Return-Path: X-Original-To: apmail-hadoop-user-archive@minotaur.apache.org Delivered-To: apmail-hadoop-user-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id C702C101F8 for ; Tue, 23 Apr 2013 00:38:51 +0000 (UTC) Received: (qmail 19909 invoked by uid 500); 23 Apr 2013 00:38:45 -0000 Delivered-To: apmail-hadoop-user-archive@hadoop.apache.org Received: (qmail 19806 invoked by uid 500); 23 Apr 2013 00:38:45 -0000 Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@hadoop.apache.org Delivered-To: mailing list user@hadoop.apache.org Received: (qmail 19798 invoked by uid 99); 23 Apr 2013 00:38:45 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 23 Apr 2013 00:38:45 +0000 X-ASF-Spam-Status: No, hits=-0.1 required=5.0 tests=HTML_MESSAGE,RCVD_IN_DNSWL_MED,SPF_PASS,WEIRD_PORT X-Spam-Check-By: apache.org Received-SPF: pass (athena.apache.org: domain of Xia_Yang@dell.com designates 143.166.148.206 as permitted sender) Received: from [143.166.148.206] (HELO ausc60ps301.us.dell.com) (143.166.148.206) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 23 Apr 2013 00:38:36 +0000 X-LoopCount0: from 10.175.216.250 X-IronPort-AV: E=Sophos;i="4.87,530,1363150800"; d="txt'48?scan'48,208,217,48?log'48,208,217,48?out'48,208,217,48";a="121173361" From: To: Date: Mon, 22 Apr 2013 19:38:13 -0500 Subject: RE: How to configure mapreduce archive size? Thread-Topic: How to configure mapreduce archive size? Thread-Index: Ac48sbhAdvlC0nm8TQm168LdWlx0tQC0ctfw Message-ID: References: <2D2295EC-2713-47C4-90B4-E16BA6467954@hortonworks.com> In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: multipart/mixed; boundary="_004_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_" MIME-Version: 1.0 X-Virus-Checked: Checked by ClamAV on apache.org --_004_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_ Content-Type: multipart/alternative; boundary="_000_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_" --_000_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi Hemanth, Could you specify steps of how to get the debug level of tasktracker? What I did is in hbase_home/conf folder, update log4j.properities, make had= oop log to debug level and also added: log4j.logger.org.apache.hadoop.mapred.JobTracker=3DDEBUG log4j.logger.org.apache.hadoop.mapred.TaskTracker=3DDEBUG and restart hbase (which include hadoop) and run mapreduce job. I can find = hadoop debug level log, but I cannot find the log for TaskTracker. Attached= is my log4j.properties and my logs. Could you help? Thanks, Jane From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com] Sent: Thursday, April 18, 2013 8:55 PM To: user@hadoop.apache.org Subject: Re: How to configure mapreduce archive size? Well, since the DistributedCache is used by the tasktracker, you need to up= date the log4j configuration file used by the tasktracker daemon. And you n= eed to get the tasktracker log file - from the machine where you see the di= stributed cache problem. On Fri, Apr 19, 2013 at 6:27 AM, > wrote: Hi Hemanth, I tried http://machine:50030. It did not work for me. In hbase_home/conf folder, I update the log4j configuration properties and = got attached log. Do you find what is happening for the map reduce job? Thanks, Jane From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com] Sent: Wednesday, April 17, 2013 9:11 PM To: user@hadoop.apache.org Subject: Re: How to configure mapreduce archive size? The check for cache file cleanup is controlled by the property mapreduce.ta= sktracker.distributedcache.checkperiod. It defaults to 1 minute (which shou= ld be sufficient for your requirement). I am not sure why the JobTracker UI is inaccessible. If you know where JT i= s running, try hitting http://machine:50030. If that doesn't work, maybe ch= eck if ports have been changed in mapred-site.xml for a property similar to= mapred.job.tracker.http.address. There is logging in the code of the tasktracker component that can help deb= ug the distributed cache behaviour. In order to get those logs you need to = enable debug logging in the log4j configuration properties and restart the = daemons. Hopefully that will help you get some hints on what is happening. Thanks hemanth On Wed, Apr 17, 2013 at 11:49 PM, > wrote: Hi Hemanth and Bejoy KS, I have tried both mapred-site.xml and core-site.xml. They do not work. I se= t the value to 50K just for testing purpose, however the folder size alread= y goes to 900M now. As in your email, "After they are done, the property wi= ll help cleanup the files due to the limit set. " How frequently the cleanu= p task will be triggered? Regarding the job.xml, I cannot use JT web UI to find it. It seems when had= oop is packaged within Hbase, this is disabled. I am only use Hbase jobs. I= was suggested by Hbase people to get help from Hadoop mailing list. I will= contact them again. Thanks, Jane From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com] Sent: Tuesday, April 16, 2013 9:35 PM To: user@hadoop.apache.org Subject: Re: How to configure mapreduce archive size? You can limit the size by setting local.cache.size in the mapred-site.xml (= or core-site.xml if that works for you). I mistakenly mentioned mapred-defa= ult.xml in my last mail - apologies for that. However, please note that thi= s does not prevent whatever is writing into the distributed cache from crea= ting those files when they are required. After they are done, the property = will help cleanup the files due to the limit set. That's why I am more keen on finding what is using the files in the Distrib= uted cache. It may be useful if you can ask on the HBase list as well if th= e APIs you are using are creating the files you mention (assuming you are o= nly running HBase jobs on the cluster and nothing else) Thanks Hemanth On Tue, Apr 16, 2013 at 11:15 PM, > wrote: Hi Hemanth, I did not explicitly using DistributedCache in my code. I did not use any c= ommand line arguments like -libjars neither. Where can I find job.xml? I am using Hbase MapReduce API and not setting an= y job.xml. The key point is I want to limit the size of /tmp/hadoop-root/mapred/local/= archive. Could you help? Thanks. Xia From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com] Sent: Thursday, April 11, 2013 9:09 PM To: user@hadoop.apache.org Subject: Re: How to configure mapreduce archive size? TableMapReduceUtil has APIs like addDependencyJars which will use Distribut= edCache. I don't think you are explicitly using that. Are you using any com= mand line arguments like -libjars etc when you are launching the MapReduce = job ? Alternatively you can check job.xml of the launched MR job to see if = it has set properties having prefixes like mapred.cache. If nothing's set t= here, it would seem like some other process or user is adding jars to Distr= ibutedCache when using the cluster. Thanks hemanth On Thu, Apr 11, 2013 at 11:40 PM, > wrote: Hi Hemanth, Attached is some sample folders within my /tmp/hadoop-root/mapred/local/arc= hive. There are some jar and class files inside. My application uses MapReduce job to do purge Hbase old data. I am using ba= sic HBase MapReduce API to delete rows from Hbase table. I do not specify t= o use Distributed cache. Maybe HBase use it? Some code here: Scan scan =3D new Scan(); scan.setCaching(500); // 1 is the default in Scan, which will= be bad for MapReduce jobs scan.setCacheBlocks(false); // don't set to true for MR jobs scan.setTimeRange(Long.MIN_VALUE, timestamp); // set other scan attrs // the purge start time Date date=3Dnew Date(); TableMapReduceUtil.initTableMapperJob( tableName, // input table scan, // Scan instance to control CF and attribu= te selection MapperDelete.class, // mapper class null, // mapper output key null, // mapper output value job); job.setOutputFormatClass(TableOutputFormat.class); job.getConfiguration().set(TableOutputFormat.OUTPUT_TABLE, tableName= ); job.setNumReduceTasks(0); boolean b =3D job.waitForCompletion(true); From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com] Sent: Thursday, April 11, 2013 12:29 AM To: user@hadoop.apache.org Subject: Re: How to configure mapreduce archive size? Could you paste the contents of the directory ? Not sure whether that will = help, but just giving it a shot. What application are you using ? Is it custom MapReduce jobs in which you u= se Distributed cache (I guess not) ? Thanks Hemanth On Thu, Apr 11, 2013 at 3:34 AM, > wrote: Hi Arun, I stopped my application, then restarted my hbase (which include hadoop). A= fter that I start my application. After one evening, my /tmp/hadoop-root/ma= pred/local/archive goes to more than 1G. It does not work. Is this the right place to change the value? "local.cache.size" in file core-default.xml, which is in hadoop-core-1.0.3.= jar Thanks, Jane From: Arun C Murthy [mailto:acm@hortonworks.com= ] Sent: Wednesday, April 10, 2013 2:45 PM To: user@hadoop.apache.org Subject: Re: How to configure mapreduce archive size? Ensure no jobs are running (cache limit is only for non-active cache files)= , check after a little while (takes sometime for the cleaner thread to kick= in). Arun On Apr 11, 2013, at 2:29 AM, > = > wrote: Hi Hemanth, For the hadoop 1.0.3, I can only find "local.cache.size" in file core-defau= lt.xml, which is in hadoop-core-1.0.3.jar. It is not in mapred-default.xml. I updated the value in file default.xml and changed the value to 500000. Th= is is just for my testing purpose. However, the folder /tmp/hadoop-root/map= red/local/archive already goes more than 1G now. Looks like it does not do = the work. Could you advise if what I did is correct? local.cache.size 500000 Thanks, Xia From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com] Sent: Monday, April 08, 2013 9:09 PM To: user@hadoop.apache.org Subject: Re: How to configure mapreduce archive size? Hi, This directory is used as part of the 'DistributedCache' feature. (http://h= adoop.apache.org/docs/r1.0.4/mapred_tutorial.html#DistributedCache). There = is a configuration key "local.cache.size" which controls the amount of data= stored under DistributedCache. The default limit is 10GB. However, the fil= es under this cannot be deleted if they are being used. Also, some framewor= ks on Hadoop could be using DistributedCache transparently to you. So you could check what is being stored here and based on that lower the li= mit of the cache size if you feel that will help. The property needs to be = set in mapred-default.xml. Thanks Hemanth On Mon, Apr 8, 2013 at 11:09 PM, > wrote: Hi, I am using hadoop which is packaged within hbase -0.94.1. It is hadoop 1.0.= 3. There is some mapreduce job running on my server. After some time, I fou= nd that my folder /tmp/hadoop-root/mapred/local/archive has 14G size. How to configure this and limit the size? I do not want to waste my space = for archive. Thanks, Xia -- Arun C. Murthy Hortonworks Inc. http://hortonworks.com/ --_000_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hi Hemant= h,

