Search Discussions

155 discussions - 509 posts

  • Hello All, We use the tarball hadoop and hbase distributions and are having some trouble upgrading to CDH4 from CDH3, specifically around the native snappy codec. It was not included with the tarball ...
    Chris TarnasChris Tarnas
    Jun 27, 2012 at 12:07 pm
    Jul 22, 2012 at 8:27 am
  • Hi I have downloaded and installed the cloudera CDH3 vm on vmware. I am trying to run the samples that come with Hadoop Definitive guide . I was successfully able to run the MaxTemperature sample. I ...
    Rachna JotwaniRachna Jotwani
    Jun 21, 2012 at 9:16 pm
    Jul 19, 2012 at 1:53 pm
  • Hi, I have a semi-working test cluster, which was previously on CDH3, I've recently upgraded to CDH4 but have only just got round to looking at HBase again, I was having problems where the master ...
    Darrell TaylorDarrell Taylor
    Jun 25, 2012 at 6:14 pm
    Jun 28, 2012 at 8:26 am
  • Hi, Whenever I try to start my datanode it gives the following error : datanode dead but pid exists. I have visited the datanode log file , here is what I found: 2012-06-07 22:06:31,480 INFO ...
    Jun 8, 2012 at 5:12 am
    Jun 10, 2012 at 8:25 pm
  • Hi, I'm using CDH4 with Yarn. When running queries in hive that require a M/R job the job is submitted to the Yarn resource manager (I see it listed there with status "KILLED") but is immediately ...
    Jun 27, 2012 at 1:38 pm
    Aug 3, 2012 at 12:11 am
  • Hey all, We've did some test on CDH4 in our test environment now as we decided to wait for CDH4.1 (mostly because of trouble with pig) we'd like to revert the system to CDH3. My simple question is ...
    Markus ReschMarkus Resch
    Jun 18, 2012 at 11:12 am
    Jul 17, 2012 at 11:14 am
  • Am trying to install Cloudera manager 4 on my cluster in order to update the upgrade from CDH3 to CDH4. However the manager install keeps getting hung up at 70% (Attached screen shot). Am not able to ...
    Rakesh Neelavar RaoRakesh Neelavar Rao
    Jun 26, 2012 at 7:10 pm
    Jul 2, 2012 at 6:34 pm
  • Hello, I'm a student working on a project using HDFS from CDH4. I am trying to use the append functionality but I encounter som issues. Do you know if this functionality us stable in CDH4 ? Could you ...
    Julien DJulien D
    Jun 25, 2012 at 11:30 am
    Dec 30, 2013 at 8:56 am
  • Hi, I just ran a lot (~250) MapReduce jobs on my cluster and now I have 2670 missing blocks and 4243 under-replicated blocks on my cluster. Four out of five datanodes are dead. "hdfs fsck" reveals ...
    Jun 21, 2012 at 4:17 pm
    Jun 23, 2012 at 2:16 pm
  • Hello, I have a 32 node cluster. Last Thursday I joined in 8 new nodes, using Cloudera Manager. I bought up hdfs, mapred, and region server daemons on these boxes (via cloudera manager). Once ...
    Mike DenningMike Denning
    Jun 4, 2012 at 2:53 pm
    Jun 4, 2012 at 7:01 pm
  • Here is an interesting use case. There is a requirement to write specific output from a map/reduce job, based off of a specific key value to be written to only a certain number of pre-identified ...
    Jun 15, 2012 at 9:45 pm
    Jun 19, 2012 at 3:02 pm
  • Hello, I have the same problem with James here ...
    Jun 21, 2012 at 6:04 am
    Jul 15, 2012 at 12:21 am
  • Hi, I have set up an distributed cluster with 5 nodes (using vmware hypervisor, centos 6.0, cdh4 first release) I was able to configure to a point where distributed MR jobs work, and confirmed with ...
