Return-Path: X-Original-To: apmail-flume-user-archive@www.apache.org Delivered-To: apmail-flume-user-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id A6DC210FFD for ; Fri, 27 Sep 2013 15:06:04 +0000 (UTC) Received: (qmail 35128 invoked by uid 500); 27 Sep 2013 15:06:02 -0000 Delivered-To: apmail-flume-user-archive@flume.apache.org Received: (qmail 35091 invoked by uid 500); 27 Sep 2013 15:06:01 -0000 Mailing-List: contact user-help@flume.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@flume.apache.org Delivered-To: mailing list user@flume.apache.org Received: (qmail 35076 invoked by uid 99); 27 Sep 2013 15:06:00 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 27 Sep 2013 15:06:00 +0000 X-ASF-Spam-Status: No, hits=1.5 required=5.0 tests=HTML_MESSAGE,RCVD_IN_DNSWL_LOW,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (athena.apache.org: domain of mfuini@spryinc.com designates 209.85.128.52 as permitted sender) Received: from [209.85.128.52] (HELO mail-qe0-f52.google.com) (209.85.128.52) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 27 Sep 2013 15:05:54 +0000 Received: by mail-qe0-f52.google.com with SMTP id i11so1908957qej.11 for ; Fri, 27 Sep 2013 08:05:33 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:content-type; bh=O8uV/Gk2L1kFVgBT9sBgJPYKU4eKGvAshdnse+LIrco=; b=T2Nx1wBZJ/o/TBBLr1/4tKiaJMtjpxudSg+0f12OEeFhCbW5NcgRx7QqplL/yshgod qhvHABkyXblUDUb3DDE+gdNDPjN3CSSdtNeH1PqfQnz43uydxw3QwgcmHzWE5Jge190m 69wzbzC5h/l7N0SCzj4+rvkXYfx7DP0LH39QaQpvXsM4ofh2p24s40zfMU9jE9/tDyHW BoQ3Og5tbZwtF+IOaatX79mF8Tc9qkFM9AIxhYrgAZ9MiivcWVWkC3AU+nJrxxRJdeRq jEST52pwnrARl+RtfmAyOlrt6tD2NL2Tos3Rn7xISpa6EgMRfOf4KgDOI60qLkTiEN6c iQig== X-Gm-Message-State: ALoCoQmp0VbL+2vIgB0T3kiAHfUm/7QVhQfv3oq3Xv2a5EfUbu3rllRcuPUYRyq8Gdwu64yU/fLz MIME-Version: 1.0 X-Received: by 10.224.125.138 with SMTP id y10mr14258934qar.40.1380294332049; Fri, 27 Sep 2013 08:05:32 -0700 (PDT) Received: by 10.140.87.245 with HTTP; Fri, 27 Sep 2013 08:05:31 -0700 (PDT) X-Originating-IP: [75.145.229.78] In-Reply-To: References: Date: Fri, 27 Sep 2013 11:05:31 -0400 Message-ID: Subject: Re: Unable to put batch on required channel From: Mark Fuini To: user@flume.apache.org Content-Type: multipart/alternative; boundary=001a11c300f45908d004e75ed165 X-Virus-Checked: Checked by ClamAV on apache.org --001a11c300f45908d004e75ed165 Content-Type: text/plain; charset=ISO-8859-1 unsubscribe On Fri, Sep 27, 2013 at 10:55 AM, Cameron Wellock < cameron.wellock@nrelate.com> wrote: > Final update, in case anyone ever has a similar problem: increasing the > transactionCapacity to a low multiple of the batch size (say 5x batch size) > seems to have fixed the problem, at least for the moment. > > Cameron > > > On Thu, Sep 26, 2013 at 12:22 PM, Cameron Wellock < > cameron.wellock@nrelate.com> wrote: > >> Hi Paul, thanks for your thoughts. The sink does not complain--at >> all--and there are no relevant errors in the logs on the datanodes. I >> haven't waited to see if flume recovers after the other write stops, as I >> took the error messages at face value and restarted flume. I will try that >> today, time permitting, and I'll let you know what happens. >> >> Thanks again, >> Cameron >> >> >> On Thu, Sep 26, 2013 at 12:07 PM, Paul Chavez < >> pchavez@verticalsearchworks.com> wrote: >> >>> Is the HDFS sink reporting any issues writing to the cluster? If you >>> leave it alone or wait until the other application stops writing will flume >>> recover?**** >>> >>> ** ** >>> >>> SpoolDir is a good source if the write performance to HDFS is variable >>> as the files in the spool directory will just sit and wait until the flume >>> channel has space again. Another option may be to add another HDFS sink or >>> two pulling from the same channel, but from what you are saying this may >>> not increase performance.