 

Could you specify steps of how to get the = debug level of tasktracker?

 

What I did is in = hbase_home/conf folder, update log4j.properities, make hadoop log to debug = level and also added:

 

log4j.logger.org.apache= .hadoop.mapred.JobTracker=3DDEBUG

log4j.logger.org.apache.hadoop.mapred.TaskTracker=3DDEBUG

 

and restart hbase (which include hadoop) and run mapre= duce job. I can find hadoop debug level log, but I cannot find the log for = TaskTracker. Attached is my log4j.properties and my logs. Could you help?

 

Thanks,

 

Jane

 

<= p class=3DMsoNormal>From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com]=
Sent: Thursday, April 18, 2013 8:55 PM
To: user@hadoo= p.apache.org
Subject: Re: How to configure mapreduce archive size= ?

 

Well, since the DistributedCache is used by the tasktracker= , you need to update the log4j configuration file used by the tasktracker d= aemon. And you need to get the tasktracker log file - from the machine wher= e you see the distributed cache problem.

 

On Fri, Apr 19, 2013 at 6:27 AM, <Xia_Yang@dell.com> wrote:

Hi Hemanth,

 

I tried http://machine:50030. It did not work for me= .

 

<= span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F= 497D'>In hbase_home/conf folder, I update the log4j configuration propertie= s and got attached log. Do you find what is happening for the map reduce jo= b?

 

<= span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F= 497D'>Thanks,

 

Jane

 <= /o:p>

From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com]
Sent= : Wednesday, April 17, 2013 9:11 PM


To: user@hadoop.apache.org
Subject: Re: How to c= onfigure mapreduce archive size?

 

The check for cache file cleanup is contr= olled by the property mapreduce.tasktracker.distributedcache.checkperi= od. It defaults to 1 minute (which should be sufficient for your requiremen= t).

 

I am n= ot sure why the JobTracker UI is inaccessible. If you know where JT is runn= ing, try hitting http://= machine:50030. If that doesn't work, maybe check if ports have been cha= nged in mapred-site.xml for a property similar to mapred.job.tracker.h= ttp.address. 

 

There is logging in the code of the tasktracker component th= at can help debug the distributed cache behaviour. In order to get those lo= gs you need to enable debug logging in the log4j configuration properties a= nd restart the daemons. Hopefully that will help you get some hints on what= is happening.

 

<= div>

Thanks

hemanth

 

On Wed, Apr 17, 2013 at 11:= 49 PM, <Xia_Yang@= dell.com> wrote:

Hi Hemanth a= nd Bejoy KS,

 

I have tried both mapred-site.xml and core-site.xml. They do= not work. I set the value to 50K just for testing purpose, however the fol= der size already goes to 900M now. As in your email, “After they are = done, the property will help cleanup the files due to the limit set. ”= ; How frequently the cleanup task will be triggered?

=

 

Regarding the job.x= ml, I cannot use JT web UI to find it. It seems when hadoop is packaged wit= hin Hbase, this is disabled. I am only use Hbase jobs. I was suggested by H= base people to get help from Hadoop mailing list. I will contact them again= .

 

<= span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F= 497D'>Thanks,

 

Jane

 <= /o:p>

From: Hemanth Yamijala [mailto:yhemanth@thoughtworks.com]
Sent= : Tuesday, April 16, 2013 9:35 PM


To: user@hadoop.apache.org
Subject: Re: How to configure mapredu= ce archive size?

 

You can limit the size by setting local.cache.size&nbs= p;in the mapred-site.xml (or core-site.xml if that works for you). I mistak= enly mentioned mapred-default.xml in my last mail - apologies for that. How= ever, please note that this does not prevent whatever is writing into the d= istributed cache from creating those files when they are required. After th= ey are done, the property will help cleanup the files due to the limit set.=  

 

T= hat's why I am more keen on finding what is using the files in the Distribu= ted cache. It may be useful if you can ask on the HBase list as well if the= APIs you are using are creating the files you mention (assuming you are on= ly running HBase jobs on the cluster and nothing else)

<= p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:= auto'> 

Thanks

Hemanth

 

On Tue, Apr 16, 2013 at 11:15 PM, <Xia_Yang@dell.com> wrote:<= /p>

Hi Hemanth,

&= nbsp;

I did not explicitly using Distrib= utedCache in my code. I did not use any command line arguments like –= libjars neither.

 

Where can I find job.xml? I am using Hbase MapReduce API= and not setting any job.xml.

 

The key point is I want to limit the size o= f /tmp/hadoop-root/mapred/local/archive. Could you help?<= /p>

 

Thanks.

 

Xia<= /span>

 

From: H= emanth Yamijala [mailto:yhemanth@thoughtworks.com]
Sent: Thursday, April= 11, 2013 9:09 PM


To: user@hadoop.apache.= org
Subject: Re: How to configure mapreduce archive size?

 

T= ableMapReduceUtil has APIs like addDependencyJars which will use Distribute= dCache. I don't think you are explicitly using that. Are you using any comm= and line arguments like -libjars etc when you are launching the MapReduce j= ob ? Alternatively you can check job.xml of the launched MR job to see if i= t has set properties having prefixes like mapred.cache. If nothing's set th= ere, it would seem like some other process or user is adding jars to Distri= butedCache when using the cluster.

 

Thanks

hemanth

 

 

 

On Thu, Ap= r 11, 2013 at 11:40 PM, <Xia_Yang@dell.com> wrote:

<= span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F= 497D'>Hi Hemanth,

 

Attached is some sample folders within my /tmp/hadoop-r= oot/mapred/local/archive. There are some jar and class files inside.=

 

My a= pplication uses MapReduce job to do purge Hbase old data. I am using basic = HBase MapReduce API to delete rows from Hbase table. I do not specify to us= e Distributed cache. Maybe HBase use it?

 

Some code here:

 

    &= nbsp;  Scan scan =3D new S= can();

       scan.setC= aching(500);        // 1 is the default in Scan, which will be bad for MapReduce jobs=

       scan.set= CacheBlocks(false);  // don't set to true for MR jobs

       scan.setTimeRange(Long.= MIN_VALUE, timestamp);=

       // set other scan attrs

 &= nbsp;     // the purge start t= ime

       Date = date=3Dnew Date();<= /o:p>

       TableMapReduceUtil.initTa= bleMapperJob(

      = ;       tableName,    &nb= sp;   // input table<= o:p>

          =    scan,         &nb= sp;     // Scan instance = to control CF and attribute selection

&n= bsp;            MapperDel= ete.class,   &nb= sp; // mapper class=

          &nbs= p;  null,   = ;      // mapper<= /u> output key

     &= nbsp;       = null,  // mapper outpu= t value

      &= nbsp;      job);

 =

       job.setOutputFo= rmatClass(TableOutputFormat.class);

       job.getConf= iguration().set(TableOutputFormat.OUTPUT_T= ABLE, tableName);

    =    job.setNumReduceTasks(0);

 &nb= sp;    

       boolean b =3D job.waitForCompletion(true);

 

From: Hemanth Yamijala [mailto:yhemanth@thoug= htworks.com]
Sent: Thursday, April 11, 2013 12:29 AM<= o:p>


To: user@hadoop.apache.org
Subject: Re: How to configure mapreduce archive size?

