Apache
/
hbase
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
hbase/hbase-hbtop
History
Bryan Beaudreault fd5b526398 Add maven and RPM building in blazar 
This uses our standard buildpack, but each module must get a custom .blazar.yaml to pass in required
properties to the build. One can distribute that .blazar.yaml using build-scripts/copy_blazar_yaml.sh.
In order to minimize the duplication, most of the customizations are consolidated in
build-scripts/prepare_environment.sh.

The RPM build is hooked into the standard build hbase-assembly, as a post-build step. The RPM build
scripts in hbase-assembly/rpm-build handle turning the assembled tar.gz into multiple RPMs. We required
some minor changes to the descriptor in hbase-assembly to account for the fact that our build processes
each module independently instead of as a single reactor build. As a result, moduleSet does not work, so
moduleSet usages were converted to dependencySet.
 		2021-09-24 15:51:16 -04:00
..
img HBASE-11062 hbtop (#476) 2019-09-07 14:17:59 +09:00
src HBASE-25452 Use MatcherAssert.assertThat instead of org.junit.Assert.assertThat (#2826) 2020-12-30 11:06:58 +08:00
.blazar.yaml Add maven and RPM building in blazar 2021-09-24 15:51:16 -04:00
README.md HBASE-11062 hbtop (#476) 2019-09-07 14:17:59 +09:00
pom.xml Preparing hbase release 2.4.6RC0; tagging and updates to CHANGES.md and RELEASENOTES.md 2021-09-03 09:41:52 -07:00

README.md

hbtop

Overview

hbtop is a real-time monitoring tool for HBase like Unix's top command. It can display summary information as well as metrics per Region/Namespace/Table/RegionServer. In this tool, you can see the metrics sorted by a selected field and filter the metrics to see only metrics you really want to see. Also, with the drill-down feature, you can find hot regions easily in a top-down manner.

Usage

You can run hbtop with the following command:

$ hbase hbtop

In this case, the values of hbase.client.zookeeper.quorum and zookeeper.znode.parent in hbase-site.xml in the classpath or the default values of them are used to connect.

Or, you can specify your own zookeeper quorum and znode parent as follows:

$ hbase hbtop -Dhbase.client.zookeeper.quorum=<zookeeper quorum> -Dzookeeper.znode.parent=<znode parent>

Top screen

The top screen consists of a summary part and of a metrics part. In the summary part, you can see HBase Version, Cluster ID, The number of region servers, Region count, Average Cluster Load and Aggregated Request/s. In the metrics part, you can see metrics per Region/Namespace/Table/RegionServer depending on the selected mode. The top screen is refreshed in a certain period – 3 seconds by default.

Scrolling metric records

You can scroll the metric records in the metrics part.

Scrolling metric records

Command line arguments

Argument Description
-d,--delay <arg> The refresh delay (in seconds); default is 3 seconds
-h,--help Print usage; for help while the tool is running press h key
-m,--mode <arg> The mode; n (Namespace)|t (Table)|r (Region)|s (RegionServer), default is r (Region)

Modes

There are the following 4 modes in hbtop:

Mode Description
Region Showing metric records per region
Namespace Showing metric records per namespace
Table Showing metric records per table
RegionServer Showing metric records per region server

Region mode

In Region mode, the default sort field is #REQ/S.

The fields in this mode are as follows:

Field Description Displayed by default
RNAME Region Name false
NAMESPACE Namespace Name true
TABLE Table Name true
SCODE Start Code false
REPID Replica ID false
REGION Encoded Region Name true
RS Short Region Server Name true
LRS Long Region Server Name false
#REQ/S Request Count per second true
#READ/S Read Request Count per second true
#FREAD/S Filtered Read Request Count per second true
#WRITE/S Write Request Count per second true
SF StoreFile Size true
USF Uncompressed StoreFile Size false
#SF Number of StoreFiles true
MEMSTORE MemStore Size true
LOCALITY Block Locality true
SKEY Start Key false
#COMPingCELL Compacting Cell Count false
#COMPedCELL Compacted Cell Count false
%COMP Compaction Progress false
LASTMCOMP Last Major Compaction Time false

Namespace mode

In Namespace mode, the default sort field is #REQ/S.

The fields in this mode are as follows:

Field Description Displayed by default
NAMESPACE Namespace Name true
#REGION Region Count true
#REQ/S Request Count per second true
#READ/S Read Request Count per second true
#FREAD/S Filtered Read Request Count per second true
#WRITE/S Write Request Count per second true
SF StoreFile Size true
USF Uncompressed StoreFile Size false
#SF Number of StoreFiles true
MEMSTORE MemStore Size true

Table mode

In Table mode, the default sort field is #REQ/S.

The fields in this mode are as follows:

Field Description Displayed by default
NAMESPACE Namespace Name true
TABLE Table Name true
#REGION Region Count true
#REQ/S Request Count per second true
#READ/S Read Request Count per second true
#FREAD/S Filtered Read Request Count per second true
#WRITE/S Write Request Count per second true
SF StoreFile Size true
USF Uncompressed StoreFile Size false
#SF Number of StoreFiles true
MEMSTORE MemStore Size true

RegionServer mode

In RegionServer mode, the default sort field is #REQ/S.

The fields in this mode are as follows:

Field Description Displayed by default
RS Short Region Server Name true
LRS Long Region Server Name false
#REGION Region Count true
#REQ/S Request Count per second true
#READ/S Read Request Count per second true
#FREAD/S Filtered Read Request Count per second true
#WRITE/S Write Request Count per second true
SF StoreFile Size true
USF Uncompressed StoreFile Size false
#SF Number of StoreFiles true
MEMSTORE MemStore Size true
UHEAP Used Heap Size true
MHEAP Max Heap Size true

Changing mode

You can change mode by pressing m key in the top screen.

Changing mode

Changing the refresh delay

You can change the refresh by pressing d key in the top screen.

Changing the refresh delay

Changing the displayed fields

You can move to the field screen by pressing f key in the top screen. In the fields screen, you can change the displayed fields by choosing a field and pressing d key or space key.

Changing the displayed fields

Changing the sort field

You can move to the fields screen by pressing f key in the top screen. In the field screen, you can change the sort field by choosing a field and pressing s. Also, you can change the sort order (ascending or descending) by pressing R key.

Changing the sort field

Changing the order of the fields

You can move to the fields screen by pressing f key in the top screen. In the field screen, you can change the order of the fields.

Changing the order of the fields

Filters

You can filter the metric records with the filter feature. We can add filters by pressing o key for ignoring case or O key for case sensitive.

Adding filters

The syntax is as follows:

<Field><Operator><Value>

For example, we can add filters like the following:

NAMESPACE==default
REQ/S>1000

The operators we can specify are as follows:

Operator Description
= Partial match
== Exact match
> Greater than
>= Greater than or equal to
< Less than
<= Less than and equal to

You can see the current filters by pressing ^o key and clear them by pressing = key.

Showing and clearing filters

Drilling down

You can drill down the metric record by choosing a metric record that you want to drill down and pressing i key in the top screen. With this feature, you can find hot regions easily in a top-down manner.

Drilling down

Help screen

You can see the help screen by pressing h key in the top screen.

Help screen

Others

How hbtop gets the metrics data

hbtop gets the metrics from ClusterMetrics which is returned as the result of a call to Admin#getClusterMetrics() on the current HMaster. To add metrics to hbtop, they will need to be exposed via ClusterMetrics.