**** >>> >>> ** ** >>> >>> Hope that helps,**** >>> >>> Paul Chavez**** >>> >>> ** ** >>> >>> ** ** >>> >>> *From:* Cameron Wellock [mailto:cameron.wellock@nrelate.com] >>> *Sent:* Thursday, September 26, 2013 8:37 AM >>> *To:* user@flume.apache.org >>> *Subject:* Unable to put batch on required channel**** >>> >>> ** ** >>> >>> Hello world,**** >>> >>> ** ** >>> >>> I've been trying to set up a test instance of flume and have been >>> stymied by recurring failures. I'm trying to use a single flume agent >>> moving about 200G of data from a spooldir into a very small hadoop cluster >>> (3 nodes). If flume is the only thing writing to HDFS, everything works >>> fine, but as soon as another application starts writing data into the >>> cluster HDFS slows down and flume barfs with an "unable to put batch on >>> required channel" exception.**** >>> >>> ** ** >>> >>> I have tried all kinds of configuration changes, to no avail. I have >>> tried memory channels, file channels, small batch sizes (down to 50), large >>> batch sizes (up to 20000), increasing timeouts, increasing channel capacity >>> (up to 150 million), you name it. Sooner or later (usually 5-10 minutes >>> after restart) flume comes to a halt. This is especially vexing considering >>> that it's copying from a file to a file--there are no realtime requirements >>> that might reasonably lead to a full channel in other circumstances. >>> Anybody have any advice? Insights? Wild guesses? Outright lies?**** >>> >>> ** ** >>> >>> Below are two exceptions from the log, one from a memory channel >>> configuration, one from a file channel configuration, and below that is the >>> most recent configuration file used. Absolutely any suggestions would be >>> appreciated.**** >>> >>> ** ** >>> >>> Thanks,**** >>> >>> Cameron**** >>> >>> ** ** >>> >>> ** ** >>> >>> 25 Sep 2013 21:05:12,262 ERROR [pool-5-thread-1] >>> (org.apache.flume.source.SpoolDirectorySource$Spool**** >>> >>> DirectoryRunnable.run:195) - FATAL: Spool Directory source r1: { >>> spoolDir: /var/nrelate/flume-spool**** >>> >>> }: Uncaught exception in SpoolDirectorySource thread. Restart or >>> reconfigure Flume to continue proc**** >>> >>> essing.**** >>> >>> org.apache.flume.ChannelException: Unable to put batch on required >>> channel: org.apache.flume.channel**** >>> >>> .MemoryChannel{name: c1}**** >>> >>> at >>> org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:200) >>> **** >>> >>> at >>> org.apache.flume.source.SpoolDirectorySource$SpoolDirectoryRunnable.run(SpoolDirectorySou >>> **** >>> >>> rce.java:189)**** >>> >>> at >>> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)* >>> *** >>> >>> at >>> java.util.concurrent.FutureTask$Sync.innerRunAndReset(FutureTask.java:351) >>> **** >>> >>> at >>> java.util.concurrent.FutureTask.runAndReset(FutureTask.java:178)**** >>> >>> at >>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(Scheduled >>> **** >>> >>> ThreadPoolExecutor.java:165)**** >>> >>> at >>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadP >>> **** >>> >>> oolExecutor.java:267)**** >>> >>> at >>> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110) >>> **** >>> >>> at >>> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603) >>> **** >>> >>> at java.lang.Thread.run(Thread.java:679)**** >>> >>> Caused by: org.apache.flume.ChannelException: Space for commit to queue >>> couldn't be acquired Sinks a**** >>> >>> re likely not keeping up with sources, or the buffer size is too tight** >>> ** >>> >>> at >>> org.apache.flume.channel.MemoryChannel$MemoryTransaction.doCommit(MemoryChannel.java:128) >>> **** >>> >>> at >>> org.apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSemantics.java: >>> **** >>> >>> 151)**** >>> >>> at >>> org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:192) >>> **** >>> >>> ... 9 more**** >>> >>> ** ** >>> >>> ** ** >>> >>> 25 Sep 2013 22:18:37,672 ERROR [pool-5-thread-1] >>> (org.apache.flume.source.SpoolDirectorySource$Spool**** >>> >>> DirectoryRunnable.run:195) - FATAL: Spool Directory source r1: { >>> spoolDir: /var/nrelate/flume-spool**** >>> >>> }: Uncaught exception in SpoolDirectorySource thread. Restart or >>> reconfigure Flume to continue proc**** >>> >>> essing.