    Jun 16, 2012 at 1:07 am
    Jun 18, 2012 at 4:12 pm
  • Hi i am getting below error when i stored gz file in hive hive CREATE TABLE app( patent STRING , gyear STRING , gdate STRING , appyear STRING, country STRING, postate STRING , assignee STRING, ...
    Jun 21, 2012 at 6:57 am
    May 28, 2013 at 8:24 am
  • On CDH3U4, what determines the permissions of the folders and files under /var/log/hadoop/userlogs? In my cluster everything has drwx--x--- 2 mapred mapred. How can I change that to be drwxr-xr-- 2 ...
    Dave ShineDave Shine
    Jun 29, 2012 at 7:56 pm
    Feb 11, 2013 at 5:35 pm
  • This post [1] on the Cloudera blog discusses how to use org.apache.hadoop.metrics.file.FileContext (now deprecated) to gather runtime metrics. Though on a recently deployed CDH4 cluster (via Cloudera ...
    Jun 29, 2012 at 3:25 pm
    Sep 21, 2012 at 5:08 pm
  • We are currently running with CDH4B1 (will be upgrading to CDH4 GA soon) and have configured HA so that the master will fail over automatically. In testing failover we've noticed that in our long ...
    Jun 13, 2012 at 8:53 pm
    Jul 6, 2012 at 2:20 pm
  • Hi All, We are running a mapreduce job in a fully distributed cluster.The output of the job is writing to HBase. While running this job we are getting an error: *Caused by ...
    Manu SManu S
    Jun 6, 2012 at 2:25 pm
    Jul 3, 2012 at 6:46 am
  • Hello, Is there a method to run hbase org.apache.hadoop.hbase.PerformanceEvaluation with LZO, or Snappy ? I use HBase 0.92.1-cdh4.0.0. i saw this jira https://issues.apache.org/jira/browse/HBASE-1902 ...
    Sebastien NahelouSebastien Nahelou
    Jun 21, 2012 at 3:50 pm
    Jun 26, 2012 at 11:08 am
  • Hi all, I am currently using hadoop cdh3u3 on a cluster with 16 nodes. Previous night, I launched a job which must take a long time (about 10 hours) on this cluster, using nohup command because my ...
    Jun 13, 2012 at 1:26 pm
    Jun 14, 2012 at 1:43 pm
  • Hello, I'm using CDH 4 and was able to set up HDFS with HA. I tested the automatic failover and that worked perfectly. The problem I'm running into now is that my hadoop/hdfs command is not working ...
    Huy DaoHuy Dao
    Jun 11, 2012 at 11:46 pm
    Jun 13, 2012 at 11:15 pm
  • I'm trying to copy data from a CDH3u3 cluster to a CDH4 cluster using the following command (ran on the CDH4 cluster) hadoop distcp -i ...
    Jun 14, 2012 at 6:27 pm
    Feb 19, 2014 at 9:41 am
  • I have a long running job, and I would like to be able to de-blacklist a tasktracker from the job, but am not seeing anything about how to do this once it happens. I've tried restarting the ...
    Jun 22, 2012 at 6:16 pm
    Mar 14, 2013 at 2:26 am
  • Hi, I am using ganglia 3.17 with cdh3u4. I have used GangliaContext31 in the hadoop-metrics.properties file, but I don't see any hadoop metrics in the web console. I am getting other information like ...
    Austin ChungathAustin Chungath
    Jun 25, 2012 at 6:44 am
    Aug 7, 2012 at 5:18 pm
  • Hi, We have a hadoop cluster running on CDH3U2. Of late, some of our map reduce jobs are failing and more than 50% of our datanodes will be shown as dead on the namenode webadmin page. Here is the ...
    Jun 25, 2012 at 11:39 pm
    Jul 4, 2012 at 10:03 pm
  • After about a week of researching, logging, etc. I have finally discovered what is happening, but I have no idea why. I have created my own WritableComparable object so I can emit it as the key from ...
    Dave ShineDave Shine
    Jun 26, 2012 at 12:59 pm
    Jun 26, 2012 at 4:49 pm
  • I tried to copy file from the local linux system into hdfs file based system from the root , I typed su - hdfs for hdfs user I created one hdfs directory say /user/cdh and have linux local files ...
