Member since
10-01-2018
802
Posts
143
Kudos Received
130
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
2332 | 04-15-2022 09:39 AM | |
1798 | 03-16-2022 06:22 AM | |
5265 | 03-02-2022 09:44 PM | |
2128 | 03-02-2022 08:40 PM | |
1331 | 01-05-2022 07:01 AM |
08-10-2020
02:24 AM
@bermo This can be your best bet regarding how to help with installation: https://community.cloudera.com/t5/Community-Articles/How-To-Path-B-Installation-of-Cloudera-Manager/ta-p/44598 Also for trial you can visit the page: https://www.cloudera.com/downloads/cdp-data-center-trial.html Here are some free tutorials to get you familiar with Cloudera: https://www.cloudera.com/tutorials.html
... View more
08-10-2020
02:18 AM
1 Kudo
@HanzalaShaikh You are most probably hitting a known bug with Java recent versions (OpenJDK 1.8u242 or JDK 11.0.6). TSB-394. To resolve this issue, take the following action on all impacted nodes solved as appropriate for the environment. Edit java.security file located in the active JDK on the clusters. Add or alter sun.security.krb5.disableReferrals parameter, to ensure that the following is set to true: sun.security.krb5.disableReferrals=true
... View more
08-10-2020
02:06 AM
@Shri23 Error 32 usually means, the referenced object does not exist. I.e. You entered a bad DN value for something that needed a correct DN value. Please look at your AD and confirm that the user exist in the path: ou=hadoop-ad,DC=hadoopsecurity,DC=local userPrincipalName=cm@HADOOPSECURITY.LOCAL Also as a side note the service account should have create, modify and delete access in AD as well.
... View more
08-01-2020
10:16 AM
@joeyfrye1 Both looks fine for me. Can you check again. What issue you are facing?
... View more
08-01-2020
09:08 AM
1 Kudo
@selvaprabhu_k The same issue has been answered in below thread, please have a look. https://community.cloudera.com/t5/Support-Questions/ImportError-No-module-named-numpy/m-p/90427#M21658 https://community.cloudera.com/t5/Support-Questions/Jupyter-notebook-gt-ImportError-No-module-named-numpy/m-p/300687/highlight/false#M220357
... View more
08-01-2020
09:07 AM
@Marek The solution your are referring can be implemented like below in Oozie service Configuration: <property>
<name>oozie.launcher.mapreduce.map.env</name>
<value>PYSPARK_DRIVER_PYTHON=/opt/cloudera/parcels/Anaconda/bin/python,PYSPARK_PYTHON=/opt/cloudera/parcels/Anaconda/bin/python</value>
</property> Though installing numpy packages should resolve the issue.
... View more
08-01-2020
08:39 AM
@Marek I tried my best to dug this issue for you. First of all I want to clear that this is not CDSW issue. This is the HDFS client configuration issue. I found some discrepancy: In the hdfs-site.xml file notice the dfs.ha.namenodes.[nameservice ID] section where you have declared the NN IDs as "namenode43,namenode57" which is causing the issue as indicated in the message. <property>
<name>dfs.ha.namenodes.namenodeHA</name>
<value>namenode43,namenode57</value>
</property> "Namenode for namenodeHA remains unresolved for ID namenode57. Check your hdfs-site.xml file to ensure namenodes are configured properly." So you should declare the IP or FQDN here because the session is not resolving this short name as there is no entry in the "/etc/hosts" file. Your "/etc/hosts" file is consist only below format: ---- RUN cat /etc/hosts 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1 localhost localhost.localdomain localhost6 localhost6.localdomain6 IP address blc-control-02.novalocal IP address blc-control-01.novalocal 👆This format is not recommended the Network names should be configured as below: 1.1.1.1 foo-1.example.com foo-1
2.2.2.2 foo-2.example.com foo-2
3.3.3.3 foo-3.example.com foo-3
4.4.4.4 foo-4.example.com foo-4 In the Ideal condition you should met below condition: Your CDSW session should resolve the localhost as: !nslookup localhost Server: IP address
Address: IP address#53
Name: localhost
Address: 127.0.0.1
Name: localhost
Address: ::1 Your "/etc/hosts" file should be correct as per the Recommended Network names Configuration Your hdfs-site.xml should use FQDN or IP. NOTE: Having the short name entries of "namenode43,namenode57" in "/etc/hosts" file might work with your existing hdfs-site.xml but I would recommend to do that with FQDN way to avoid other issue. Please modify and let me know how this goes, I am waiting for this issue to be resolved 🙂🤞
... View more
07-28-2020
02:05 AM
@Marek The only thing I can think of form the logs is issue with client Configuration. What’s the hdfs-site.xml saying can we have a copy. What if you ran /opt/cloudera/parcels/CDH/bin/hdfs dfs -ls / form session? What engine version is running? Some engines did not have the hdfs client installed. Run below command form Session and provide the output echo $PATH which hdfs I am wondering if the path /opt/cloudera/parcels/CDH/bin is missing form default paths set in session then you have to export the path /opt/cloudera/parcels/CDH/bin manually and run the hdfs listing again to see if that works.
... View more
07-28-2020
12:15 AM
2 Kudos
@HanzalaShaikh This seems a very old cosmetic issue in CM. What version you are using? It appears in older versions we had some early bugs caused by the addition of a new jar file in the classpath directory (/usr/share/cmf/schema/../lib/* ) used by the scm_prepare_db.sh script. This new jar contained a log4j.properties file that was not configured to use the log4j appender A, and overwrote the original properties from the proper log4j.properties file in the original jar file.
... View more
07-26-2020
05:25 AM
3 Kudos
@emeric Apologise for inconvenience the issue is expected as there is not bits for Ubuntu in the archive. The CDP Data Center is version supported on below operating System. RHEL/CentOS/Oracle 7.6, 7.7 Debian Not Supported I would like you to try CDP on supported platform and follow below doc for installation. https://docs.cloudera.com/cloudera-manager/7.1.1/installation/topics/cdpdc-run-cm-server-installer.html
... View more