Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Kafka >> mail # user >> runit and Kafka


Copy link to this message
-
Re: runit and Kafka
Acttully, I found out that there are 3 process that are created when
running:

/opt/kafka/bin/kafka-server-**start.sh /etc/kafka/server.properties

guyd      4883  4658  0 09:36 pts/5    00:00:00 /bin/bash ./kafka-server-start.sh ../config/server.properties
guyd      4885  4883  0 09:36 pts/5    00:00:00 /bin/bash ./kafka-run-class.sh kafka.Kafka ../config/server.properties
guyd      4887  4885  2 09:36 pts/5    00:00:00 java -Xmx512M -server -Dlog4j.configuration=file:./../config/log4j.properties -Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.authenticate=false -Dcom.sun.management.jmxremote.ssl=false -Dcom.sun.management.jmxremote.port=9999 -cp :./../project/boot/scala-2.8.0/lib/scala-compiler.jar:./../project/boot/scala-2.8.0/lib/scala-library.jar:./../core/target/scala_2.8.0/kafka-0.7.0.jar:./../core/lib/zkclient-20110412.jar:./../core/lib_managed/scala_2.8.0/compile/jopt-simple-3.2.jar:./../core/lib_managed/scala_2.8.0/compile/log4j-1.2.15.jar:./../core/lib_managed/scala_2.8.0/compile/zookeeper-3.3.3.jar kafka.Kafka ../config/server.properties
Why do we need these 3 processes?

On 07/08/2012 09:35 AM, Guy Doulberg wrote:
> Hi Jay,
> Thanks for your replay.
>
> Not exactly,
>
> I can see that the pid the runit save, is deferent from the pid that I
> acctually have running on the machine , could it be that the launcher of
> the kafka broker forks another process with a different PID?
>
> Thanks.
>
> The only way I could kill the process is by using kill -9 , I suspect
> that runit doesn't use the
> On 07/05/2012 10:19 PM, Jay Kreps wrote:
>> I think what you are saying is that you don't have the process id for the
>> broker so you don't have the right thing to kill?
>>
>> Two solutions I know:
>> 1. Change your script to store the process id somewhere, and use that for
>> the kill
>> 2. Use pkill and just kill all kafka instances (only works if there is only
>> one broker per node)
>>
>> -Jay
>>
>> On Thu, Jul 5, 2012 at 2:24 AM, Guy Doulberg <[EMAIL PROTECTED]>wrote:
>>
>>> Hi guys,
>>>
>>> In my deployment of Kafka I am using runit to control the service of the
>>> kafka broker.
>>>
>>> In the run script I have this line:
>>> #!/bin/sh
>>> exec 2>&1
>>>
>>> exec /opt/kafka/bin/kafka-server-**start.sh /etc/kafka/server.properties
>>>
>>>
>>>
>>> It works fine, if I want to start the broker, but if I want to stop or
>>> restart the process, the kafka broker doesn't get kill...
>>>
>>> So eventually the old kafka broker prevents from the new broker to start
>>> (Because the ports are already binded).
>>>
>>>
>>> Of course the running the same line of code in the shell and then ctrl-c
>>> ing works good (as you know kafka broker is not a daemon)
>>>
>>> I was wondering if someone has encountered this phenomena and solved it.
>>>
>>> Thanks,
>>>
>>> Guy Doulberg
>>>
>
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB