Search Discussions

187 discussions - 812 posts

  • HI, How I can setup cdh4.1 with eclipse Many thanks Best regards --
    Mukhtaj KhanMukhtaj Khan
    Nov 20, 2012 at 12:28 pm
    Mar 28, 2014 at 6:07 pm
  • Hi, On our dev environment both namenodes are in standby mode. After some investigation I decided to restart them but they are started in standby mode too. In one of my zookeeper logs I have also ...
    Marcin SmialekMarcin Smialek
    Nov 21, 2012 at 6:32 pm
    Jun 26, 2013 at 3:39 pm
  • Hi, Can I use Java 7 for CDH4.1. Actually, I want to use the Directory Watch Service in my java program. I use this service to detect the new file in watched directory. But this is only available in ...
    Mukhtaj KhanMukhtaj Khan
    Nov 21, 2012 at 7:21 pm
    Nov 23, 2012 at 2:06 am
  • Anybody ever try to load CSV files compressed using PKZip into a Hive table stored as Sequence Files? Is there a SerDe out there for this? Thanks, Ben --
    Nov 13, 2012 at 5:17 pm
    Feb 22, 2014 at 3:28 am
  • All we have our application running on the same host as the flume agent. Application is fine in connecting to base, flume uses the same hbase-site as the application. But flume seems to timeout using ...
    Nov 29, 2012 at 6:13 pm
    Nov 30, 2012 at 8:04 am
  • Hi, we just upgraded a cluster from CDH 4.0.1 to 4.1.2 on a number of nodes running on Ubuntu 12.04 (Precise). We first upgraded Cloudera Manager (now 4.1.0), then ran apt-get dist-upgrade on all ...
    Nov 13, 2012 at 11:03 am
    Nov 27, 2012 at 10:47 am
  • Hi All, I am trying to create a connection from java to Hive. But I am getting this error. java.lang.ClassNotFoundException: org.apache.hadoop.hive.jdbc.HiveDriver at ...
    Vidyasagar GudapatiVidyasagar Gudapati
    Nov 7, 2012 at 9:11 am
    Nov 7, 2012 at 11:36 am
  • Hi, I was using cloudera manager 4.1 free edition to setup cdh4u1. I was to setup a local cluster (ubuntu 12.04). Everything went good except that datanode is not getting started. The log files logs ...
    Nagarjuna KanamarlapudiNagarjuna Kanamarlapudi
    Nov 6, 2012 at 2:55 am
    Nov 6, 2012 at 4:47 pm
  • All Trying to connect to HIVE using the following code Class.forName(driverName); } catch (ClassNotFoundException e) { // TODO Auto-generated catch block e.printStackTrace(); System.exit(1); } ...
    Nov 27, 2012 at 11:27 am
    Nov 28, 2012 at 10:09 pm
  • Hi, As per the release build version: 3.1.3-cdh4.0.1, any user can kill the job run by a different user. Consider two users, abc and xyz. One user, say abc belongs to a group mentioned as ...
    Nov 2, 2012 at 1:13 pm
    Dec 19, 2012 at 7:57 pm
  • Hi all, I need to have 2 nodes to store approx. 100 million rows of data using Hbase/Hadoop. I know by default hadoop replicates in 3 nodes. Due to money crunch we are thinking about 2 nodes for ...
    Nov 20, 2012 at 9:12 pm
    Nov 30, 2012 at 6:59 pm
  • *Specs of the setup:::* 1 Manager, 2 Agents CentOS 5.8 VMs Installing CDH4 DNS and reverse DNS seem to be working fine *Steps::: * Downloaded and installed Couldera manager Used Cloudera Manager for ...
    Nov 27, 2012 at 11:44 pm
    Jan 14, 2014 at 2:49 am
  • When trying to install hue-common on Ubuntu 10.04, either through the Cloudera Manager, or using apt-get, the installation fails with an identical error. In particular, the /home partition on that ...
    Andreas PitsillidisAndreas Pitsillidis
    Nov 27, 2012 at 12:44 am
    Feb 5, 2013 at 2:47 pm
  • Hi all I am working through the tutorial of Jeffrey Breen (R + hadoop). I installed cloudera demo vm CDH3u4 and rmr1.3.1. Now I'd like to run a mapreduce job with an airline dataset. However I got ...