    Jun 20, 2012 at 12:47 am
    Jun 20, 2012 at 7:21 pm
  • Hello, I am trying to install CDH3u3 in pseudo-distributes mode on a Redhat 6.2 system. When I execute start-all.sh It looks like it works and I can access the namenode and job tracker via the web ...
    Tim PietyTim Piety
    Jun 13, 2012 at 7:54 pm
    Jun 14, 2012 at 1:59 pm
  • hi I am new to hadoop when i am typying the command bin/hadoop fs -ls /usr/mohit/wordcount/input it is showing 12/06/13 11:23:21 INFO security.UserGroupInformation: JAAS Configuration already set up ...
    Jun 13, 2012 at 5:54 am
    Jun 13, 2012 at 12:43 pm
  • After downloading the CDH4 package, I execute "sudo apt-get update" and get the following errors: W: GPG error: http://archive.cloudera.com squeeze-cdh3 Release: The following signatures couldn't be ...
    Jun 8, 2012 at 7:51 pm
    Jun 11, 2012 at 5:10 pm
  • Hi all, I have a test cluster with 1 namenode and 2 data nodes. security is enabled and it works properly. in one node i have configured HUE also. Now how can i make the cluster public accessible. I ...
    Sarath P RSarath P R
    Jun 2, 2012 at 5:53 am
    Jun 4, 2012 at 7:07 am
  • Hi, Wanted to know the experience of people who have used sqoop export from HDFS to Netezza RDBMS. Does it generate insert statements for each row to be inserted? We did a small POC and saw from the ...
    Neelavar Rao, RakeshNeelavar Rao, Rakesh
    Jun 1, 2012 at 2:11 pm
    Jun 1, 2012 at 11:13 pm
  • We are planning to run CDH4 with MRv1 due to some legacy code that is not compatible with MRv2. Will we still be able to take advantage of HA Name Node if we use MRv1? I believe so since Name Node is ...
    Jeremy BennettJeremy Bennett
    Jun 14, 2012 at 4:01 am
    Jun 28, 2012 at 12:37 am
  • I am testing a small HDFS cluster of 4 machines. I successfully installed CDH3 (and now CDH4), configured and started HDFS, etc. As part of my testing, I filled HDFS with files to full capacity (100% ...
    Aaa aaaAaa aaa
    Jun 24, 2012 at 1:43 am
    Jun 26, 2012 at 9:28 pm
  • Hello, I'm trying to run some job, which requires lot of multiple stream writings ( jsons of various lengths ), and I'm getting an OOM exception(not a java heap space), which looks like it comes from ...
    Jun 24, 2012 at 4:20 pm
    Jun 25, 2012 at 5:18 pm
  • can any one send me some map reduce program in java like wordcount.java. --
    Jun 22, 2012 at 6:11 am
    Jun 22, 2012 at 7:35 am
  • I have a Hive query (hive .7 & hadoop .20) which seems to be stretching the limits of my 7 node configuration as of today. It's been running okay... The error I'm seeing this: *stdout logs* # # There ...
    Tim R. HavensTim R. Havens
    Jun 21, 2012 at 2:31 pm
    Jun 22, 2012 at 12:42 am
  • I have a problem about "hdfs fsck -move" command of CDH4. I executed "hdfs fsck -move" command to a file which has 6 blocks and loses one block. I expected that fsck command copied files into ...
    Yokoi HiroshiYokoi Hiroshi
    Jun 17, 2012 at 10:00 am
    Jun 21, 2012 at 11:38 am
  • Hi , I am using snappy for compression can any one tell me how i start this in java..
    Jun 19, 2012 at 10:39 am
    Jun 21, 2012 at 6:59 am
  • Hello, I was looking working with pseudo-distributed mode in CDH4 (first release) hadoop and was able to run MR jobs in it. Then after this, I was setting up distributed mode with 6 nodes and is ...
