Member since
01-09-2014
283
Posts
70
Kudos Received
50
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
1714 | 06-19-2019 07:50 AM | |
2731 | 05-01-2019 08:07 AM | |
2780 | 04-10-2019 08:49 AM | |
2697 | 03-20-2019 09:30 AM | |
2363 | 01-23-2019 10:58 AM |
01-06-2017
11:51 AM
Take a look at the preferred leader election tool: https://cwiki.apache.org/confluence/display/KAFKA/Replication+tools This assumes that the desired leaders are listed first in the partition list. If 30 is listed first, it will still be the leader for all those partitions. -pd
... View more
01-03-2017
11:52 PM
i was pointinn to a wrong broker so the zk was unable to find the topics
... View more
09-16-2016
03:57 AM
thank you for your detailed reply. sorry if i am digging this up, but i followed have a similar problem (topic partition 5 referencing a non-existent leader broker) and a replication factor of 1 i think this happened during a move of the topic from my colleagues. anyhow i followed your instructions* and enabled the unclean leader election before restarting kafka as cloudera service via manager. however, i am still receiving this line in the log of the supposedly new leader: Broker xxx received LeaderAndIsrRequest with correlation id 1 from controller 1 epoch 1035 for partition [mytopic,5] but cannot become follower since the new leader [ID] is unavailable. where ID is probably an old broker (i wouldnt know) but clearly not an active broker id (of which i have 3). * with the exception that i did not generate a json file but wrote it myself, since only 1 partition needed id-correction and i entered the broker that already holds the partition data on disk
... View more
08-22-2016
05:43 PM
yes, thanks for the reply! I figured out the same thing earlier today as I went back to the Flume User Guide and started copying and pasting the properties in again... When I reviewed my config initiall, i didn't look before the attribute name to even see I was missing "hdfs". Definitely an ID10T and PEBKAC error. 🙂 Thanks for keeping me honest!
... View more
08-09-2016
10:14 PM
Yes I found the solution. There was one more parcel "CDH5", which I activated first and then I was able to activate Kafka. But one more issue is when I tried to add the Kafka as a service, then I am facing error as "Supervisor retuned fatal. Please check the role log file, stderr or stdout". What is this issue again. Please help me.
... View more
08-09-2016
11:02 AM
1 Kudo
This is actually a logging regression issue, that was introduced in CDH5.7.0, but fixed in CDH5.7.1, and the messages are benign, but can fill up the logs quickly. The recommendation would be to upgrade to CDH5.7.1 or higher where this logging regression was fixed. Alternatively, you can suppress these INFO messages by adding the following to the"Solr Server Logging Advanced Configuration Snippet (Safety Valve) ": log4j.logger.org.apache.solr.servlet.SolrDispatchFilter=WARN -pd
... View more
07-14-2016
02:02 PM
1 Kudo
If you are using the exec source to tail a file, keep in mind that it is not a very reliable source. I would suggest using the taildir source (https://archive.cloudera.com/cdh5/cdh/5/flume-ng/FlumeUserGuide.html#taildir-source) to tail files reliably. -pd
... View more
06-24-2016
08:34 AM
Your understanding is correct. You either need to ensure flume can write to that directory, or create a directory that flume owns and can write to. -pd
... View more
06-13-2016
10:20 PM
But how would he fix it, if he did need it?? I am experiencing the same thing.
... View more
06-07-2016
04:23 PM
The flume http source is for creating a REST API that you can post data to from the upstream sender. If you are looking for a source that will consume from SQL server via accessing the SQL server API, you'll need to write a custom source for that, or possibly try this: https://github.com/keedio/flume-ng-sql-source Additionally, if you don't need real time processing, you may want to consider using sqoop to import data via batch processing, and it can handle incremental updates. -pd
... View more