    Nov 6, 2012 at 11:45 am
    Nov 11, 2012 at 5:19 pm
  • Hi, I am using Cloudera Manager to get the installation done, have reached upto the point where i was asked to login to the web-browser to proceed with my CDH installation. Since i would just be ...
    Nov 24, 2012 at 12:52 pm
    Jul 29, 2013 at 4:17 pm
  • i'm on the latest release of cloudera distribution. I initiated a distcp call: hadoop distcp -i -update -pr hdfs://SourceNN:8020/Sourcehdfs/path hdfs://DestNN:8020/Dest/hdfs/path/ when I look @ the ...
    Nov 14, 2012 at 9:43 pm
    Jan 8, 2013 at 8:29 pm
  • Hi, While benchmarking our solution, I realized that the WAL is a huge bottleneck. We write one 100Kb value per row. The keys are random therefore well distributed among the region servers. When the ...
    Pierre-Luc BertrandPierre-Luc Bertrand
    Nov 22, 2012 at 6:52 pm
    Dec 3, 2012 at 11:46 pm
  • All with the new FlumeNG has anyone built a log reader that would replace tail? --
    Nov 20, 2012 at 4:38 pm
    Nov 26, 2012 at 4:19 pm
  • Hi everyone I have installed a mapreduce service in cloudera manager however neither hive nor hbase seem to be using it. In th opensource distibution I had to add my direcotry containing amongst ...
    Nicolas MaillardNicolas Maillard
    Nov 21, 2012 at 3:12 pm
    Nov 21, 2012 at 4:52 pm
  • We are running cdh3u3 where we are hitting this. - Inder -- - Inder "You are average of the 5 people you spend the most time with" --
    Inder PallInder Pall
    Nov 16, 2012 at 1:17 pm
    Nov 21, 2012 at 12:42 pm
  • Hello Everyone , i was trying to invoke workflow2 from workflow and could not but when i try it say the following error xxxxxxx@bbbbbbbbb.:/u/users/aaaaaa $ oozieJob.sh -info ...
    Vijay rachalaVijay rachala
    Nov 14, 2012 at 11:05 pm
    Nov 16, 2012 at 3:07 pm
  • We previously had CDH3u3 installed on the 6 node test cluster where we performed a series of HDFS/Pig/HBase test. We kick started/re-imaged the cluster with same OS/setup/etc ... before installing ...
    Mike DobroMike Dobro
    Nov 8, 2012 at 10:09 pm
    Nov 9, 2012 at 5:25 pm
  • All When i run a hive "select * from table" it runs fine. When I introduce a where clause I get 2012-11-25 05:44:30,900 INFO org.apache.hadoop.mapred.JobInProgress: Choosing data-local task ...
    Nov 28, 2012 at 9:06 pm
    Feb 1, 2014 at 7:32 pm
  • Hi I am following Path B Installation steps for Cloudera Manager. In step 2 I am unable to install cloudera-manager-server and cloudera-manager-daemons packages. Here is how I added the cloudera ...
    Pankaj GuptaPankaj Gupta
    Nov 19, 2012 at 3:34 am
    Oct 29, 2013 at 8:05 am
  • Hi, I am trying to start a Secure standalone Zookeeper server instance. I have followed the steps mentioned at: https://ccp.cloudera.com/display/CDH4DOC/ZooKeeper+Security+Configuration I am not able ...
    Subroto SanyalSubroto Sanyal
    Nov 13, 2012 at 10:51 am
    Dec 14, 2012 at 6:13 pm
  • Ben, What's the query you are issuing? Can you try a different query (say with group by) but no where clause. select * doesn't launch a MapReduce job. Group by/where clause does. Mark --
    Mark GroverMark Grover
    Nov 29, 2012 at 8:14 pm
    Dec 3, 2012 at 2:23 pm
  • I'm using flume-ng I have one agent getting some tweets according to keyword (I used Analyzing Twitter Data with Hadoop example flume part), and I want to create second agent working with diffirent ...