    Jun 14, 2012 at 2:42 am
    Jun 18, 2012 at 9:32 pm
  • Dear Experts, Does anybody know how to configure the high availability for YARN Resource Manager with Zookeeper on CDH 4.0.0? Or is it not implement yet? Thanks in advance, Taka
    Taka TazawaTaka Tazawa
    Jun 17, 2012 at 4:33 am
    Jun 18, 2012 at 9:10 pm
  • Hadoop: cdh3u2 OS: centos 5.5 64bit hardware: master & slave: mem:32GB + disk 2TB * 12 + 1000MB network card --------------------------------------------------------------------------------- my ...
    Jun 17, 2012 at 4:20 pm
    Jun 18, 2012 at 2:33 pm
  • Does MRUnit only support the new API? I'm running CDH3U4 and very confused by the results I'm getting. I was hoping MRUnit could help me out, but everything is coded with the old API. If MRUnit does ...
    Dave ShineDave Shine
    Jun 15, 2012 at 2:10 pm
    Jun 18, 2012 at 12:29 pm
  • I sent this to the cdh-dev list without much luck...maybe the cdh-user will be more helpful. I'm attempting to rebuild RPM packages from the SRPM's of CDH4 on OpenSUSE 12.1 and it's failing during ...
    Darin PerusichDarin Perusich
    Jun 12, 2012 at 12:32 pm
    Jun 13, 2012 at 4:23 pm
  • I have upgraded my 50 node cluster from CDH3 to CDH4, but I think there is a mix now :-( I find the documentation for CDH4 confusing and varied. Can someone tell me the best way to check all of my ...
    Fred johnFred john
    Jun 11, 2012 at 5:38 pm
    Jun 11, 2012 at 8:04 pm
  • Hi, I am trying to use Snappy Compression on a per job basis and want to override mapred-site.xml (more specifically mapred.compress.map.output and mapred.output.compression.codec) from the command ...
    Sid KumarSid Kumar
    Jun 6, 2012 at 10:23 pm
    Jun 7, 2012 at 5:37 am
  • Hi All, I have a general question. Does the execution time of a Map-Reduce job decrease linearly with the increase in number of nodes? e.g say I have 2 nodes and I am running job 1, when I run it ...
    Jun 5, 2012 at 6:44 pm
    Jun 6, 2012 at 6:28 pm
  • our cluster is very busy for we are doing a lot etl work in parallel , but the hive server down for the following reasons (from hive-server.log), may be there are too many jobs submitted through the ...
    Jun 6, 2012 at 8:47 am
    Jun 6, 2012 at 10:08 am
  • Hello J. Ryan, I'm encountering the same problem while using rsync and hdfs-fuse, on cdh4. Have you got a solution to this problem? Thanks. Fantoccini
    Jun 28, 2012 at 4:46 am
    Aug 8, 2012 at 5:11 am
  • Hi, I am currently using CDH3u3 and make heavy use of Oozie and Pig. I realize the current CDH4 is a beta release, but it appears to be tracking the hadoop 2.0 alpha build. Is this correct? Ideally I ...
    Jun 25, 2012 at 10:13 pm
    Jun 25, 2012 at 10:58 pm
Group Navigation
period‹ prev | Jun 2012 | next ›
Group Overview
groupcdh-user @

152 users for June 2012

Harsh J: 38 posts Todd Lipcon: 25 posts Joey Echeverria: 22 posts Krishnanand Khambadkone: 15 posts Roman Shaposhnik: 14 posts Mohit Kumar: 13 posts Viji R: 11 posts Dave Shine: 10 posts V v: 10 posts Sarath P R: 9 posts Alejandro Abdelnur: 8 posts Darrell Taylor: 8 posts Hmy: 8 posts Sabrina: 8 posts Sandeep Reddy P: 8 posts Huy Dao: 7 posts Kevin: 7 posts Rakesh Neelavar Rao: 7 posts Subroto Sanyal: 7 posts Yang Y: 7 posts
show more