**** >>> >>> org.apache.flume.ChannelException: Unable to put batch on required >>> channel: FileChannel c1 { dataDir**** >>> >>> s: [/var/lib/flume-ng/.flume/file-channel/data] }**** >>> >>> at >>> org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:200) >>> **** >>> >>> at >>> org.apache.flume.source.SpoolDirectorySource$SpoolDirectoryRunnable.run(SpoolDirectorySou >>> **** >>> >>> rce.java:189)**** >>> >>> at >>> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)* >>> *** >>> >>> at >>> java.util.concurrent.FutureTask$Sync.innerRunAndReset(FutureTask.java:351) >>> **** >>> >>> at >>> java.util.concurrent.FutureTask.runAndReset(FutureTask.java:178)**** >>> >>> at >>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(Scheduled >>> **** >>> >>> ThreadPoolExecutor.java:165)**** >>> >>> at >>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadP >>> **** >>> >>> oolExecutor.java:267)**** >>> >>> at >>> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110) >>> **** >>> >>> at >>> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603) >>> **** >>> >>> at java.lang.Thread.run(Thread.java:679)**** >>> >>> Caused by: org.apache.flume.ChannelException: The channel has reached >>> it's capacity. This might be t**** >>> >>> he result of a sink on the channel having too low of batch size, a >>> downstream system running slower **** >>> >>> than normal, or that the channel capacity is just too low. [channel=c1]* >>> *** >>> >>> at >>> org.apache.flume.channel.file.FileChannel$FileBackedTransaction.doPut(FileChannel.java:46 >>> **** >>> >>> 8)**** >>> >>> at >>> org.apache.flume.channel.BasicTransactionSemantics.put(BasicTransactionSemantics.java:93) >>> **** >>> >>> at >>> org.apache.flume.channel.BasicChannelSemantics.put(BasicChannelSemantics.java:80) >>> **** >>> >>> at >>> org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:189) >>> **** >>> >>> ... 9 more**** >>> >>> ** ** >>> >>> ** ** >>> >>> ** ** >>> >>> ** ** >>> >>> # define the pipeline parts >>> --------------------------------------------------------------**** >>> >>> ** ** >>> >>> agent.sources = r1**** >>> >>> agent.sinks = k1**** >>> >>> agent.channels = c1**** >>> >>> ** ** >>> >>> agent.sources.r1.channels = c1**** >>> >>> agent.sinks.k1.channel = c1**** >>> >>> ** ** >>> >>> # the main source, a spooldir >>> ------------------------------------------------------------**** >>> >>> ** ** >>> >>> agent.sources.r1.type = spooldir**** >>> >>> agent.sources.r1.spoolDir = /var/intheworld/flume-spool**** >>> >>> agent.sources.r1.batchSize = 10000**** >>> >>> agent.sources.r1.deserializer.maxLineLength = 10000**** >>> >>> agent.sources.r1.interceptors = i1 i2**** >>> >>> ** ** >>> >>> # parse out the timestamp and add to header**** >>> >>> agent.sources.r1.interceptors.i1.type = regex_extractor**** >>> >>> agent.sources.r1.interceptors.i1.regex = ^.*\\"ts\\":(\\d+).*$**** >>> >>> agent.sources.r1.interceptors.i1.serializers = s1**** >>> >>> agent.sources.r1.interceptors.i1.serializers.s1.name = timestamp**** >>> >>> ** ** >>> >>> # also set host (hostname doesn't work properly, so set explicitly)**** >>> >>> agent.sources.r1.interceptors.i2.type = static**** >>> >>> agent.sources.r1.interceptors.i2.key = host**** >>> >>> agent.sources.r1.interceptors.i2.value = Ess003726**** >>> >>> ** ** >>> >>> # the sink, HDFS >>> ------------------------------------------------------------------------- >>> **** >>> >>> ** ** >>> >>> agent.sinks.k1.type = hdfs**** >>> >>> agent.sinks.k1.hdfs.path = hdfs:// >>> a.host.in.the.world.com/events/raw/%Y-%m-%d >>> **** >>> >>> agent.sinks.k1.hdfs.filePrefix = %{host}**** >>> >>> agent.sinks.k1.hdfs.rollInterval = 0**** >>> >>> agent.sinks.k1.hdfs.rollSize = 0**** >>> >>> agent.sinks.k1.hdfs.rollCount = 0**** >>> >>> agent.sinks.k1.hdfs.batchSize = 10000**** >>> >>> agent.sinks.k1.hdfs.txnEventMax = 10000**** >>> >>> agent.sinks.k1.hdfs.idleTimeout = 900**** >>> >>> agent.sinks.k1.hdfs.callTimeout = 300000**** >>> >>> agent.sinks.k1.hdfs.fileType = DataStream**** >>> >>> agent.sinks.k1.hdfs.writeFormat = Text**** >>> >>> ** ** >>> >>> # the channel >>> ---------------------------------------------------------------------------- >>> **** >>> >>> ** ** >>> >>> agent.channels.c1.type = file**** >>> >>> agent.channels.c1.capacity = 150000000**** >>> >>> agent.channels.c1.transactionCapacity = 10000**** >>> >>> agent.channels.c1.write-timeout = 360**** >>> >> >> > --001a11c300f45908d004e75ed165 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
unsubscribe