 

Could you paste the content= s of the directory ? Not sure whether that will help, but just giving it a = shot.

 

What= application are you using ? Is it custom MapReduce jobs in which you use D= istributed cache (I guess not) ? 

&nbs= p;

Thanks

H= emanth

 

O= n Thu, Apr 11, 2013 at 3:34 AM, <Xia_Yang@dell.com> wrote:

Hi Arun,

 <= /o:p>

I stopped my application, then restarted my hbase (w= hich include hadoop). After that I start my application. After one evening,= my /tmp/hadoop-root/mapred/local/archive goes to more than 1G. It does not= work.

 

Is this the right place to change the value?

=

 

"local.= cache.size" in file core-default.xml, which is in hadoop-core-1.0.3.ja= r

 

Thanks,

 

Jane

 

From: Arun C Murthy [mailto:acm@hortonworks.com]
S= ent: Wednesday, April 10, 2013 2:45 PM

<= p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:= auto'>
To: user@hadoop.apache.org
Subject: Re: How to configure ma= preduce archive size?

 

Ensure no jobs are running (cache limit is on= ly for non-active cache files), check after a little while (takes sometime = for the cleaner thread to kick in).

 

Arun

 <= /o:p>

On Apr 11, 2013, at 2:29 AM, <Xia_Yang@Dell.com> <Xia_Yang@Dell.com> = wrote:

 

Hi= Hemanth,

 <= /o:p>

For the hadoop 1.0.3, I can only find &qu= ot;local.cache.size" in file core-default.xml, which is in hadoop-core= -1.0.3.jar. It is not in mapred-default.xml.

 

I up= dated the value in file default.xml and changed the value to 500000. This i= s just for my testing purpose. However, the folder /tmp/hadoop-root/mapred/= local/archive already goes more than 1G now. Looks like it does not do the = work. Could you advise if what I did is correct?

 

&n= bsp; <name>local.cache.size</name>

<= div>

  <value>500000</value>

 

Thanks,

 

Xia

 

From: Hema= nth Yamijala [mailto:yhemanth@thoughtworks.com] 
Sent: Monday, = April 08, 2013 9:09 PM
To: user@hadoop.apache.org
Subject:&nb= sp;Re: How to configure mapreduce archive size?

=

 

Hi,

<= /div>

 

T= his directory is used as part of the 'DistributedCache' feature. (http://hadoop.apache.org/docs/r1.0.4/mapred_tutorial= .html#DistributedCache). There is a configuration key "local.cache= .size" which controls the amount of data stored under DistributedCache= . The default limit is 10GB. However, the files under this cannot be delete= d if they are being used. Also, some frameworks on Hadoop could be using Di= stributedCache transparently to you.

 

So you could check = what is being stored here and based on that lower the limit of the cache si= ze if you feel that will help. The property needs to be set in mapred-defau= lt.xml.

 

Thanks

Hemanth

 

On Mon, Apr 8, 2013 at 11:09 PM, <Xia_Yang@dell.com> wrote:

<= /div>

Hi,

 <= /o:p>

I am using hadoop which is packaged within hbase= -0.94.1. It is hadoop 1.0.3. There is some mapreduce job running on my ser= ver. After some time, I found that my folder /tmp/hadoop-root/mapred/local/= archive has 14G size.

 

How to configure this and limit the size? I do not want=   to waste my space for archive.

 = ;

Thanks,

&= nbsp;

Xia

&= nbsp;

 

 

--

Arun C. Murthy

=

&= nbsp;

 <= /o:p>

 

 

<= /div>

 

=

 

