@uri ben-ari Please try the curl command with double quotes (escaping the double quotes needed as part of curl rrequest) curl -H "X-Requested-By: ambari" -u admin:admin -X PUT -d "{\"RequestInfo\":{\"query\":\"ServiceComponentInfo/component_name.in(${LIST})\"},\"ServiceComponentInfo\":{\"recovery_enabled\":\"true\"}}" http://localhost:8080/api/v1/clusters/HDP01/components? ... View more

Here you have no problem in the log. It's only a warning. And this is because you have another Spark application running and then the port 4040 (which is the default port for the Spark Monitoring UI) is already in use. But then it will try on port 4041 and so on, until it finds a free port, as you can see in the following log, The only problem I can see is that you are specifying an invalid port number (100015 is not a valid port number, I guess you meant 10015). ... View more

you can restart after you set all parameters. but might be difficult to troubleshoot if some of the services fail to start. ... View more

@uri ben-ari To list all the requests use the following API and see what is the maximum request ID (which will be the recent one, Higher the requestID more recent one it is): # curl -sH "X-Requested-By: ambari" -u admin:admin -i http://localhost:8080/api/v1/clusters/Sandbox/requests?fields=Requests/request_status | grep -A 1 id | awk '{print $NF}' > /tmp/requests_details.txt; sed -e '1,2d' -e s'/--//g' -e 's/\n//g' -e 's/"//g' -e '/^$/d' /tmp/requests_details.txt | awk 'NR%2{printf "%s ",$0;next;}1' Example Output: 185, COMPLETED 186, COMPLETED 187, COMPLETED 188, FAILED 189, COMPLETED 190, COMPLETED 191, COMPLETED 193, COMPLETED 195, COMPLETED 196, COMPLETED 197, COMPLETED 198, COMPLETED 199, COMPLETED 200, COMPLETED 201, COMPLETED 202, COMPLETED 203, COMPLETED 204, COMPLETED 206, ABORTED . . Then if you want to see only the IN_PROGRESS requests then you can make the following API call. # curl -sH "X-Requested-By: ambari" -u admin:admin -i http://localhost:8080/api/v1/clusters/Sandbox/requests?fields=Requests/request_status | grep -A 1 id | awk '{print $NF}' > /tmp/requests_details.txt; sed -e '1,2d' -e s'/--//g' -e 's/\n//g' -e 's/"//g' -e '/^$/d' /tmp/requests_details.txt | awk 'NR%2{printf "%s ",$0;next;}1' | grep -i 'IN_PROGRESS' Example Output: # curl -sH "X-Requested-By: ambari" -u admin:admin -i http://localhost:8080/api/v1/clusters/Sandbox/requests?fields=Requests/request_status | grep -A 1 id | awk '{print $NF}' > /tmp/requests_details.txt; sed -e '1,2d' -e s'/--//g' -e 's/\n//g' -e 's/"//g' -e '/^$/d' /tmp/requests_details.txt | awk 'NR%2{printf "%s ",$0;next;}1' | grep -i 'IN_PROGRESS' 207, IN_PROGRESS . ... View more

@uri ben-ari Following should list all the components that has stale config, and need to be restaretd to get the config changes reflected. # curl -i --user admin:admin -i -H 'X-Requested-By: ambari' -X GET http://localhost:8080/api/v1/clusters/Sandbox/host_components?HostRoles/stale_configs=true Here please replace the following: 1. "localhost" : with the Ambari FQDN 2. 8080 : with ambari port. 3. Sandbox : With cluster Name . ... View more

@uri ben-ari The error indicates that you are not passing the correct "zookeeper.connect" property value to the flag "--zookeeper" in the command line. You can get the "zookeeper.connect" value from Ambari UI --> Kafka --> Configs --> Kafka Broker --> "zookeeper.connect" In my case the value was : "amb25103.example.com:2181,amb25102.example.com:2181,amb25101.example.com:2181" So i had to use the following command: Example: # /usr/hdp/2.6.0.3-8/kafka/bin/kafka-preferred-replica-election.sh --zookeeper amb25103.example.com:2181,amb25102.example.com:2181,amb25101.example.com:2181 . ... View more