Sorry, my bad. I meant to ask what "get" returns?
/var/lib/ambari-server/resources/scripts/configs.sh -u admin -p admin@DEV get mstd01.corp.com<http://mstd01.corp.com/> BDP2_DEV hdfs-site Similarly, what does http://mstd01.corp.com:8080/api/v1/clusters/BDP2_DEV/configurations?type=hdfs-site return? Also, I think the delete may have some issue - what version of Ambari are you using/? This is what I see ... [root@c6402 vagrant]# /var/lib/ambari-server/resources/scripts/configs.sh -u admin -p admin delete localhost c1 hdfs-site dfs.namenode.rpc-address USERID=admin PASSWORD=admin ########## Performing 'delete' dfs.namenode.rpc-address: on (Site:hdfs-site, Tag:version1) ########## Config found. Skipping origin value ########## PUTting json into: doSet_version1446520420271646783.json { "resources" : [ { "href" : "http://localhost:8080/api/v1/clusters/c1/configurations/service_config_versions?service_name=HDFS&service_config_version=2";, "configurations" : [ { "clusterName" : "c1", "stackId" : { "stackName" : "HDP", "stackVersion" : "2.3", "stackId" : "HDP-2.3" }, "type" : "hdfs-site", "versionTag" : "version1446520420271646783", "version" : 2, "serviceConfigVersions" : null, "configs" : { "dfs.replication" : "3", "dfs.namenode.audit.log.async" : "true", ... ________________________________ From: Shaik M <[email protected]> Sent: Monday, November 02, 2015 6:05 PM To: [email protected] Subject: Re: Rebalance HDFS - Issue I have restart HDFS after executing this. The output of this command: [root@mstd01~]# /var/lib/ambari-server/resources/scripts/configs.sh -u admin -p admin@DEV delete mstd01.corp.com<http://mstd01.corp.com> BDP2_DEV hdfs-site “dfs.namenode.rpc-address” USERID=admin PASSWORD=admin@DEV ########## Performing 'delete' “dfs.namenode.rpc-address”: on (Site:hdfs-site, Tag:version1446466158434) ########## PUTting json into: doSet_version1446466579436571830.json ########## NEW Site:hdfs-site, Tag:version1446466158434 But still that property is exists in Ambari->HDFS->Config. On 2 November 2015 at 21:01, Sumit Mohanty <[email protected]<mailto:[email protected]>> wrote: What do you get when you use this - /var/lib/ambari-server/resources/scripts/configs.sh -u admin -p admin@DEV delete mstd01.corp.com<http://mstd01.corp.com> BDP2_DEV hdfs-site Did you restart after the delete? ________________________________ From: Shaik M <[email protected]<mailto:[email protected]>> Sent: Monday, November 02, 2015 4:23 AM To: [email protected]<mailto:[email protected]> Subject: Re: Rebalance HDFS - Issue Hi Sumit, Thank you for your assistance. I have tried to remove "dfs.namenode.rpc-address" property using below command, it went successfully, but still that property exists. [root@mstd01~]# sudo /var/lib/ambari-server/resources/scripts/configs.sh -u admin -p admin@DEV delete mstd01.corp.com<http://mstd01.corp.com> BDP2_DEV hdfs-site “dfs.namenode.rpc-address” [sudo] password for root: USERID=admin PASSWORD=admin@DEV ########## Performing 'delete' “dfs.namenode.rpc-address”: on (Site:hdfs-site, Tag:version1446466158434) ########## PUTting json into: doSet_version1446466579436571830.json ########## NEW Site:hdfs-site, Tag:version1446466158434 Please help me to fix this issue. https://issues.apache.org/jira/browse/AMBARI-13373 Thanks, Shaik On 2 November 2015 at 19:04, Sumit Mohanty <[email protected]<mailto:[email protected]>> wrote: Likely "dfs.namenode.rpc-address" in hdfs-site has a wrong value. If it is, you can delete it - see https://cwiki.apache.org/confluence/display/AMBARI/Modify+configurations (section Edit Configurations with configs.sh) ________________________________ From: Shaik M <[email protected]<mailto:[email protected]>> Sent: Sunday, November 01, 2015 8:17 PM To: [email protected]<mailto:[email protected]> Subject: Rebalance HDFS - Issue Hi, I am trying to run HDFS balancer from Ambari 2.1.2 with HDP 2.3.2, it's failing with following exception. 15/11/02 03:07:00 INFO block.BlockTokenSecretManager: Setting block keys 15/11/02 03:07:00 INFO balancer.KeyManager: Update block keys every 2hrs, 30mins, 0sec java.io.IOException: Another Balancer is running.. Exiting ... Nov 2, 2015 3:07:00 AM Balancing took 1.935 seconds I have verified respective Hadoop NN, there is no balancer on NameNode. Could you please help use to fix this issue. Thanks, Shaik M
