----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/53629/#review155558 -----------------------------------------------------------
ambari-server/src/main/resources/common-services/HIVE/0.12.0.2.0/package/scripts/webhcat.py <https://reviews.apache.org/r/53629/#comment225542> Can you try to do "kdestroy" for webhcat user. And than start Webhcat if that works +1. - Andrew Onischuk On Nov. 10, 2016, 6:49 a.m., Shi Wang wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/53629/ > ----------------------------------------------------------- > > (Updated Nov. 10, 2016, 6:49 a.m.) > > > Review request for Ambari, Andrew Onischuk, Laszlo Puskas, and Robert Levas. > > > Bugs: AMBARI-18836 > https://issues.apache.org/jira/browse/AMBARI-18836 > > > Repository: ambari > > > Description > ------- > > The Smoke and “Headless” Service users are used by Ambari to perform service > “smoke” checks and run alert health checks. > The permission for hdfs.headless.keytab is 440. But it will cause security > concern to allow other service user in hadoop group to kinit hdfs headless > principal using hdfs.headless.keytab. In this way, other service user could > "pretend" to be hdfs user and be granted hdfs user's authorities. > > > Diffs > ----- > > > ambari-server/src/main/resources/common-services/HDFS/2.1.0.2.0/kerberos.json > 3d6e25c > > ambari-server/src/main/resources/common-services/HIVE/0.12.0.2.0/package/scripts/webhcat.py > 816b6af > > Diff: https://reviews.apache.org/r/53629/diff/ > > > Testing > ------- > > Hi Robert, > What will be your opinion on this? > Is it necessary to set 440 permission to hdfs headless keytab? The kinit hdfs > headless principal operation for webhcat seems useless, I kdestroy the hdfs > ticket for hcat user and hive service check/start/stop all works fine. Is > there any other tests I should try? Thanks. > > > Thanks, > > Shi Wang > >