On Fri, Sep 27, 2013 at 10:55 AM, Cameron Wellock <cameron.wellock@nrelate.com> wrote:
Final update, in case anyon= e ever has a similar problem: increasing the transactionCapacity to a low m= ultiple of the batch size (say 5x batch size) seems to have fixed the probl= em, at least for the moment.=A0

Cameron


On Thu, Sep 26, 2013 at 12:22 PM, Cameron Wellock <cameron.wellock@nrelate.com> wrote:
Hi Paul, thanks for your th= oughts. The sink does not complain--at all--and there are no relevant error= s in the logs on the datanodes. I haven't waited to see if flume recove= rs after the other write stops, as I took the error messages at face value = and restarted flume. I will try that today, time permitting, and I'll l= et you know what happens.

Thanks again,
Cameron


On Thu, Sep 26, 2013= at 12:07 PM, Paul Chavez <pchavez@verticalsearchworks.com> wrote:

Is the HDFS s= ink reporting any issues writing to the cluster? If you leave it alone or w= ait until the other application stops writing will flume recover?=

=A0<= /p>

SpoolDir is a good sou= rce if the write performance to HDFS is variable as the files in the spool = directory will just sit and wait until the flume channel has space again. A= nother option may be to add another HDFS sink or two pulling from the same = channel, but from what you are saying this may not increase performance.=

=A0<= /p>

Hope that helps,

Paul Chavez=

=A0<= /u>

=A0<= /p>

From: Cam= eron Wellock [mailto:cameron.wellock@nrelate.com]
Sent: Thursday, September 26, 2013 8:37 AM
To: user@flume.apache.orgSubject: Unable to put batch on required channel

=A0

Hello world,

=A0

I've been tryin= g to set up a test instance of flume and have been stymied by recurring fai= lures. I'm trying to use a single flume agent moving about 200G of data= from a spooldir into a very small hadoop cluster (3 nodes). If flume is th= e only thing writing to HDFS, everything works fine, but as soon as another= application starts writing data into the cluster HDFS slows down and flume= barfs with an "unable to put batch on required channel" exceptio= n.

