FAQ

Search Discussions

44 discussions - 162 posts

  • Hi all, My setup: CDH 5 Hadoop 2.3, Hive 0.13, Impala 1.4.0, and I install Impala *without* Cloudera Manager. After create and insert data into a table with parquet format, it works fine for ...
    Zhe LiZhe Li
    Sep 28, 2014 at 9:08 am
    Nov 5, 2014 at 9:45 pm
  • Can you create an Impala external table in Parquet format? With the HDFS block size perhaps rewritten for parquet format. To unsubscribe from this group and stop receiving emails from it, send an ...
    Vidya BalasubramanianVidya Balasubramanian
    Sep 24, 2014 at 12:27 am
    Sep 30, 2014 at 11:52 pm
  • Hi Everyone, I am trying to faster impala query by using HDFS caching. The data files in HDFS are stored as Parquet format. The query performance are almost the same whether the data are cached. I am ...
    David ZhouDavid Zhou
    Sep 9, 2014 at 6:19 am
    Sep 24, 2014 at 10:43 am
  • Hi, We have 21 Data Node Hadoop cluster and with impala v1.4.0-cdh4-INTERNAL. I have created on external table and loaded the dataset into it. which has 3.6 B records, and it takes 607.15s to run ...
    RoyRoy
    Sep 4, 2014 at 2:38 pm
    Sep 8, 2014 at 7:08 pm
  • Hi all, I'm part of the data team in a middle-sized web based company. We collect logdata and convert those once per hour in .lzo format before we put them in a YYYY/MM/DD/HH directory format on ...
    Erik VandeputteErik Vandeputte
    Sep 5, 2014 at 1:19 pm
    Oct 27, 2014 at 8:07 pm
  • I have created following table CREATE EXTERNAL TABLE search_tmp (time_stamp BIGINT,id INTEGER,........., keyword STRING) PARTITIONED BY (year INT, month INT, day INT, hour INT) row format delimited ...
    RoyRoy
    Sep 8, 2014 at 4:47 pm
    Sep 8, 2014 at 5:34 pm
  • Hi, I am trying to connect Tableau to Impala. So far I am able to access impala data/tables from client machine via datanode. but when I tried to do same for Tableau I failed to connect. Anyone can ...
    RoyRoy
    Sep 2, 2014 at 9:05 pm
    Sep 8, 2014 at 4:09 pm
  • create external table airflight_stats_text_lzo ( Year string, Month string, DayofMonth string, DayOfWeek string, DepTime string, CRSDepTime string, ArrTime string, CRSArrTime string, UniqueCarrier ...
    Vidya BalasubramanianVidya Balasubramanian
    Sep 24, 2014 at 12:25 am
    Sep 29, 2014 at 4:20 pm
  • If I use dynamic partitioning and insert into partitioned table - it is 10 times slower than inserting into non partitioned table. Any ideas to make this any faster? CASE 1: Create table x ( c1 ...
    Giri TataGiri Tata
    Sep 11, 2014 at 7:38 pm
    Sep 16, 2014 at 4:24 pm
  • I use Cloudera Manager, there is no option to update the impala log directory. I tried to update /etc/default/impala file, the IMPALA_LOG_DIR entry. But after I restart the service, it's still ...
    Rangga SobiranRangga Sobiran
    Sep 8, 2014 at 10:08 am
    Sep 15, 2014 at 7:28 pm
  • Hello CDH Users, We are pleased to announce the release of the Cloudera JDBC drivers for both Apache Hive and Impala. These drivers include the following: - More complete JDBC API coverage - Easier ...
    Wendy TurnerWendy Turner
    Sep 15, 2014 at 7:48 pm
    Oct 29, 2014 at 7:15 pm
  • Hi Impala users, We're battling some perf issues that I suspect are in part related to a huge catalog (tons of tables with hourly data that never get deleted, too many files per partition, etc. I'd ...
    Sean O'BrienSean O'Brien
    Sep 30, 2014 at 10:41 pm
    Oct 11, 2014 at 10:32 pm
  • Hi Luke, If you allocate memory in Init(), you do need a Serialize() method that frees the allocated memory and returns a serialized StringVal allocated using the StringVal constructor: // Creates a ...
    Matthew JacobsMatthew Jacobs
    Sep 27, 2014 at 7:17 pm
    Sep 30, 2014 at 8:55 pm
  • I'm a beginner of Impala. Please forgive me if the answer is too obvious. Referring to impala tutorial ...
    Susie ZhaoSusie Zhao
    Sep 26, 2014 at 6:11 pm
    Sep 30, 2014 at 6:22 am
  • This is my query profile,and i find thrift_transmit_timer is very long,can any one tell me how can i to reduce the time: Query (id=a340a05de7f0ed35:e57a19da5e47078f): Summary: Session ID ...
    Zhangxin19880228Zhangxin19880228
    Sep 25, 2014 at 7:41 am
    Sep 28, 2014 at 7:09 am
  • Hi, Using impala 1.4, I created a simple table as follow: CREATE TABLE x (c string) stored as avro TBLPROPERTIES ('avro.schema.literal'='{ "name": "my_record", "type": "record", "fields": [ ...
