Member since
01-08-2018
133
Posts
31
Kudos Received
21
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
17285 | 07-18-2018 01:29 AM | |
3096 | 06-26-2018 06:21 AM | |
5245 | 06-26-2018 04:33 AM | |
2706 | 06-21-2018 07:48 AM | |
2231 | 05-04-2018 04:04 AM |
05-04-2018
04:04 AM
2 Kudos
Just select "None" in Sentry service section of Kafka. You don't have to delete rules. The rules are stored in sentry and since kafka will not ask, rules are useless.
... View more
05-04-2018
02:27 AM
1 Kudo
You are using sqoop1. Sqoop1 is not a service, it is a tool, that submits the job to YARN. So, apart from your stdout and YARN logs, there are no sqoop logs. Number of mappers (-m 4) means that your job will open 4 connections to your database. If there is no indication in logs about out of memory or illegal value on a column (I mean yarn logs), then you should check that your DB can accept 4 concurrent connections.
... View more
05-03-2018
11:55 PM
According to Cloudera "https://www.cloudera.com/documentation/enterprise/release-notes/topics/rn_consolidated_pcm.html#concept_ihg_vf4_j1b" JDK 8
All JDK 8 updates, from the minimum required version, are supported in Cloudera Manager/CDH 5.3 and higher unless specifically excluded. Updates above the minimum that are not listed are supported but not tested. So, as about your question, 8u171 is supported but not tested. Not tested means that you may be the first which will face some issues. In such case, were issue is due to a bug in JDK version, Cloudera will add this version as "excluded". If you want to be safe, then use the latest tested, mentioned by @csguna
... View more
04-27-2018
12:18 AM
There is another topic that may be related to your issue http://community.cloudera.com/t5/Interactive-Short-cycle-SQL/After-upgrading-to-cdh-5-14-2-Impala-daemon-stopped-suddenly/m-p/66472#M4357 In a few words Impala 2.11 has an issue that causes minidumps and a ticket was opened https://issues.apache.org/jira/browse/IMPALA-6882 Can you check if you have the same problem? Does your cpu has the "popcnt" flag?
... View more
04-26-2018
11:55 PM
1 Kudo
There is no need to have a separate area only for oozie workflows. I understand that we were used to it. But I think the new interface is better, once you get familiar with. So, if you go to documents and select the one you want to copy, press the three vertical dots icon and the following menu will appear. If you want to see only the oozie workflows, you can customize your view. On the left side of search button, there is a drop down (by default is "All") menu where you can select only oozie workflows to be displayed.
... View more
04-18-2018
06:11 AM
There should be no impact. It is the same private key. You just encrypt it with a password.
... View more
04-18-2018
02:42 AM
You can add a password into your private key file. Suppose that you private key file is test.pem. Its contents should be like: -----BEGIN PRIVATE KEY----- . . . -----END PRIVATE KEY----- or -----BEGIN RSA PRIVATE KEY----- . . . -----END RSA PRIVATE KEY----- Run the following command $ openssl rsa -des3 -in test.pem -out test1.pem -passout pass:test This command will create the test1.pem file which is protected by password. Its contents will be similar to : -----BEGIN RSA PRIVATE KEY----- Proc-Type: 4,ENCRYPTED DEK-Info: DES-EDE3-CBC,3716DAF995B742A4 . . . -----END RSA PRIVATE KEY-----
... View more
04-18-2018
01:57 AM
1 Kudo
No you don't have to stop all Management Services. Only Navigator Metadata Server is enough, but you can stop also Navigator Audit server if you feel more safe.
... View more
04-18-2018
01:54 AM
If you delete the log files that end with a number (e.g. mgmt-cmf-mgmt-NAVIGATORMETASERVER-my.node.com.log.out.1), then you will have no issue. Don't delete the ones with ".log" suffix, because they are open for writing by services. Suppose that you have Max Logs Size = 200MB Maximum Log File Backups = 10 That means that each time your .log file reaches 200MB will be rolled, for ten times in total (will be copied to .log.1, .log.2,.....,.log.10). So in total, the log files of a service can occupy up to ((10* 200MB) + 200MB)=2200MB. Depending on your disk size you can reduce only one or both of them, depending on what is better for you. E.g. if you don't want so big files you can set "Max Log Size"=100MB, that means ((10*100MB) + 100MB) = 1100 MB. So just reducing this parameter into half, you save 1100MB of disk space. The same applies to all services.
... View more
04-17-2018
10:01 AM
Use the example I wrote above. It will change the owner of /inputnew directory in hdfs to "cloudera" sudo -u hdfs hdfs dfs -chown cloudera /inputnew
... View more