    Nov 12, 2012 at 3:04 pm
    Nov 12, 2012 at 4:34 pm
  • Hi, For some reason, I need to move the name server and the secondary name server in one of our production clusters to other hosts. My plan is as follows. 1) Add the two new nodes to the cluster 2) ...
    John FangJohn Fang
    Nov 12, 2012 at 3:32 am
    Nov 12, 2012 at 3:05 pm
  • I am not able to use Oozie on a standard VM for cdh4.1.1 from Cloudera. If I go from Hue to the Oozie screen (http://localhost:8888/oozie/) I get the error <urlopen error [Errno 111] ECONNREFUSED If ...
    Nov 28, 2012 at 1:05 pm
    Dec 4, 2012 at 8:18 pm
  • Hi, We had HDFS block size defaulted for some reason at 64Mb on our cluster. When we realized it, we changed it to 128Mb. I dropped all hbase tables and recreated them. They were still at 64Mb. I ...
    Pierre-Luc BertrandPierre-Luc Bertrand
    Nov 22, 2012 at 6:36 pm
    Nov 30, 2012 at 4:37 pm
  • Hi, I have recently noticed that FS image file generated on the NameNode is not updated, I have downloaded it on 2 consecutive days using this URL ...
    Wojciech LangiewiczWojciech Langiewicz
    Nov 28, 2012 at 8:51 am
    Nov 30, 2012 at 10:13 am
  • Actually they don't appear neither in live nodes nor in dead nodes I just configured 10 nodes via cloudera manager, and the only node that appear in the list is the same as the Namenode There is some ...
    Gabriel DinizGabriel Diniz
    Nov 24, 2012 at 5:42 pm
    Nov 28, 2012 at 6:49 pm
  • Hi, I installed java jdk1.6.0_31 using below step : 1. Download jdk-6u31-linux-x64.bin from Oracle website 2. chmod +x jdk-6u31-linux-x64.bin 3. ./jdk-6u31-linux-x64.bin 4. mkdir /usr/lib/jvm 5. mv ...
    Martinus MartinusMartinus Martinus
    Nov 12, 2012 at 11:26 am
    Nov 21, 2012 at 6:32 pm
  • Hi guys, i was plan to transfer the log file from Apache server to hdfs.i make a configuration file based on this website http://flume.apache.org/FlumeUserGuide.html. when i was started to run this ...
    Nov 19, 2012 at 8:07 am
    Nov 20, 2012 at 3:21 pm
  • We're doing testing with CDH4.1.1 on a 32 node cluster. ( Quorum based Storage with 3 journal nodes, and 'shell(/bin/true)' for fencing) In one of our tests we 'kill -STOP' the active NN process ...
    Ron BuckleyRon Buckley
    Nov 8, 2012 at 1:27 pm
    Nov 20, 2012 at 1:27 am
  • I've spent quite a bit of time setting up and configuring our CDH4.1.1 cluster to the point where I figured I'd break down and ask here before sinking yet more hours into trying to troubleshoot ...
    Chris IngrassiaChris Ingrassia
    Nov 7, 2012 at 8:17 pm
    Nov 9, 2012 at 3:46 pm
  • Hello, I have a cluster running Hadoop cdh4.0. Distcp was running normally untill for the past two days. I configured s3cmd in the node and tried to access the s3 using command line it worked. Don't ...
    Nov 8, 2012 at 12:09 am
    Nov 8, 2012 at 2:12 am
  • Hi all, kindly in my lap i installed the cdh4 with two HA namenode implemented with QJ method and 5 zookeper *ensemble + 4 datanode and task tracker* * * *now i need to go install hbase with two ...
    Muhamed GhareebMuhamed Ghareeb
    Nov 6, 2012 at 1:53 pm
    Nov 6, 2012 at 5:38 pm
  • how do i assure in both cdh3 and cdh4 that all the jars that are included in my jar (so in lib folder) are on the classpath before hadoop's jar both locally and on tasktrackers when i use "hadoop ...
    Nov 3, 2012 at 9:04 pm
    Nov 4, 2012 at 3:00 pm
  • Hello. We have a problem with Flume + Hive realtime data collecting and analytics. Flume collect data and write files to HDFS folder. This folder is data location for Hive table. Flume roll file ...