    Stephane DrouinStephane Drouin
    Sep 23, 2014 at 7:24 pm
    Sep 27, 2014 at 3:49 pm
  • create view total as select sum(b) as sum from t1; Does not work, encountered SUM expected IDENTIFIER. -- Abhi Basu To unsubscribe from this group and stop receiving emails from it, send an email to ...
    Abhi BasuAbhi Basu
    Sep 16, 2014 at 5:41 pm
    Sep 16, 2014 at 5:45 pm
  • Currently, I am using CDH 5 and with impala 1.3.0 and need of median function while converting some legacy code. I did see from github that there is approximate median implemented - but i am not sure ...
    Giri TataGiri Tata
    Sep 15, 2014 at 6:56 pm
    Sep 15, 2014 at 7:36 pm
  • mx, For the slow query, would you mind sending us the query and its profile, if possible. Additionally, could you send us the results of the describe (describe formatted gives more information). How ...
    Ishaan JoshiIshaan Joshi
    Sep 8, 2014 at 6:10 pm
    Sep 15, 2014 at 3:29 pm
  • This is more of a developer question than a user question, so I'm not sure this is the right mailing list, but here goes! For an advanced databases class project, I'm planning on doing work to ...
    Michael JohnsonMichael Johnson
    Sep 10, 2014 at 6:21 pm
    Sep 13, 2014 at 4:34 pm
  • Hi, here's is the log message when I starting Impala. E0910 16:43:45.421779 17410 impala-server.cc:208] Could not read the HDFS root directory at hdfs://10.4.17.210:5060. Error was: Failed on local ...
    Zhe LiZhe Li
    Sep 10, 2014 at 9:23 am
    Sep 12, 2014 at 4:20 am
  • I am using CDH 5.1 evaluation VM. I don't see the port open for external access, just for local access how do I change this so I can connect to it from external IP? [root@quickstart cloud]# netstat ...
    Soheil EizadiSoheil Eizadi
    Sep 6, 2014 at 2:01 am
    Sep 8, 2014 at 7:32 pm
  • Hi All, We are facing the issue when we try to open impala through POPEN in python. It is able to connect to the impala daemon but before executing any queries it exits stating "Unable to save ...
    Sumit awkashSumit awkash
    Sep 4, 2014 at 10:29 am
    Sep 30, 2014 at 12:39 am
  • Are there any available for Impala for functions like Chi Squared, or do we need to write an UDF using python libraries? Thanks. To unsubscribe from this group and stop receiving emails from it, send ...
    Abhi BasuAbhi Basu
    Sep 29, 2014 at 10:20 pm
    Sep 30, 2014 at 12:37 am
  • Hi Syed, Thanks for the reply. Got it working when using these options: 1. I'm using CDH 5.1.3, Parcels. 2. Cluster is RHEL6 - 64 Bit and client is the same, as I am submitting the job to Oozie via ...
    Aare PuussaarAare Puussaar
    Sep 27, 2014 at 10:28 am
    Sep 27, 2014 at 9:47 pm
  • Hello CDH and Impala Users, We are pleased to announce the release of version 2.5.12 of the ODBC Driver for Apache Hive and version 2.5.20 of the ODBC driver for Impala. These versions contain bug ...
    Wendy TurnerWendy Turner
    Sep 25, 2014 at 6:56 pm
    Sep 26, 2014 at 8:31 am
  • All, Currently, we are sending the queries to impala daemon on one of the nodes and it seem to work as query coordinator as well as execution node. However - if there are thousands of queries which ...
    Giridhar TGiridhar T
    Sep 16, 2014 at 2:21 pm
    Sep 16, 2014 at 3:47 pm
  • Hi, I'm running Impala with Llama and Yarn. My problem is that llama is taking more containers in each node than containers that had been condifured in yarn-site.xml. *Yarn-site.xml:* <property <name ...
    Albert FranziAlbert Franzi
    Sep 15, 2014 at 2:01 pm
    Sep 16, 2014 at 2:25 pm
  • Hi All, Have anyone tried to build Impala 1.4 CDH 5.1.2 ? The /impala/fe/src/main/java/com/cloudera/impala/util/FSPermissionChecker.java should be ...
    Andrew LeeAndrew Lee
    Sep 8, 2014 at 5:01 pm
    Sep 15, 2014 at 11:45 pm
  • Hi, Is there an equivalent of Hive's recover partitions? I cannot seem to find anything related. Is the concept of keeping your data in directories that contain the partition names (such as ...
    Dimitris TheodorouDimitris Theodorou
    Sep 11, 2014 at 12:35 pm
    Sep 11, 2014 at 5:19 pm
  • Hi, I am using EMR with impala version 1.2.4. I am using a 60 node cluster with r3.xlarge instances. I am triggering several impala queries in parallel continuously from multiple machines. My data ...