= --_000_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_-- --_004_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_ Content-Type: text/plain; name="hadoop log.zip.txt" Content-Description: hadoop log.zip.txt Content-Disposition: attachment; filename="hadoop log.zip.txt"; size=16145; creation-date="Mon, 22 Apr 2013 18:08:51 GMT"; modification-date="Mon, 22 Apr 2013 18:08:51 GMT" Content-Transfer-Encoding: base64 UEsDBAoAAAAAAABYlkIAAAAAAAAAAAAAAAALAAAAaGFkb29wIGxvZy9QSwMEFAAAAAgAp1iTQpDR WrNdAQAAIQMAACQAAABoYWRvb3AgbG9nL2hiYXNlLXJvb3QtbWFzdGVyLUFGTS5sb2d1kjFvwjAU hHf/ijc6UiPZgbSFDalC6lAVqR26muQRLJzY2A6C/vrGaSgmUM+f7u6db2klLIwFPgPO5vnTPOOw evmEjPEJfHhhvWwqqIXzaEE3sFi+kUJbhI1UCE5+I/w9ula62LkHSIsEGCmFF+CwGlFAd+uTx4CV CbSNkrX0WBJXbLFsVbAzVmor/QniR1MMqnd8R+abWNVgUwZJJ6tGKAfjR1OZwITlfEZqcYSggSXU WGt7tr/kVQk8TntuAK5yXLg6TqC7CH1bt+ZDgibpus+mxEhzexrNeQZnYZPAM1m9f7x+dQmcExXC vsV2kKZnbN9hfJYxRiwKlXpZ4/1OO3MbSnVeFLs7rV5Ocr8ZGSlMC73giHRY6KYMqI+vD2W1rtuO sbroIo9boGk7XH+Q1rdCXXd/leEQC/c76L/8v1qPMb6Mds7nLJ/zPN65NiasZLsWDoEepBgmn5Af UEsDBBQAAAAIAK5Yk0J1gT4QjzYAAC4/FAAkAAAAaGFkb29wIGxvZy9oYmFzZS1yb290LW1hc3Rl ci1BRk0ub3V07d35cxtXgiDo/XXqr8iN8mzZHSSIxA26NWGXbJdV5Wsluz3Rno6OJJAgYYMABwBl qSp6//Z9mQnwEG4QpNQ931dlEQSQ7718dx4vGVdPyrWTuB3F8Wm5fho3ohffffV9dD3tD0r/ko4n /dHwxbA3Oo2+/nMySaNyqV0rxX+It93q1fXZ6+L36GI6vZqcnpxMXg9LyVXSuUhLo/H5yTi9Gk1O kknv5OIsxHByNk6G4bPJSRZVdDyO4mqjXonL28f5fHR51R+k3ejsbXRx3Z2EuMP/f7xOo79eD6JK LYpbp7XyaVyOfvrxeVQpx5UlYX/x5Z9/+kvUGQ17pefhn/759TiZhihOo1+T10mpPyq9+P7LN530 qnizk3/n40/+8N+SaRR2a76HF0l3NLoqLYZT+uf+sD/9Hx/ffzML+7RSaR4mnLi2Kpw8q0t5kd7f sDNOk2n68ZJP8iDjuLw+yLw0XqXjUOihGC6TYfeb/jAtdUffJv3hx4sf5KE2G+sDvUwm03Rc+vrb 4udlFtT8lyJZrXLlkz8crBSLH4cpg8aGDNujDCobguwM+ulwWoT8efeyf5O+O+/kIbVbu+T73WKb TEdXxbsfL/m0SGajvW/o4+vh6mDL9VXB5nXvx9Fo8PJ6OAwhZ8Hc+TXfvLFy6/9C9ff/+F5o3gLC tsO0k234bTJMzkN2/3P4qGgLi58VYdc2NIrtW9fe9V/r0roe3romaed63J++Lf00Scd/GY+ur7Lp 0fiy2DodTq7H6YsQbj8Z9P+edj9e+rVZBu4ZSX/yavb5l8PkbLA2kkp1ZTmvj+Q8nX4zOu8Psw/X hF9r7B/+8+vxODT6TTHUK0UMk+thaZz2BqF7KX0XvvA6/TadXoy6n3c66WQyGr+4vBqU+sPXo9/S 8sfFF6LiGztv//FPw9+Go9+H0avR9biTLm7/RTpIz0MYw/PdwshbwCAZnt+EVGy/bpu1zb7YfFLq JIPBx/Nf8nyrrmwKxeb3iqfYPn+Vb1xvbegeFrd+NQ350bkbxqYu5n4YSZ6FH9XK5bthbBiS7gXx Ud4s0q/zL+RhzFr4bXjVdvXAAbY2DGz3d/K21t8JI65uGHVWDrwf3Xnrb+nbmzFzxShciyv7jvBZ wm/eXD3MN7fcE8P8f5Fh/qZ655335DSKnmezzNAxRsP096h4Nxqd/Rpqy/aB5D+jy+TqKguoHzrV Z2vGl1cX6WCQ1Z/uT8P+myKMb4ttP4062SY/9i/T0fX0WbWc2ZyOZaPRaVTEGw2yYfEQYYSpz+Vl f4tsWR7U9STLnMEodL7hdTo+zfb+h3F/2OlfJYPTaDwa7R32T395UaQx+/h0bUh/H4WBK70K9eVf //ZTqKJhHzs3vUM0ukqHWTLvvDUdRf86Gv0t3yb6vT+9iMKsKb0ME5no43xvLkaTaZiCteJPVp2d uhPnPKTT6Hnet4TAXvfHo+FleH16+73Z6bJn1VKtVD2OK7Vyu1k5is6u+4Npdh6rXDkpN06ys1ZR XD6ttaK/fPvjQ2PP9qM0TC7TZze79dAg8/nDfF/iUqNU/vdK40CBDruj8bNX18Po235nPJq8DX3A 5SR6MeyUDhLBxShkxMnVb+fxya/j9CBBdgbJZFK6SqYXz06uJ+PZqc783+PilOrJWX94UiqdZLP/ 05vITwb9s5Np6H6zqdL4dP22Gz6+994Wwd1773iaTqbbJCJPcRLaz+u8gR7H28WVbzW5PK7u8PXX 41EIvl6q7rRF/6qz21ZZ3zcaTo7P0mSYjWyTsHmzVN5/885onIYwWruHESYlYcPKzpuNumm207Xd NxwMir5wEkqmsn3Z3AZw54j1OOsGdgyg2z9PsyE+z68dt00HYaud8zi7TFFM/napjPOt+6PjPbIp O84K29V33m50Hka+87yR7RxpGEEv9knsMJ3mdWmHDUN9r+6SxPPr0Gkex1npbV3Zi/nKvHGVt2/f F/3zi+NJGPbS4/Drbnl5p7JkGbJ9YrPtspRmW22d0l+Tzm+T0I7yLZPJIG8UWzeL+da/Jm/Gk/02 zea5oS3uHfWbzq7bhRp3HA5BpmFaHOrqDg3y12SSJbVTXAscH9dD66rskNP51vOYd934zdlxctXf pXHl22SHD2GjSjbv23rDMMVK396OKbtt9usk75Z33WySH8LttuF02p/sNh/Itnl73Mja1dbjRrFN Ns7uuuH4+uxtXl0G6TTNh6qtu+NfJ1dZWecxxlsPstlWs1qy85bjarl+nPUc7a23uR72806qfPz1 nz9/9eX2NSz8N70Y93vT8NkOc5YwNtV+zWcr8daFcJlOx/3ObIZU2aVDHeblns1RaqWv+sNksO2G V+PRdHR23TvO5unH2eblZNtts2YQasusFOs7luJk0AsZlG2bjaZb9y7FZnnuxpUdNx2G/vttsafF ELl1/k6myZtZWsvbt+DXaTieDMfu+Yx5y23eXA7SYSd8UN86cTeHYsfV+9kRXkxOrrKD0DCHHHbS fOJfPn713ec/vPr6+x9DPQvFX5wPzzc6yLFeSNE4Gb/d6mgvr7r5Gf6Tb/rD65DFl91G7bhRO0hK +qPS9PKq2x8/Owk/D3MgOxtPn/3zd5//j4eGOJoUJx3yXT9AYMm4c/Esz8EDBDY/d1EJI0G1clxp tkuVrEtKB43Sm1bj3x8eSXYqrMiAFaetdg+tOHVxsODyqrO6Bu8UR3GpMT/fWsyY75xpO5q/fhX6 /+H5s/sn16JJaKHhW/Nzo3ErOzca/Z5MOxehJt45ibfFqb9iV58P3wxP77zb7U+y65OfX09HP2fB vkwn6TTqT6JeMphscQrobqjfz04kTkad39LpO+cTi5lTdBJXslMIoSvN93BJBD9//vK7WUbd5uKr MPMuojqNbq6s3l6gvr1mt+TDn/IrsFkastydplFyc273zjF6NOp0sss+3ej3izSkePw225WwUa8/ 