Member since
07-07-2020
15
Posts
3
Kudos Received
4
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
3348 | 12-14-2020 06:42 AM | |
1144 | 11-06-2020 04:57 AM | |
10836 | 08-31-2020 05:14 AM | |
3050 | 07-20-2020 11:27 PM |
02-11-2021
11:22 PM
Druid is not part of CDH 6.2.1 packaging. You have to install it manually or using Cloudera Manager by using Cloudera Manager Extensions (creating your own parcel and CSD).
... View more
02-11-2021
11:10 PM
Since you are using Ambari, you can you can try to use Rebalance HDFS action, or directly the Hadoop Balancer tool.
... View more
02-11-2021
10:58 PM
Since January 01 2021, CDH 5 and 6 are behind paywall. You can download CDP trial version.
... View more
02-11-2021
10:54 PM
Yes, you can set up a local mirror for offline installation. It is possible with CDP version and with older CDH 5 and 6 version. But even theses legacy versions are behing paywall now.
... View more
12-14-2020
06:50 AM
NiFi don't need to be on the same machine than HDFS. But the PutHDFS processor need informations to connect to HDFS : hdfs-site.xml and core-site.xml. And for sure, network have to be configured.
... View more
12-14-2020
06:42 AM
1 Kudo
Try to use this link : http://archive.cloudera.com/csa/1.0.0.0/parcels/ I think you need credential and cloudera 7 to use latest versions of CSA. I'm using a trial version on with CM 6.2 and CDH 6.2.1 You can get it from this link https://www.cloudera.com/downloads/cdf/csa-trial.html
... View more
12-13-2020
04:17 AM
Do you have the CSD ? You also need it to manage the service. https://docs.cloudera.com/csa/1.2.0/installation/topics/csa-installing-parcel.html
... View more
11-13-2020
08:49 AM
Hello, You can use DetectDuplicate processor and only route non-duplicate to HDFS. But using the GETSFTP processor should work fine as you configured it to run on primary node only. What errors were you facing back then ?
... View more
11-06-2020
04:57 AM
Hello, I'm not sure to understand your problem, but i'll try to answer to "Files ingested into HDFS but I don't see Files in HDFS". When you process a flowfile with a processor, the processor will route the flowfile to relationship according with what it managed to do with the flowfile. The PutHDFS processor have two relationship : success and failure. If the processor managed to put the flowfile to HDFS, the flowfile is routed to the relationship success and you can continue to process it. This unless you check automatically terminate the relationship success, then NiFi won't do anything more with the flowfile. Same for the failure relationship. Here you automatically terminate both success and failure relationship. For success relationship this is ok, because your file was put to HDFS. But for failure relationship, the file is not put to HDFS and you don't handle the failure. Just route the failure relationship to any other processor, you don't even need to use it at this point, you just want to see if flowfile are route to the failure relationship, and so if the queue corresponding to this relationship is filled. You can also see the failure if a red error box appear in the top right corner of the processor, but previous method if to understand the concept of relationship routes. If so, the flowfile are not put to HDFS, probably because of a bad configuration of the PutHDFS processor. In this case, i'll be happy to help you with this configuration (properties tab).
... View more
09-17-2020
11:54 PM
1 Kudo
Hello, one way to do it is to use the ReplaceText processor to add a line break after each object, and so having one line for each object in your FlowFile. To do so just replace "}{" by "} -linebreak- {". Note that you have to escape brackets in your Search Value : \}\{ And that you have to use Shift+Enter in your Replacement Value to add the line break : }
{ Then just use the SplitText processor with Line Count Split to 1 to split your input flow file into one flow file for each line. Hope it helps 🙂
... View more