    AMAL G JOSEAMAL G JOSE
    Sep 10, 2014 at 11:05 am
    Sep 10, 2014 at 11:57 am
  • I have created following table CREATE EXTERNAL TABLE search_tmp (time_stamp BIGINT,id INTEGER,........., keyword STRING) PARTITIONED BY (year INT, month INT, day INT, hour INT) row format delimited ...
    RoyRoy
    Sep 8, 2014 at 4:06 pm
    Sep 8, 2014 at 4:44 pm
  • Are there plans to support S3 as a valid location for external tables? To unsubscribe from this group and stop receiving emails from it, send an email to impala-user+unsubscribe@cloudera.org.
    Alex LeeAlex Lee
    Sep 4, 2014 at 10:03 pm
    Sep 4, 2014 at 10:22 pm
  • Hi, I am doing many inserts across an impala 1.4.1 cluster. I am seeing periodic crashes on different nodes with the following log messages. I don't know if the errors are related to the crash but I ...
    Luke C.Luke C.
    Sep 30, 2014 at 9:01 pm
    Sep 30, 2014 at 9:01 pm
  • Hi all, I am just a newbie in Impala. I am trying to install Impala (1.4.2) in CDH 5.1.3 but when I am trying to start the impala-server daemon, it is showing like "Impala Server is dead and pid file ...
    Athira ashokAthira ashok
    Sep 30, 2014 at 6:25 am
    Sep 30, 2014 at 6:25 am
  • Is it possible to run Impala on Mesos? Has anyone tried this before? I know there is Llama <http://cloudera.github.io/llama/ for running Impala on YARN. Is there something similar with Mesos? Would ...
    NightNight
    Sep 25, 2014 at 3:16 am
    Sep 25, 2014 at 3:16 am
  • ​Hi, i have some data in pail format, there is a way to use it with impala? May i create some tables over that files? Thanks Camillo​ To unsubscribe from this group and stop receiving emails from it, ...
    SiacoSiaco
    Sep 22, 2014 at 9:50 am
    Sep 22, 2014 at 9:50 am
  • Why does this not work? As far as Cloudera docs, this should work. create table table_partition like table_source partitioned by (fieldname string) stored as parquetfile ; Thanks. To unsubscribe from ...
    Abhi BasuAbhi Basu
    Sep 18, 2014 at 8:46 pm
    Sep 18, 2014 at 8:46 pm
  • I am testing parquet file format and inserting data into parquet file using impala external table.Following is the parameter set that may affect the parquet file size: NUM_NODES: 1 ...
    Vikas AVikas A
    Sep 17, 2014 at 9:43 pm
    Sep 17, 2014 at 9:43 pm
  • I'm seeing very high memory usage (10GB+) with catalogd on a 10-node Impala 1.4.0 cluster co-located with CDH4.6.0. This is despite enabling a 2GB cap at startup via IMPALA_CATALOG_ARGS (mem_limit) ...
    Norbert BurgerNorbert Burger
    Sep 16, 2014 at 4:59 pm
    Sep 16, 2014 at 4:59 pm
  • I am using Impala (1.2.4) on Amazon EMR. This is the Impala that comes with EMR. I am seeing the following in my logs even though I have enabled dfs.datanode.hdfs-blocks-metadata.enabled. Is there ...
    Sabarish SasidharanSabarish Sasidharan
    Sep 15, 2014 at 1:17 pm
    Sep 15, 2014 at 1:17 pm
  • Awesome!!! Thanks, Henry! To unsubscribe from this group and stop receiving emails from it, send an email to impala-user+unsubscribe@cloudera.org.
    Alex BehmAlex Behm
    Sep 13, 2014 at 1:20 am
    Sep 13, 2014 at 1:20 am
  • Hi, I'm trying to run a TPC-H benchmark of Impala at scale factor 1000 (~1TB dataset) on a large number of EC2 instances, but most of the join queries take forever / silently fail. I have run the ...
    Marco SlotMarco Slot
    Sep 12, 2014 at 1:21 pm
    Sep 12, 2014 at 1:21 pm
  • Hi, Anyone knowing more about impala picking up new files automatically from directories, as the files are being streamed in? This would solve a lot of trouble with periodic refreshes on tables etc ...
    Laurens BronwasserLaurens Bronwasser
    Sep 10, 2014 at 9:48 am
    Sep 10, 2014 at 9:48 am
Group Navigation
period‹ prev | Sep 2014 | next ›
Group Overview
groupimpala-user @
categorieshadoop
discussions44
posts162
users68
websitecloudera.com
irc#hadoop

68 users for September 2014

Vidya Balasubramanian: 13 posts Roy: 11 posts Vikas A: 7 posts Zhe Li: 6 posts Albert Franzi: 5 posts Alex Behm: 5 posts Giri Tata: 5 posts Abhi Basu: 4 posts Alan Choi: 4 posts Dimitris Tsirogiannis: 4 posts Lenni Kuff: 4 posts Aare Puussaar: 3 posts David Zhou: 3 posts Ishaan Joshi: 3 posts John Russell: 3 posts Luke C.: 3 posts Matthew Jacobs: 3 posts Sammy Yu: 3 posts Soheil Eizadi: 3 posts Uri Laserson: 3 posts
show more