7EZ//fzzV/e/v/KM27qkz6rb7/3BIBqOptGrz199c5xcT0OM036etrO0k4QaGIW3om7aS64H0yWR Z+eG859/KkL8U/j4etDNwzxLo97oetgtRS960dvRdZSM0/yD4mR0FuVR/v5l8jbqnw/DjCPE1p+U ou+HebSj8M84ukiG3aOoXwSRvrkKEYa8ybbOMuX30fi3oyjMGEO7CDn0JvvWeJdcuq1BL9POKNSM rIzuNJofQ0L63Sxbev2QmlEvT2M0G7WyClqt1Oqfff7Vt7vV0VcLdTMNI/zZoD+5CLs3qyN5C3yn poa8uG3Hs3a5Y6t7VWx1G2PW30Tj9H9fh3dCmMP8NPeqBOy4n0vjms/yo7wK5Q1y9f7OdrLfjZ5F 5TdxNa1Um3Grmp6FPqh2FA3T81GWH1m2Ff1T+F5t7cWbJd3jz0Vfdv+KRKgSaZgZdO9chPjydUj/ UTR9e5U++240TEPypiHuZ6/eDjuzK51pqK/51GN4PRg8NA3HC3s8rzNpd8dif5km3azSjNOrwdvb XD1dkqdXSVY7T09nncgPYW9Os72ZFdbt76FTymvsad5NRxchirADp1F8VI2KiL6+fSuuVo7K0byi hbf+NBvWiiuLRz9mn02uRsNJGj6c/KN91A7bNBrVdq1Sq7Sb5Xd+K/7XatdrrXJcLdfCz7jWrBzV 6uHt9n9E+12HWpLjr0JLnQ12WY1N34TZbJaTfx+Ouml0byc+4DIJGfNumVR2LJM/9mZCktrVarVc bVZb1VatXItrjVq3XG50Gr1GtRGHn61Gr1lt1iqdai3/Xvg3vI7DVuEL4d92eKeS/9cIn9eOPqDy fpkdBadZw6/G0dnbafrx5JOs6+8m0yTqjUeXy0o+CgNVNLmtKZ9G/+tNuZz9d9O1HdXKzdpsr0ql ZWPSh1JXqovtt7q6rmTT3eNxep7lZhHDQluuVI8q1dvybNbi5ju/zcu3Vg3/VKpP3XwX9+EDLp7a YlOuPaB4NjTrpY26Wq2Hb8XV9opGnb1qHL3fgr9tx5X2Fu14MZ8W2/RtW67Wm/G9/qq27Or6h1Jj 6osNur6yxhx99W7rParc2dPWvf3Ofgt92rwU463L8H1lRWMxKxqrG0/+b7+7kCVxliO32VCNK/d+ q4bfZjlSz7Ol8kHnSXOxQ2numiebepF66CQq1bhRC+/WQr/RDZlVy3qRSjf7vdFoVMKrVtgufBBe lcMWcXgZtgivaqF7Cb1No370WHm/89ygsUWfMsuqT/PPn6XVs7jbqrWPe7V687jW7p0dt8Kc6bic xr143Yygf9Up7oR8+cPz0+in/Ah61X162Xd/Hvez46305VXny+F5f5gdi4/XbjG7k+/FMPzTSzrL Trm8k5T5yYTsELm4+S7b9nUyCEfFjez463Kyeof+q97zP8vQUemi1x+kpWwl5LhouR99PjgfhWK5 uLwN+ebDWbjlDTeb7hlue8OdneuD/SheHfCGG8c3pXe+FmjngL9+PhpcXw6/SCedcf9qOrq/ruid j4oQ4w0h/pi1l6UBvvvJbAnUvBLcnvR7nt9kFlrad8nl8oUFK7679Gb+j34Yj968rd8mZNNKgyK+ UF8n0/F1J6T03lKDYfr7i/BJMuxsu+Zhi5B2WP2wb2iL6yDuhLTH1nmuZtvlL9ZuvFMPm933ngf5 8eJHszveN9z/f7eTvw3t5p1iEUF1w63fWwWy6eb+rQJpbOgEt1qKkFWBNDsneLPUZ35D/op1A41N 975vvW4g3tTbWjjwYSwcMFf4zzNXqKwMOI4fFPKaycLGkM0WzBbMFswWzBbMFswWPqDZQnX1bKG6 YTn3/rOFTSGbLZgtmC2YLZgtmC2YLXxAs4Xa6tlCY1Ons/dsYVPIZgtmC2YLZgtmC2YLZgsf0Gyh vnq20NowAu8/W9gUstmC2YLZgtmC2cL/AbOFhTsl51WjWNC5sRIXm310p9J1P2sklUovaba2j/TF D8/n6z0/rjU/eWc57LJld/kqleJe1uyW+GxFYH4j8h9Xr6g7fKyn2cK6cZZXR9FF8vr+gwEn0bJl iI+WAefhv9fJ4DrdlAWhkyl6zlDWyWBwGhV90HTUGQ1mfyEjqj1lLt6W3fvKsPXxbpNhT5rym/xa /SdJHje/1se7kF/9SdEhZZ1llu4nTfZNZlXfU2atj3djZr2fmrXssRVPkVnr492YWe+n26q/p8xa H+9CZk0urqfdbBoex3/4vwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAACApxNXT8q1k7gdxfFpuX4a16MvvvzzT3+JJun4dToufdUfJoOX6f++TifT H8ajTjqZjMan0exlf3gejYsPT0/DJuGd0bDfPS2/iatppdqMW9X0rFwuV6Lp26v09Cr7fudN/oXe fWn09xXvT98M842vh78NR78Ps/iukunF6fAk+cN+if9w0vn30ei3NL0KSX0+6KfD6fPhm+Fp9JfR NLoqsnZyNRpO0qg3Gt9JdbSY7KQ3TcdRfDlZiKoRvfjuq+/nOfLdi+9f5a+ymL5KOtPR+O1p9Hmn k15N0240GXV+S6dRZzQcpp1piC3qjUeX0UlcaZbK4X/xab0c1+MlkdzL938djf6W71YR12n0qkh7 FMohORv0JxeXYWfnNaeIo5NnwEJUs/f/NIkGyWT6ryHvo/4kZEB5w44uJKHI4CiZTtPLq2mWu9PR bXqiYfr7PIfDd3be48M0k1pR/TrjNJmm8zyb1cPyvOK1ajdV7Tgur61m2yXyKdOztoi+nJdGVhNn sS0m6ff+NCuu89G0n2R1dtq/TEfX06gWPi3nLWVFVXrSMkzf9CfTyTyz4p17jZOLs2SSnlyGSp+O H6tgP4RE7pe95+n0i2SazJNe+TDz94NI5SHqb3XflI9Ho+nxOD0PcR0XyXyauvz+E3yQel1777vx ny/Fh6jv9T334mkq9xOn7hAZ2ti3WuT/9rtPk7HvKZUH6SqaH2gOP1Uy+1ed0td/Dt8pJvqn0Ysf ns8n/R/Xmp/cPagJM//BqJMMLkaT6Z1ZYq3crJWLo5GsNwr7NuxmhfLHSrW1Od75LDYKM9MoBD7I tzuK4katFZ29nabLjs1WhJGlvXgdXSTD7iA7totCwvMUnob3JnngRVrvTHMb7XL1YLHc7EOUvkk7 1/kxU4j4u5+++Sa6GveHnf5VMlgZ252YStNx0knDIdg8wE+LaMOR7Nff5lObl3m//2IYXvbCV/9Y jAPF1i/Tq9F4GoWKd53+GOb6z8oh9nmDmb0RynYaCnryLI6+Lrb6ZpR0j6Jq/ICcf5kffXezd4sD 8e7sqPGwWb97NNHP49E0DRWr2LnSU7aJrHK/TgbX6eZW8fKH518Oz/vDWclnu7dQqqsPyhaybJZd 756buJc39fan0R+jJHz8Or3zzclpVMk+yOtQN6/Y4a1lpxA2t+nyUdRuHKA9V95tz+Wl+3OwOGap P2xbLi+05Ns2HDr+ME5MR53R4F/ScX4wf9uG47Vt+JvR8Dx0nNGraUja+YFb1QHzeddIZg23Utm6 4S6pgPGBKmDt3QoYH74C1hYqYHzYChj/J6yAB8znXSM5SAWshHH1EDOa6sKM5vAVsLo4nznwbGZN BexPZrOb6+Ewi267KcyhJxFPUduWRzKfp9QfUtuqB6pt9XdrW/Xwta2+UNuqh61t1b1q27q+7sAV 4YCZumskB6lttQPVtsa7ta12+NrWWKhttcPWttqH3rcdMFN3jeQgta3+wCPUeU1ovlvb6oevbc2F 2lbfsbbl1/6KSxmlWX3KLsteZy+iycX1tDs77ZOdYkqXnem5G0KRvG+TYXKehfNz0s8TMhrODjOL 01YfTz7JCvF8FOWB3xzhHlXrzbh9FFcbjWq7VqlV2rXGdkn+qHJ6J5g8Q+4E0yjXjv+cDJJhJx0/ vxiN05BHecI2BP48mSaD0flfk2F/mp1WWx/D/W+vieIRzj6Wi9N63XSQhvo/v3QY35yTbtVvzuRV HvXi9METsvpmjOFo2u/1O0l+1mF1ilafzFgd9M/JtHORdr98nZ2TmUyTaXr66u2w87w4d5Fd3s52 