=A0

I have tried all kinds of configuration changes, to no avail= . I have tried memory channels, file channels, small batch sizes (down to 5= 0), large batch sizes (up to 20000), increasing timeouts, increasing channe= l capacity (up to 150 million), you name it. Sooner or later (usually 5-10 = minutes after restart) flume comes to a halt. This is especially vexing con= sidering that it's copying from a file to a file--there are no realtime= requirements that might reasonably lead to a full channel in other circums= tances. Anybody have any advice? Insights? Wild guesses? Outright lies?<= /u>

=A0

Below are two exceptions from the log, one from a memory cha= nnel configuration, one from a file channel configuration, and below that i= s the most recent configuration file used. Absolutely any suggestions would= be appreciated.

=A0

Thanks,

C= ameron

=A0=

=A0

25 Sep 2013 21:05:12,262 ERROR [pool-5-thread-1] (org.apache= .flume.source.SpoolDirectorySource$Spool

DirectoryRunnable.run:195) =A0- FATAL: Spool Directory source r1: { spoolDi= r: /var/nrelate/flume-spool

=A0}: Uncaught exception in SpoolDirectorySource thread. Restart or rec= onfigure Flume to continue proc

essing.

org.apache.flume.ChannelException: Unable to put batch on= required channel: org.apache.flume.channel

.MemoryChannel{name: c1}

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.channel.ChannelProce= ssor.processEventBatch(ChannelProcessor.java:200)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.source.SpoolDirectory= Source$SpoolDirectoryRunnable.run(SpoolDirectorySou

=

rce.java:189)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java.util.concurrent= .Executors$RunnableAdapter.call(Executors.java:471)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java= .util.concurrent.FutureTask$Sync.innerRunAndReset(FutureTask.java:351)

=A0=A0=A0=A0=A0=A0=A0=A0=A0= =A0=A0 at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:178)<= u>

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java= .util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201= (Scheduled

ThreadPoolExe= cutor.java:165)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java= .util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(Schedu= ledThreadP

oolExecutor.j= ava:267)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java= .util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110)=

=A0=A0=A0=A0=A0=A0=A0= =A0=A0=A0=A0 at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPo= olExecutor.java:603)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java= .lang.Thread.run(Thread.java:679)

Caused by: org.apache.flume.ChannelException: Space for commit to= queue couldn't be acquired Sinks a

re likely not keeping up with sources, or= the buffer size is too tight

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.channel.MemoryC= hannel$MemoryTransaction.doCommit(MemoryChannel.java:128)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.= apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSeman= tics.java:

151)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.channel.ChannelProces= sor.processEventBatch(ChannelProcessor.java:192)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 ... 9 more

=A0

=A0

25 Sep 2013 22:18:37,672 ERROR [pool-5-t= hread-1] (org.apache.flume.source.SpoolDirectorySource$Spool<= /p>

DirectoryRunnable.run:195) =A0- FATAL: Spool Di= rectory source r1: { spoolDir: /var/nrelate/flume-spool

=A0}: Uncaught exception in SpoolDirectoryS= ource thread. Restart or reconfigure Flume to continue proc

essing.

org.apache.flume.ChannelException: Unable to put batch on= required channel: FileChannel c1 { dataDir

s: [/var/lib/flume-ng/.flume/file-channel/data] }

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.f= lume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:200)<= u>

