 192.168.1.2:61000] enter the enterbarrier
>
>
> Any suggestions? I am running the latest Hama packing on OSX snow leopard.
> I don't believe the os is the issue, since I've been running hadoop on osx
> with no issue.
>
> Chris
> cmhill@umd.edu
>
--
Best Regards, Edward J. Yoon
edwardyoon@apache.org
http://blog.udanax.org