4LtRN/0i39FudPek6E2ruXeDyMJF820SdnM1vkjR+PReOzrN28vxQsCh3+yk/TBERDfbR/m+HOUJ f3Yn4UfF7j27t3tH+e48e2d3Hq+IVp/zec9FtE3CFouo6D2Kk3QLhVP+IArnZZp0i77lavB2Xclk sWV3H4Xep7iB44e8wV6Hoaaog7e/9/rD/C6R014ymKTRRYgi5MJpVGk2jypREdXXd9+Mq9WjcnSn f/vTOzt1dBxHN7dbhc+zaHafMNXXTJhu6sGamVL8eFOlA84Ido1k69PVjzFOvXOLS7z3Jb6N1f+B 49iHktAdszlelvpk71uJHi2XnzqdD+sM4wN0hnEjPqq+2xlmb2ad4XFcjtf2hz/e6w6jbUanv/00 7Q+2nDa8Ozh2R+kkG8OjvIBKszF3nN3rOUmnq7uMD2HMaS1mc/bmI2fzhqH/oRncjH7+/OV3S28b zo7/rs8vQlDDbjTqhZnDOE0uQ8DZTcQho//w5bD7fe9V/u6X8zdPo5+GydkgzcaJ8EE3SrrdfvZB Moi62f0Bd28CXj1Nqh1Fg/5vaSjV2Vezw93OYDS5uXn5D/8tmUaj8XkpCUV7kZZuc27ZvpS6oxff f3z/rV+T18lppVL+ZLegZndTl8bXw4+XfjALt1aEm/1WCgeK56UfL7IMybf7adbbvBpdjzvpJ0tK ZfPwPAjlmw7Di9vj9W5/Mr8Ymt3YXuTczShdir67vjzLNugtv3S6JBn3jsSvBv3pN6Pz2bH4Rz8W N+R+O9rmWLY0ub/1/Y1XHto2V9/Vnp1TuFMf7l0wXn1zcPT7Rb9zEapTd13tW1kgyxPSPx+OsgtI t20j6l7nb/SHV9fTm5Mef8grwzC0w1d5mu+0mh/HyXCSXyoPre1q1A9J7xcNuTOfpufVaXI9LA37 o1LnYhbG84skfD54cXk1KM3j+fi7JC/eb9Ppxaj7yfYbvsiSu1A7l27+eTe5CiW3xabbN9K8gWfB L2upcXnnploEuCywdrO5a1irepBqvfJB9iBPfBKtVUR4N7zS1y/v3AlyGr3qZDVuHNUa7Tj0Ce1W 3GxUy81KduwwSEMvF9rPVX8cmvRNwqJvQ5eWDsLeh95iEurzuN+ZHGVJqZcrtUY1bjVbpWqrWUvD ZO7srBZacL16ljbTTqddrsaNZi9tthdHvdZT5057c+5kvVkW7XX4eP7FJeEU/dDsbG0p7wKLcL4u 3ro3s8/rf9bff5H0B2+/SF/3O+lCJlbKzUazVOumZ9VqPW7HtaRebteSNK73Os1uklRb9dZZazET 901LOIbu3BwvL6amWas2Smmj1+k1Q4s/S9tn7U692gs1pdWuduNeuds4665OzbIsvs3cA+RDPsxm 61eyfRpdXiXF+Bn9P1FvcB1mjqsLbXnSfroK86IwaStCDZU/G71mtf8xi+1xb6RcPOWwEO+yWw6q 7bU3Uq4MY8mFjNbijZTt7e7m27diH7/8/vsfj4+OyqVQc6uNcn1xRrXvDtxkz07XYtqbTmSFAB92 j+b622u2u0dzmzxZd/PkjqV62Gg2nvR6xOZ27x7N1Q3uXrfzKkww0tP5odTF2uRuc2Pn6nxe3xEf ZBB4yq74sUatlfOCm+ObxxwE1hfSYof2lDm+Q3e6rqee5eITjKi7jhdrqlO1HKpTpVqtdsr18E+t 2YkrvWq9ldQ6lVrzrNvtdNu1w6XkcWbZ64v+1TQZZxfRLtPLSdYpFTUorwCLJX8Uda7H42IxeV6O N1tN+n9PwxFFrbR4ZW27Sv7gUjh0m5gn7CCF8pQN9vHq8+5pecIavXlI3dzHP+4x0fbd4/tLx7Ju 6ec0/W11bx23w8wo7bSTXiizclpNqkmj3e4l3Uq32Ttr1UNy4p17pXnjO0TUT9n23kNWfTW7UhJN hsnV5GI0nS6fMxxlO3+ZhmOU8PHvSX+ap/vydadzVGRE1oU/C134YWewOyT6XgrzJGUv8vFlyzI6 RB/wuF3W9n3A+0vHsj7g69H1ePB2TZ9Ui+N66SxppdkcLW6X016n0ywnjUpIWrfWqTXSVn3noWXe DRwo9qfsCZ40ww5R79/PNPh+nf9wpuJFR74mPVkLLNW7vUraqHfLtWYjpKRSO6s3WpXWWbXbLbfT XvNhw96DY3/6ke8xMmxz1X3UUXf72vvekrHyUsOGeWS7UWuXamch59NKo9xrnCWds1YSBpbeWbUd l7u1atroPOgc/0ES8OSn+p8w2w7Rcz/2ULN9C3ifKVnaCJa3xbhVKzdKZ/X2WblVSZqtuJr26u16 s1Vu1uvtMMCcxd1OsnfFf1CkB6rsD5uer20hT5Cnh2gVj5TOHc5xPn0CVs/cVyWmXW83S2HyVA0N r9etJ/UkqTbC8WK716p2Q6dWbqbts4dN2x8W9dPP2Z8iqw5SwT+4oWqvqwCPtQfbjpuPVNx7zdfW 5EO9EtdKnbjdKtd61SSudKtntU63lpZrtWqrXA6fhwQdbgz/QHKlSMyGKtKqN8qlTq0SMiGuhBQ1 wkFFu9eodcqtzln3rN7phC8+5Mji8Q4Ddz24eD8pWX2AvL5kquVmKJl2u9Hu1iv1bqWZhOSUO42k ndbPQrk0Qn3qxA86wHhwi3msy0YHrLlPfvzzeN3QvmPyY/UA25+MOUhNf/oTMo/bQK+n/UHpq1fZ EolJyLDswdX5ufv+ID3N/znJFhvMFkfkL4vrEifzixInpenl1Uml20visziMIJVmLYwkrXpy1omr 9Wq1cdZspq38GdRX6fiyn98qfjr+/U3x/0c8FjrEDO1xi2DLGeKHMKf4YBLyFI1i15H0m6wJTxen Ws1quVJKa42zTqfW7dZbcTdpdXvtSq2TxulZtVNvpvFTHlA/RU+86zT1fY4LDyu3pxwKnr6GLbsJ dnMNe6R0bl+pHisB2VhY+vqr8O/P436orf9SOY1eDPvTfjLo/z1Enw9wz7M1K89Hw17//DQdFoX4 Syd7M3ui7/fDbJnls+n4Ov23e2/nIT7Ll0nOP3gx7KZv0snSz/48GI0ul3/0ZTjkm34/zLPk/ifP QwUdh34r7c7e36p48yqR7fRpVDz3IPoquQwDQZSnIZslZI+iydb95pX1gBOGUK4hxaPrYTeP6qs8 piLrd035d99Hf8nW9yWDWbJDzcneLPZotkO/J5NshWX2cIpRlJdz9PEBd+eTxZWqK1N9Gn2V9x/d 6CiaZCthh520331WqTWP8rv1Zjd5lMpHUX8YUhviz3P+kNm/U4/xMtT1y5s7PrKOK0tNMj1kgrJi 2S64/rA32mf/VpbG53m1OGDkR1E6DP1TOnlWWSzhLJ68iFv1+tJbMdd0wjf35LxzK2iYovx/eY05 Cf/e3AGax1IOb5bX3ywcallUbV1OFtN6O+rNl2zPu5d9Rpws7/YadB7l/uatAn3axW+V8lNHGD91 hJU1M6T5MpM78cblSqvZqsfthcK5CelRV2fVFhroYrzLVmfVQt8dr1qctTqIJWub2guLs2rLHqW8 uIxn71hudmGXFVQ3sa1eQVVbfDZ2Ue5Pt4JqqzxZs7TpkFm/ezSbVlA9ZpO4u4JqQ6PYZjHUsub8 xD1RdVUZPb9IO7/ls50waI4GxdPxbx5KsiTbq/eS/s3o/PkgTYZZULNnOpY/Xf/EhxDJ7VarnvDw TjRPkEO17ReqHzdrzbjcbrTD8XytUomrzZXr1B/j4lVI6pcvX37/cqtl45NZmrPiPUhi7jzO4CLp jkZXpXwCWfpuNM3iDnEWCbnzEIvijfmDK0bDQd5cDpKadx6wcC9Fq7OndJ5Oi98/vv9+8eSGuLL4 6IYtQ/5oVke+mT2FpZRXjC+LerEssjDzbO8Q2TdZcJP8UQ+zl7OHYJT3fyjEvOo/ZVs7VG9U36LV /vj9Dz98+UW2smx0dZV2P81q4NsQcv4Q1fmm4edl0h8We5pV0FVRLf5hkizYvKkVQ13IqDwPtg5g yQylfBPK6coecofgKocNrnbY4OLDBtc4bHDVwwbXPGxw9cMG1zpscO0HBPfDyxeHbhkrgnxICa8I 8iFVekWQD6nWK4J8SLewIsiHdA0rgnxIFV8R5EMa4YogH9JyVgT5kNZzMyiteDLc2hHq/lmt2QPa