> On March 14, 2015, 1:24 a.m., Robert Levas wrote: > > ambari-server/src/main/resources/common-services/HIVE/0.12.0.2.0/kerberos.json, > > line 24 > > <https://reviews.apache.org/r/32060/diff/1/?file=894702#file894702line24> > > > > ${clusterHostInfo/hive_metastore_host} may resolved to a > > comma-delimited list of hosts, so the resulting string may be > > ``` > > ...thrift://host1,host2,host3:9083,... > > ``` > > which is probably not what you want.
New patch takes care of this case in the FE. Although in long term we should think on handling this on the server side. - Jaimin ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/32060/#review76471 ----------------------------------------------------------- On March 16, 2015, 7:36 a.m., Jaimin Jetly wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/32060/ > ----------------------------------------------------------- > > (Updated March 16, 2015, 7:36 a.m.) > > > Review request for Ambari, Robert Levas, Srimanth Gunturi, and Yusaku Sako. > > > Bugs: AMBARI-10068 > https://issues.apache.org/jira/browse/AMBARI-10068 > > > Repository: ambari > > > Description > ------- > > - Singlenode cluster c6402 with hdfs, yarn,mr,hive,tez,zk,mertics, kerberos > enabled. > - After all is running fine, add host c6403 with just DN, NM and clients. > - After add host says I need to restart hive > - I go look at the configs that have changed, and seems Ambari modified the > webhcat-site, templeton.hive.properties. > - It's saying the metastore moved to c6403, which it has not. It should not > have modified the webhcat-site. > > > Diffs > ----- > > > ambari-server/src/main/resources/common-services/HIVE/0.12.0.2.0/kerberos.json > ad5ee3e > ambari-web/app/controllers/main/admin/kerberos/step4_controller.js 064eafc > > Diff: https://reviews.apache.org/r/32060/diff/ > > > Testing > ------- > > tested the patch manually by going through STR on a cluster > > > Thanks, > > Jaimin Jetly > >
