Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hadoop, mail # user - Task process exit with nonzero status of 134


Copy link to this message
-
Re: Task process exit with nonzero status of 134
alo alt 2011-12-27, 14:07
Anthony,

personally I haven't tested yet, jdk7 have bugs already. Was only a
hint to see if the error occurs.
I would focus on memory issues, the installed RAM are okay? No errors?
My next step would be to downgrade into one JDK earlier to check for a
bug. Did you update the OS before?

- Alex

On Tue, Dec 27, 2011 at 2:54 PM, anthony garnier <[EMAIL PROTECTED]> wrote:
>
> Alex
>
> -XX:+UseCompressedOops option is the default in 1.6.0_24 and above on 64 bit JVMs (http://wiki.apache.org/hadoop/HadoopJavaVersions)
> Anyway, I tested it but same result.
> Is it wise to test hadoop with the new jdk7_2 ?
>
> Anthony
>
>
>> Date: Tue, 27 Dec 2011 13:47:03 +0100
>> Subject: Re: Task process exit with nonzero status of 134
>> From: [EMAIL PROTECTED]
>
>> To: [EMAIL PROTECTED]
>>
>> Anthony,
>>
>> 134 depends mostly on JRE (Bug) or defect RAM. _30 is the newest
>> update, could be a bug inside. Can you test SE 7u2?
>> Todd mentioned in a older post to use -XX:+UseCompressedOops
>> (hadoop-env.sh). Another option could be to take a closer look at
>> garbage collection with compressed option.
>>
>> - Alex
>>
>> On Tue, Dec 27, 2011 at 1:20 PM, anthony garnier <[EMAIL PROTECTED]> wrote:
>> > Alex,
>> >
>> > Memory available on namenode / Jobtracker :
>> > Tasks: 435 total,   1 running, 434 sleeping,   0 stopped,   0 zombie
>> > Cpu(s):  0.0%us,  0.0%sy,  0.0%ni,100.0%id,  0.0%wa,  0.0%hi,  0.0%si,
>> > 0.0%st
>> > Mem:     15360M total,    11609M used,     3750M free,      311M buffers
>> > Swap:     2047M total,        1M used,     2046M free,     8833M cached
>> >
>> > On datanode / Tasktracker :
>> > top - 13:15:27 up 6 days, 21:11,  1 user,  load average: 0.03, 0.28, 0.26
>> > Tasks: 377 total,   1 running, 376 sleeping,   0 stopped,   0 zombie
>> > Cpu(s):  0.0%us,  0.0%sy,  0.0%ni, 99.9%id,  0.0%wa,  0.0%hi,  0.0%si,
>> > 0.0%st
>> > Mem:     72373M total,     4321M used,    68051M free,      348M buffers
>> > Swap:     2047M total,        0M used,     2047M free,     2771M cached
>> >
>> > src/mapred/mapred-default.xml :
>> > <property>
>> >   <name>mapred.child.java.opts</name>
>> >   <value>-Xmx200m</value>
>> > </property>
>> >
>> >
>> > So there should be enough memory
>> >
>> > Anthony
>> >
>> >
>> >> Date: Tue, 27 Dec 2011 11:58:46 +0100
>> >
>> >> Subject: Re: Task process exit with nonzero status of 134
>> >> From: [EMAIL PROTECTED]
>> >> To: [EMAIL PROTECTED]
>> >
>> >>
>> >> Anthony,
>> >>
>> >> How much memory you have available? Did the system going into swap?
>> >>
>> >> - Check mapred.map.child.java.opts (mapred.xml) for given MaxSize (xmx).
>> >> - what says "top -Hc"?
>> >>
>> >> - Alex
>> >>
>> >> On Tue, Dec 27, 2011 at 11:49 AM, anthony garnier <[EMAIL PROTECTED]>
>> >> wrote:
>> >> > Hi,
>> >> >
>> >> > I got Nothing in the dmesg
>> >> > I've checked the Tasktracker and this is what I got :
>> >> >
>> >> > /************************************************************
>> >> > STARTUP_MSG: Starting TaskTracker
>> >> > STARTUP_MSG:   host = ylal2960.inetpsa.com/10.68.217.86
>> >> > STARTUP_MSG:   args = []
>> >> > STARTUP_MSG:   version = 0.20.203.0
>> >> > STARTUP_MSG:   build >> >> >
>> >> > http://svn.apache.org/repos/asf/hadoop/common/branches/branch-0.20-security-203
>> >> > -r 1099333; compiled by 'oom' on Wed May  4 07:57:50 PDT 2011
>> >> > ************************************************************/
>> >> > 2011-12-23 15:11:02,275 INFO
>> >> > org.apache.hadoop.metrics2.impl.MetricsConfig:
>> >> > loaded properties from hadoop-metrics2.properties
>> >> > 2011-12-23 15:11:02,330 INFO
>> >> > org.apache.hadoop.metrics2.impl.MetricsSourceAdapter: MBean for source
>> >> > MetricsSystem,sub=Stats registered.
>> >> > 2011-12-23 15:11:02,331 INFO
>> >> > org.apache.hadoop.metrics2.impl.MetricsSystemImpl: Scheduled snapshot
>> >> > period
>> >> > at 10 second(s).
>> >> > 2011-12-23 15:11:02,331 INFO
>> >> > org.apache.hadoop.metrics2.impl.MetricsSystemImpl: TaskTracker metrics
>
Alexander Lorenz
http://mapredit.blogspot.com

P Think of the environment: please don't print this email unless you
really need to.