    Nov 9, 2012 at 6:50 am
    Jan 20, 2013 at 3:46 pm
  • Is there a way to set the user run context when executing a pig, hive, or mapreduce job within hadoop? The only way so far I've found is to use su when running the command to assume another user ...
    Nov 30, 2012 at 9:33 pm
    Nov 30, 2012 at 10:49 pm
  • hi, I have two cluster, one of them test and another is prod and they is located diffirent network area I want to move my hdfs data test to prod. is there any way to do this operation? thanks. --
    Enes yücerEnes yücer
    Nov 27, 2012 at 9:19 am
    Nov 27, 2012 at 4:43 pm
  • Hi all, My cluster has a ton of small files but does this explain why my NN heap is constantly full? I only have ~4million total objects but the namenode heap is constantly full and our monitoring is ...
    Harold wooHarold woo
    Nov 7, 2012 at 12:31 am
    Nov 9, 2012 at 6:54 am
  • This is CentOS 5.8. I realize most people probably use the CentOS 6 packages, but there is a CentOS 5 CDH4 package so that's what I'm trying to use. The problem is, yum can't even *uninstall* CDH3, ...
    Keith WileyKeith Wiley
    Nov 7, 2012 at 7:12 pm
    Nov 7, 2012 at 8:51 pm
  • I am on CDH4.1 using Hive against HBase. Regular Table scan queries go thru fine, any MR type of queries fail with a NPE as shown below. hive select distinct (user) from hbase_tweets; Total MapReduce ...
    Seshu AdunuthulaSeshu Adunuthula
    Nov 2, 2012 at 5:50 pm
    Nov 5, 2012 at 10:22 pm
  • iam running flume-ng agent but hasit@hasit:/etc/flume-ng/conf$ flume dump console No command 'flume' found, did you mean: Command 'dlume' from package 'dlume' (universe) flume: command not found --
    Varaprasad yadavVaraprasad yadav
    Nov 10, 2012 at 5:26 pm
    Jan 31, 2013 at 10:41 am
  • Hi, I've noticed a higher DNS load since enabling HA on our cluster... Doing some packet dumps there appears to be a lookup to nameservice1.example.com from any node/client carrying out HDFS ...
    James HogarthJames Hogarth
    Nov 29, 2012 at 1:42 pm
    Dec 17, 2012 at 7:55 pm
  • We have CDH4 running nicely on CentOS but now I'm trying to install CDH4 on Fedora via yum repository and getting Python version mismatch errors. Cloudera-manager and hue are expecting Python 2.6 ...
    Danny D'AmoursDanny D'Amours
    Nov 27, 2012 at 8:37 pm
    Nov 29, 2012 at 6:55 am
  • Hi, We have been testing CDH4.1 batch write performance and have found that it is almost an order of magnitude slower than existing clusters on CDH3. We have a similar configuration in terms of ...
    Mark GreeneMark Greene
    Nov 28, 2012 at 8:47 pm
    Nov 28, 2012 at 9:48 pm
  • Hi, We are facing the problem while connecting Impala demon with hive. Please suggest any solution. FYI, we are using the below configuration CDH 4.1.1 Impala 0.1 --
    Mohammed NaseeruddinMohammed Naseeruddin
    Nov 26, 2012 at 5:04 pm
    Nov 27, 2012 at 10:59 am
Group Navigation
period‹ prev | Nov 2012 | next ›
Group Overview
groupcdh-user @

177 users for November 2012

Benjamin Cuthbert: 39 posts Harsh J: 35 posts Nicolas Maillard: 34 posts Joey Echeverria: 32 posts Ben: 25 posts Mukhtaj Khan: 24 posts Mark Grover: 21 posts Kevin O'dell: 19 posts Marcin Smialek: 19 posts Andy Isaacson: 18 posts Brock Noland: 17 posts Srinivasan Ramalingam: 16 posts Todd Lipcon: 16 posts Roman Shaposhnik: 14 posts Aaron T. Myers: 13 posts Enes yücer: 13 posts Pierre-Luc Bertrand: 12 posts Alejandro Abdelnur: 11 posts Keith Wiley: 11 posts Robert Kanter: 10 posts
show more