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache= .flume.source.SpoolDirectorySource$SpoolDirectoryRunnable.run(SpoolDirector= ySou

rce.java:189)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java.util.concu= rrent.Executors$RunnableAdapter.call(Executors.java:471)

<= /div>

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java.= util.concurrent.FutureTask$Sync.innerRunAndReset(FutureTask.java:351)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java= .util.concurrent.FutureTask.runAndReset(FutureTask.java:178)<= /p>

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at j= ava.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$= 201(Scheduled

ThreadPoolExecutor.java:165)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at= java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(S= cheduledThreadP

oolExecutor.java:267)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java.u= til.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at java.util.= concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603)

=A0=A0=A0=A0=A0=A0=A0=A0=A0= =A0=A0 at java.lang.Thread.run(Thread.java:679)

Caused by: org.apache.flume.ChannelExcept= ion: The channel has reached it's capacity. This might be t

he result of a sink on the channel = having too low of batch size, a downstream system running slower=A0<= u>

than normal, or that the channel capacity= is just too low. [channel=3Dc1]

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.channel.file= .FileChannel$FileBackedTransaction.doPut(FileChannel.java:46<= /p>

8)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.channe= l.BasicTransactionSemantics.put(BasicTransactionSemantics.java:93)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.channel.BasicChannelS= emantics.put(BasicChannelSemantics.java:80)

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 at org.apache.flume.c= hannel.ChannelProcessor.processEventBatch(ChannelProcessor.java:189)=

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 ... 9 m= ore

=A0

=A0

=A0

=A0

# define the pipeline parts --------------------------------= ------------------------------

=A0

agent.sources =3D r1=

agent.sinks =3D k1

agent.channels =3D c1=

=A0

agent.sources.r1.channels =3D c1

agent.sinks.k1.channel =3D c1

=A0

# the main source, a= spooldir ------------------------------------------------------------

=A0

<= div>

agent.sources.r1.type =3D spooldir

=

agent.sources.r1.spoolDir =3D /var/inthew= orld/flume-spool

agent.s= ources.r1.batchSize =3D 10000

agent.sources.r1.deserializer.maxLineLeng= th =3D 10000

agent.sourc= es.r1.interceptors =3D i1 i2

=A0

# parse out the timestamp and add to head= er

agent.sources.r1.inte= rceptors.i1.type =3D regex_extractor

agent.sources.r1.interceptors.i1.regex =3D ^.*\\"ts\\":(\\d+).= *$

agent.sources.r1.= interceptors.i1.serializers =3D s1

=A0=

# also set host (hostname doesn't wor= k properly, so set explicitly)

agent.sources.r1.interceptors.i2.type =3D static

agent.sources.r1.interceptors.i2.key =3D host

agent.sources.r1.intercep= tors.i2.value =3D Ess003726

=A0

# the sink, HDFS ------------------------= -------------------------------------------------

=A0

agent.sinks.k1.type =3D hdfs

agent.sinks.k1.hdfs.path =3D hdfs://a.host.in.the.world.co= m/events/raw/%Y-%m-%d

agent.sinks.k1.hdfs.filePrefix =3D %{host= }

agent.sinks.k1.hdfs.ro= llInterval =3D 0

agent.s= inks.k1.hdfs.rollSize =3D 0

agent.sinks.k1.hdfs.rollCount =3D 0

agent.sinks.k1.hdfs.batchSize= =3D 10000

agent.sinks.k= 1.hdfs.txnEventMax =3D 10000

agent.sinks.k1.hdfs.idleTimeout =3D 900

agent.sinks.k1.hdfs.callT= imeout =3D 300000

agent.= sinks.k1.hdfs.fileType =3D DataStream

agent.sinks.k1.hdfs.writeFormat =3D Text<= u>

=A0

# the channel ------------------------------= ----------------------------------------------

=A0

agent.channels.c1.type =3D file

=

agent.channels.c1.capacity =3D 150000000

agent.channels.c1.transactionCapacity =3D 10000=

agent.channels.c1.write= -timeout =3D 360




--001a11c300f45908d004e75ed165--