We do not currently recommend the use of StorageBasedAuthorizationProvider. While Sentry's initial setup (esp. with HDFS ACL sync enabled) may seem a little involved, note that its much simpler than ending up in a longer term situation of managing several HDFS paths and keeping them controlled manually. Currently that fix is not in scope of a backport, since this plugin is not supported for use in a CDH environment, but it may be added in future (such as if/when a rebase occurs). ... View more

Unless "hdfs groups x" returns both "A" and "B" in its results, HDFS/HBase/etc. would not be aware of such a relationship. If you use SSSD+LDAP to resolve groups for your OS, you can request it to query a certain level of nesting. See ldap_group_nesting_level under http://linux.die.net/man/5/sssd-ldap. ... View more

> - is the average load in byte?/kb? > - the average is done on each day/ week? It is the average of number of regions hosted by each RS. Its computed when you run the status command. > - why is the average different than when I do only status and the same on status 'summary' and status 'replication' ? Am not sure I follow. Could you post the observed difference? > - what is the meaning of the 'aggregate load' indicator? The load is aggregated from the number of requests per second ( requestsPerSecond) > - does the compactionProgressPct correesponds to major_compact ? Yes, and it measures the progress for a specific major compacting region at the point of time of running the status command. > - What is the meaning of: totalCompactingKVs / currentCompactedKVs Compactions cause rewrites of KV pairs inside HFiles, into new (fewer/single) HFiles. These numbers track the total KVs participating in the tracked compaction, and the count of how many that have completed persistence so far. ... View more

You can get your live RegionServer IDs with startcodes included via the HBase Shell command: status 'simple' An output line from this, such as the below: host.cloudera.com:60020 1455726247381 Can then be converted into the right format: host.cloudera.com,22101,1455726247381 ... View more

The "admin" user is something usually used within Hue (it could of course be a valid user in your environment, but this is the only assumption I can draw). HS2 cleans the temporary elements if the session holding the query that created it, has terminated. With Hue, especially on versions prior to CDH 5.2.0, you may have a situation where the admin user's sessions have never been closed/terminated, and the HS2 continues to hold references of the queries that user ran in past, whereas the other usernames are likely ending their Hue backed sessions correctly (depends on how they're working over Hue). If you have CDH 5.2.0 or above, consider setting the various idle server-side timeouts under CM -> Hive -> Configuration (search "idle"). ... View more