fh6Nf8tDnp0u6GfHsePrq2k2DZiE6KLpRRr9nn8rvMwmH3uN1vOEZ+fPijeXBFNMHy7DwdVZ8rYU hvHTPAlX2dw7TKPmH/yaTqdv8+eTvUoH4RBw9myz2ROOR+OP4s/qjWpcby4e1M6Sei+K2QQiWh7Y Z+X8aLJ82iiXq4sHjWsTvWeI25f9zfH344ayvubc+/22AmXXmy6yc+y/z6aa2eRvmkx+O5pX+9M7 89oXt5vdHGLskNlZ+MXefNY861Zr7XTx6HyPnVk7/13Tetcl83l2vufNdHZG9/loMCjOYnzWqHbP GiHwHcNbbBnZvg3SaWkW02eVSqfS6pw1/3EStpwcFRcJrifju3+D87hcatdKcf6Nk//YI+dDhLNd +iyuJLV6rVp/aAuf78cXaS+5HkxnsRzHjXq11a7VqgtLiDe1xoemMtvXL8fj0XgeRi05a1XP0h27 hQME8QEU+uYkxLVWuZwk8T9OssfQ9zvrE1H88nt6llxdTWZbPLQmNirVdjdpLfx1iPWZO0l66fTt Y9XeuFZrlmvxkjOuu9TePfZsofY2Gu16pV3esZ0eIIgPsuosJqoIsPRzevb51dU8ZfV296zdq9X/ cbJLmu4OOrvX6fx2xHmWV1uNermx5PTLplA61+P+9O1NycVn9XZ76WxxhwZWb9cbjaTRefwG9tv4 rF7cYLF7XVt2/u08v9siTDfun4m7+8u//zq5WhnZvMb+dXI1y5PT6PZ1qZtOpuPR248XT8utTe26 GB/U73z03YvvwzwvP1WY3wH0WSWN271O5cFN951Qz3qtXquzcO/C+lC7RVp33Oj6csfMynqG68UL autL5HLh9t/1G4Rfvklfp4tXFzclrfPbjkk7SFs8QL+wthp+nUwuZj3Y7OTzZ81yM5R3e5ceLE/o IbrBfPT6ISTj3jBWKTfrnWZr2aHtppHwgWFlw0rSzw7fV5RmdDzrLrPr170dOsBtQr7tVB8j9I+/ elYk/eiXk3/6t6Nf/u0orn9yG9G3SV4kB43vdoey6B4UY7HJd8llGjZ7tuxPBy3fLvuDIkUSJjts NWsuX83Tufum2R+1yDb8xz+V/uerH56FkeQo+qdS+HHn5Zv567+++uGr4vW9qUn+9QeOmCHIZ7P+ PIvpzTwBJ0UX96z4EX6fd5TP5i9miezdSeT/nG2b9fXPsn+yBE+Oi478WfEjvBU66Wfhv9lG2Tbb TrKKvJuX8j9usmu2B7d7cicFN7G9k4wlO/TAvHx4WdzL9C0zZZu29sCEZW1ylveHS9NjhTurA48T dqhRjxJwUSkfJ58v3zxKuPNm81j5EVrAowT98asHt9NfFrrhf/vkTloPP1S+yru5X/IOd97vzgeJ Wbc9603nw8W8R85+hgHm0dM373l/OXn8qLJu/Zd8kHn0uGaDxS+3w9jjl3QYq37JxshHj+lm4Pvl ZnR/igzN5hO/zOYX+8f3n3XCt/OGu+/i6sO8D+AcWfZA3QNeU9n2aPZFd348W+1WO9XqWXP/SIoL q6X/N7tPuFvc4vXDaDT4rNFpN7vV3s6XHTddCLu7Xuyb8fWfB6POb/n5m9Mofx3ly7Gib17+FGU3 8hV/frd4gven0efTaXp5NZ3d6dsbp2l0HLfa5WatVK1Ff/tztpJjOpomg2eVUrkVffvnra7EfTM6 f5mV2/g0unk5v8K23Q1i36aXxYKqfIHPeH6D4uzK5937V5dczCvlezzbdNN1+TU3Cj6/WW2S3/S2 ezLe3X7/tNw56zGZ3Sqw0+18N9Vkx4XGpW+//PHz0tFRXFqz8mEe+Ox+vfmfSv9xnBSZdnOden5D 37T4ZOWBx/KAPqs0WpVuJV7dgtYvvl22Lzsvqd2mqO6uF7m5eTE0rzxXd0r8PMR/3Kb42bLdWN3F 7r7yd6ciXx78mj8QsCz0DX8ioFWKf9sxAcsfcL0i7i0ecd2q1Fe3qR2SsMvjqm/C3/2JGcWOntzu ZbHksdNs1pJGvdNqJu1amlSSXqscN6pxGPAqlbP4bOtnZrwz6vxXXqW8/E6TQ6xS3rOQdlinvCnt B16nvOcOLa5UXp3ulSuVW7crlbPuYruFynsmeLuOYL+lynsmafVi5cUA8xXDe+zjXsuV94z+zoLl 1mJJ36xYrpXKy0aGPRcsZxXnJOvpNyxYXjaMZEuW49rCkuXWLkuW19ajNUuW1+/xbJHxTqP69ouW twz2iRev3I/w8zCanQ/DqDvd7+8fT7f6g8eVxoHn8J9GxcKaYvpYlP/iwdMs1t+TQenrb7IDx52O UMLk+tXbUFnX7VVRJW4j6Mym0xfZTP73fODJav+K9p/fjrZb6S1G+e3odahycXZSuWj8WdVYjGs0 6GbRbVUyxdKs3bLrZtfzFWOPHE9298GqaJZ0E7Nj1fye1+Kkw23Tyvqu/IOo+KQYLbKOsFRaVqP2 CP6bZHyeRrcHzI8X06vLrFnsGdMsjjAKJIOXRW88O+od3T8AnnXVp6e3f2v89N2/NR7nM7zTbjED 7LzJv5JUor8Xr1qNaPpmmH+lkoWXnVo8HZ4s3uW3XeKeLh23fwK7WKJd/KX0v4ym2drQfq8fjs7z c0grE7R4gL456J+TaZiOd798nR0GZufS09OsZ5rdaJ/NPrMd+G7UTYvZaTc/VXs6a//jDX1MXnNW /9X4bRL8r6PR3/JXRUpDiRTDS7Ey/XghyNDiwyQg67dutozyvTvKd+XZnV0pLj6nz+7t8FG+g8+2 3MH/rIX5/KI/6IapVrau4nyhUPcotrwHvk3Z3W735nzU3Tej9OoivcyPgYYhPVF3XiS3z3Yo1owX +fTL2lJYPHpsrJv6PA9Tz2z0nFxcT/NZziSdfrpxnUK+TvnT2RLYIsTJs5VZMYt59nyLVXHeDWxp WOtKPDt8L/rMq3C8uKafyqp0KIGsV+3k2/+Qd0XXoT8v+r7b33uzefppPlOOLkIUIfWnUdyoHFWi Iqqv774ZV2tH5ehOv/2nJS3nv1ee53kaft7L1aPjOLpZwhy2XXox5Wm6hHcaxFZdw8qkvlv499ZB 9zvpbBHYypnMNgup833dOoBVy0Vnj3rZMLfeejHw4YKLDxtc87DBVQ4bXPuwwdUfENym1ZEPrnat w+5rbavgHmEWWi4G0/N0Ou865lPAOG7O54C9+9KbGeH17LkT83nh5j5l31nqU6fzYWNV+QBjVaXZ Pmq9O1Zlb64Zq45+vDcQvf7HfyyeHV0xsm+zLHeX1rZ6JemhQjncEuG43ayfNXuLN7rv0IIb23cI ey45XnLlcm0mHDjEO+tqK0kc18pL1ous33rlrRn1tNrI/oDEjuFtXjtW7XTL3biSPmDl4+YMuXM7 UjWu12uV6uJ6vh1r63w/3l0wWK3GrUo7rsa71oSHpnJhwWAcJ71uu7O4wmRtQnYLYslU+c5jr4rO 93ThsG7HvPkAKtEWSah10l6jd9g1kDvV7GbvLBzx1HYs7xWr+w7UGpq1ZrVRqezY2h6+YwuNodmL 0/jsrPeAxrBfEB9kzdnyzsBKo9tLq+XqbncGFtOW3SvzvQVjjVq3kj2sf+dQ7q+Pq9bOymk33XUY fKcC1s66lWa5sWzyceCWtWbd7MZatv6O9tlk8u+/LV0ne5jAix+PGME0u7XkscOffJFOk1Ddu2sj mrfinVcUr03pXru234riOO32zs5qD54I3Q+1WWn2qmdLnnSwNtRVK4rXb7RsRfH6Frl8RfH6Elmy onjtBitXFG9K2rIVxbsMk/v1UgfoMXddUVwp1yvdOFl8IMyGhB5igFi+Cjju1uJmo73j0dshwlq5 bGJemtusKN475K1WFO8d+p4rih8S334ripfGuMUCk6XbbVxgsrxObbXAZN2mNyuKT/7+25Yrg+/N CWYrxRaWIp/cHxR3C3pxQF2ztjn/8h7hz9Yxz97aafvbGcujL4ReXPV8k+b7K6S3nTnfXx70jx2L e/fq8YBCf1iFefpl3u/Wrr0q5DsVZ486uU893rL2bNOf7lhD1q3Wffw03ObA+03HTQX4AJJxrx09 SnoeK9wtFvHvH/aGRfz7z9M2LuLfP5/XL+LfO9xtFvE/JD82LOLff6q3zSL+++PXL7N50VZLjp8k