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

Switch to Threaded View
Flume >> mail # user >> Re: Call for documentation


Copy link to this message
-
Re: Call for documentation
I am currently working on the web site. I believe that once that is in place both 1 and 2 will not be required.

Ralph

On Jul 4, 2012, at 7:40 AM, Mohammad Tariq wrote:

> Hello Mike,
>
>      Before creating the JIRA, I need your green signal. Actually, I
> just wanted to know if these things are worth mentioning in the
> official documentation as they may seem very primitive to some of the
> folks. For example, many newbies face following 4 problems quite often
> :
> 1 - Maven dependency problem.
>     FIX : Use maven3 in place of maven2.
>             We can also specify the maven version to be used in the
> pom.xml file-
>             <plugin>
>        <artifactId>maven-compiler-plugin</artifactId>
>        <configuration>
>          <executable>${JAVA_HOME}/bin/javac</executable>***
>        </configuration>
>    </plugin>
>
> 2 - Out of resource issue while trying to build from trunk.
>     FIX : Increase the memory before building the project -
>     xport MAVEN_OPTS='-Xmx512m -XX:MaxPermSize=1024m'
>
> 3 - "hostname" directory is not getting created even after using the
> %{host} sequence parameter.
>     FIX : Using the  HDFS interceptor that adds a host header key to
> each event, along with the corresponding hostname as the header value.
>     Add the following lines in the configuration file to achieve this -
>     agent.sources.tail.interceptors = hostint
>     agent.sources.tail.interceptors.hostint.type > org.apache.flume.interceptor.HostInterceptor$Builder
>     agent.sources.tail.interceptors.hostint.preserveExisting = true
>     agent.sources.tail.interceptors.hostint.useIP = false
>
> 4 - Unable to move data to the Hdfs and getting the following error -
>     ERROR hdfs.HDFSEventSink: process failed
> java.lang.NoSuchMethodError:
> org.apache.hadoop.io.SequenceFile$Writer.syncFs()V
>     FIX : Use a version of Hadoop that supports sync() operations,
> which is required for sinks that flush periodically. Apache Hadoop
> 0.20.205
>     or higher or any CDH3 version for sync() support on 0.20.x/1.x
> must be used.
>
> Regards,
>    Mohammad Tariq
>
>
> On Wed, Jul 4, 2012 at 1:14 AM, Mohammad Tariq <[EMAIL PROTECTED]> wrote:
>> Sure Mike..And thanks again for the support and valuable guidance.
>>
>> Regards,
>>    Mohammad Tariq
>>
>>
>> On Tue, Jul 3, 2012 at 10:05 PM, Mike Percy <[EMAIL PROTECTED]> wrote:
>>> No problem. If you decide to work on this, please feel free to create a JIRA
>>> and post tentative patches if you want early feedback.
>>>
>>> Also, don't hesitate to ping me or send a message the list if you have any
>>> questions.
>>>
>>> Regards,
>>> Mike
>>>
>>>
>>> On Tue, Jul 3, 2012 at 12:49 AM, Mohammad Tariq <[EMAIL PROTECTED]> wrote:
>>>>
>>>> Thanks a lot for the pointers Mike. Since it is the first time I am
>>>> trying something like this, so I needed these.
>>>>
>>>> Regards,
>>>>    Mohammad Tariq
>>>>
>>>>
>>>> On Tue, Jul 3, 2012 at 4:19 AM, Mike Percy <[EMAIL PROTECTED]> wrote:
>>>>> By the way - if you want guidance on how to submit a patch, this page
>>>>> should
>>>>> help:
>>>>>
>>>>> https://cwiki.apache.org/confluence/display/FLUME/How+to+Contribute#HowtoContribute-ProvidingPatches
>>>>>
>>>>> Regards,
>>>>> MIke
>>>>>
>>>>>
>>>>> On Mon, Jul 2, 2012 at 2:50 PM, Mike Percy <[EMAIL PROTECTED]> wrote:
>>>>>>
>>>>>> Hi Mohammad,
>>>>>> That's a great idea. Since you recently ran into such issues, please
>>>>>> feel
>>>>>> free to add that in to the User Guide and I would be happy to review
>>>>>> it.
>>>>>>
>>>>>> The RST source code for the User and Dev guides can be found here:
>>>>>>
>>>>>>
>>>>>> https://svn.apache.org/viewvc/incubator/flume/trunk/flume-ng-doc/sphinx/
>>>>>>
>>>>>> Regards,
>>>>>> Mike
>>>>>>
>>>>>>
>>>>>> On Mon, Jul 2, 2012 at 1:53 PM, Mohammad Tariq <[EMAIL PROTECTED]>
>>>>>> wrote:
>>>>>>>
>>>>>>> Hello Mike,
>>>>>>>
>>>>>>>         How about adding a "common errors and possible solutions"
>>>>>>> section?? Many of the newbies like me faced a few hiccups while trying