Member since
09-29-2015
94
Posts
117
Kudos Received
35
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
2449 | 08-04-2017 06:13 PM | |
5413 | 03-21-2017 08:00 PM | |
2472 | 11-30-2016 07:38 PM | |
1271 | 11-03-2016 05:56 PM | |
2631 | 10-11-2016 06:44 PM |
06-06-2016
11:47 PM
4 Kudos
You can just omit the TTL parameter (meaning that do not specify) in which case it will default to FOREVER. I am not sure whether we used to parse TTL => 'FOREVER' as an expression before 2.4.
... View more
06-06-2016
08:32 PM
HBase base version in HDP-2.3 abd HDP-2.4 is very similar, so this should work with HDP-2.4 as well.
... View more
05-16-2016
06:46 PM
2 Kudos
You can either hbase.thrift.info.port and hbase.regionserver.thrift.port in your hbase-site.xml, or start the thrift server by passing -p <port> and --infoport <info_port> as command line arguments. Note that port and info port are different ports. Port is used for RPCs while the info port is for the web UI.
... View more
05-13-2016
12:57 AM
I would suspect a version mismatch between an installed version of ruby / jruby and the one required by HBase.
... View more
04-29-2016
06:26 PM
2 Kudos
You can check how many regions you have from the HBase master's web UI. A good rule of thumb is to keep number of regions per regionserver to be under 1000. You can also inspect the start and end keys of regions and regions sizes from the Master web UI or by going in to a RegionServers WebUI and checking various tabs. HBase splits the regions based on range boundaries of the keyspace. HBase ALWAYS does range-splitting, not hash-based splitting. This means that depending on your key design, you maybe temporarily hotspotting some parts of the keyspace causing excessive region splits. It is likely that you have a timeseries based key design that you have to revisit. You can check out the HBase book, and there are also presentations available that talks about row key and schema design.
... View more
03-24-2016
05:07 PM
2 Kudos
Each region flushes independently of each other. If a region flushes, all of its column families are flushed together by default (unless a different flush policy is used in recent code bases). A region's flush does not affect other regions.
... View more
03-18-2016
08:35 PM
2 Kudos
The SyncTool is not in HDP releases yet, but we are tracking it to bring the tool to released versions.
... View more
03-09-2016
09:29 PM
3 Kudos
In HBase, you do not have to pre-declare the set of columns as you would in a RDBMS. You can have each row have a different set of columns which is one of the powerful features of HBase. Phoenix exposes this, through a feature called "dynamic columns". You can declare a set of columns in the Phoenix table schema, but at query time or insertion time, you can do querying by specifying the columns on-the-fly. Check out https://phoenix.apache.org/dynamic_columns.html for syntax.
... View more
03-04-2016
12:03 AM
3 Kudos
@Artem Ervits is right. There is no purge of snapshots by design. Snapshots are used for referring to data at a later time. Taking a snapshot and keeping a snapshot is an explicit decision that depends only on operations. HBase in no case will delete a snapshot without an explicit command.
... View more
02-23-2016
09:57 PM
3 Kudos
You can check the official compatibility documentation here: https://hbase.apache.org/book.html#hbase.versioning. What you are looking for is "wire compatibility" between the client and server. In short, 1.x releases among different versions are client / sever wire compatible. 0.98 to 1.x might have some rough edges but it is still supported. HDP-2.2 is 0.98 based, and HDP-2.3 and 2.4 is HBase-1.1.1 based, thus inherits the upstream compatibility gurantees.
... View more