YXcmH7/cmb18OAm8Hcx+uZ2cfWDJuz9X+GXJbPXRE/wUz2Z4SPp2fDbDg6La8dkMD4nr5tkM9ybR j1/aOz2f4SEx7ft8hgdm6s7PZ/gvdfi884a77+KOz2d4qqsw657MsMfl/92fzFCvd5rtbm+bq813 7/iZX24u7uK5e0fxdlee8xs+Zjc9/TBOr1besJWdpi0CjqbZGtphccfwvVuYD3qxe9WDJuJ2r9Zq 9fa4xWCbO0MeYwlFvt5lVuA3CxiqNwsYmje3gR3H8VMspXji9Ox9x1+2Tvr2LrrWvnfRFZ3BI97x 9zTp/BDuTq8eZVVi4f706lFcrd+/5y8PaPd7zt/bzYyt8lFt4WbGVnlxx/50r7TeuaHxj/PCDqlq V6th3GpWW9VWrVyLa41at1xudBq9RrURh5+tRq9ZbdYqnWot/174N7yOw1bhC+Hfdninkv/XCJ/X jib/aB/dWy7RLL/zW/G/Vrtea5XjarkWfsa1ZuWoFnbhqL30Vst1+f9FfzL/w5UPHWCW3in1amUw sxWCq0I74BrUm4Td/RudGxajrot+p3WQG27DnMXw3Yvvi/BuB/xsLB5lNf5uqrOSmRXSnT8sWg8V sZb9dZPORRQOKNcsN9qi8O7Wjnxlx2xxYrbaJl/IvM2tcn/LnvSxxVqSbIF8tphk9sdlsxXTo16U LYUt/k7q3/MFTfdaY/6Eh0nIlt+LFSufRv/rTbmc/bd0CfS6pZTLd/rzYSi02Z98yR8dESYpWU9e /PWYzs2tfyEZxfTlbntZUs9Po5/yB4Zkyzvzryfdbj8LPBnc2dNZVV5TdEfRoP9bmveSs5t+50uq ZxXlSYfu+0s14xDpfGrR+nDWaj5w2H+KfXzYotFZk3j40tA9s371zQnvdY+3SdbiyreL2452obOq HHYt7NbTwPe7ZPdRZ6uHmdLFi2spszdD3792SrfLasnHfiZc3Kkn6Vl39W1NB+2/K0X7CvOSyXQy 79vq3YMf0OTTiqJcS18/v2lZ8+eX3HnrxeXVIM2ebZL3Njezn9tp283OPNu+oe/b9T9V9jxkdcGu ve4u5492DXvb1l45QGtvV4+q7zb2djVv68dxOd76CG6bg6T5FHb9oHB/UOqO0uKPj+e1pzQb6cbZ nyQPE9aVxxqPd2Zu3+axX5ey7JRTPb2ZM7Wf7BTYY6XnA2gBtWUnZ8K7cbV58HMzu58bWFbf9jg3 UNlwbuDAB6s79Lwrl2s/4Di+fHr3f+FovtluV7Y6mn/McWDTnj5Cb7Xvcdx+B7DLeoc4Tm66h+TJ uqtHS9CHMD+vLOuwsrfjausD6LG2eP7ONj1WeUOP9e5C99mL8H5/OD+L9HhnCpf0MK3WVj3MFtlz iC748U6/lQ91+q28z+m3lQX3VZItg397Gt17N+oVb+dVIXtA5XX2HOnpxWhlvZqFfVNFb86Qh8zP 28CKAri78axSz45Gs4OxUNF32H75SLBLCNkZiIeFsGo0Wnx3nreDcFj+f++XriXvzkLdFOCqoWH+ IKtRb94fzp/dlbXiUXZ8PF181upCU/w29MI3VWH2lKwiH/PQ84//Fipt8fis/LnW2YMIw3tF+hZX Z7U215SF99bncGvZQ3hnKfw67Mmd9F6EX4s/8xBqwadRcQ5lnlel7NP8idufRr3J3RCeFR3QL8WP 40r5qH6U9bOLTzZblpZ3rvW8+vH7H3748ot3UrXPXs32IyT2trzzPZzdA7EkzKKb7E1CvRmkr96G Uru8E1BnkCbjrMrcfjr7Oxl5ycZRWpxVmWwd9Mv0cvR6/lz2SRHi7cO/T06eOpwvRsO02M3iLFvY twfXp/lMpfSH/x9QSwMEFAAAAAgA7laTQlh5EYYDBQAAmREAABsAAABoYWRvb3AgbG9nL2xvZzRq LnByb3BlcnRpZXOtV1tv4jgUfudXHJVBmpGG9Da7WiHlgV7YtgPtiHS0zwYbcMeJI9uhw1b89z12 HCAkFIq2UtXGPjd/596EGzbhCQMtYwaUTUgmDMyJyJgGMyMGxiSBEQM5Z0pxShl+LUAvtGExpEqm TBnOdKM5I1TKNFBSmkDI6ZSp8P6x9/R1LBMtBWvU3N/cXv38e5sA7wLKVRhsHky4YOH6u9FoNprw IEcQZXFM1AK6acoSyhS4m5+awUQKIV95MoVcGxgJGmlAexb3nRJFDBKjfPt6RILa96EMry0mqWI0 G7PgRY4868okUKgEOSjhYtHZz7QC5SHqInXjAPJPb1XglvsYN/Fq7yJCEAvnm5l9ijQbUNmjspcX cFI15SRo4D/fXtxZ/x2Tv8LtnCXmWmaJYcrqttTWPc8zxfRMCuolGfdtD8Juv5+7+sYCDENE23L0 LPSFx/HeMxJ/EtwMe91QqimekPGMBfm1k+FFWAkrATXcgSVYP8TH5PK0dGJBXtqXWKGYHoDZEnOa 8OnM1Aq9wVD7QQy+PwmDBf60B4M2pVbE5VmbkgWMyPhXljaaddwD8vvKXd/j2e/w8qxWhyALmZnq 673evru2Gv0B5omKiemANc76pM/mTEDuykcSu7MB05pM2TsKg2uZIASay6R4YYu+3UdPf/15dr6E VgqtcQdacStxUTfKct+Xjah/92EK2n9YFW8XS/jc6nVag8+t/pcv0M5V5kFUFJomdCmFE/95YqPd RqqPfTLCUgd8AqgVXkli7H2mbZJwDTXR5uVUIb/OL3bFmecLDFFTZsLIJVuAVXYX4WGufZe5HsXF 4nQwOEVM7u46cdzR2jsM0fQ+c/g9E/0Lg6GUec2bUssoCpNBUpsi9i+nYZIJsX3F9VgwkmRpOCFC s+3rRH5nLI1SwY0Ov1XkSkME2mLTNOL/svD87GybJs0QV6TxMozKKkrwVw+ZITy5k5nS4flFxbvP /VIpKRXewCOyy8PI61Td03UpKSOzrGPh+tojU+FagVbLWEGlqnWLYln33uMCbc34gVpg4ypi40xx bC8ko9wAWUeXN157gnVnK1i6mZkFjq2uNEWH9wCo5d/bBSqWVfHMBR2F6CbrR+prwl59I/cS84+g QM236adebzV7vUMXEIr48jke+Ex1xWBnL96u4bW9uK4N1zB+rA2jvW4o0xjYQHDSs7+lxqprlWBj XdeRwdUumq3mW0t2mJ93cx7e5Xxn+z9E1TVM6+JeZPu/HwK7LjlFPrPZFiqET1hm5zoNRLE86qid g+yUC8KOEiiqFF3VYjqjE42ppNDaIEGNiaQs2FSep3j4T3f4aP18nWkj49X86LToxmoDiZlRfKxd bLg7N3I3dt66LWSfjb7g48rxrDAOiuXlQDbbJ8p8zT2MiMgmAgdqy992sc6Yymtdojwwoy/RmXyk cBeoSn1x984lHNcGHqciwJncBDNj0rHg6O9giN/RZZSThLfD4dMwj5pH7PUbBQHw3SkUhbJYMGTi FozcSshFYt5StoqWVURbeYW4amLNieKkRJNb4ZYN8NvGpjkR/qNxFMxszMpJEdEQ5yOutsFL+WTC lDMJMcvtdkFmR8G73OaBR7Zq8OaeUzM6WFcUbtlaiXbstFUVuD4ev98g8966+u5iWZk9rMSjOtya 8djBtEbgri3r4Ay/T34oOcWA15/wy3sj3IuNW8y9PauWeYSivMn+B1BLAQIUAAoAAAAAAABYlkIA AAAAAAAAAAAAAAALAAAAAAAAAAAAEAAAAAAAAABoYWRvb3AgbG9nL1BLAQIUABQAAAAIAKdYk0KQ 0VqzXQEAACEDAAAkAAAAAAAAAAEAIAAAACkAAABoYWRvb3AgbG9nL2hiYXNlLXJvb3QtbWFzdGVy LUFGTS5sb2dQSwECFAAUAAAACACuWJNCdYE+EI82AAAuPxQAJAAAAAAAAAABACAAAADIAQAAaGFk b29wIGxvZy9oYmFzZS1yb290LW1hc3Rlci1BRk0ub3V0UEsBAhQAFAAAAAgA7laTQlh5EYYDBQAA mREAABsAAAAAAAAAAQAgAAAAmTgAAGhhZG9vcCBsb2cvbG9nNGoucHJvcGVydGllc1BLBQYAAAAA BAAEACYBAADVPQAAAAA= --_004_DEE652818364114E9649496951412C471444E64217AUSX7MCPS302A_--