메뉴 건너뛰기

검색창 닫기

Cloudera, BigData, Semantic IoT, Hadoop, NoSQL

Cloudera CDH/CDP 및 Hadoop EcoSystem, Semantic IoT등의 개발/운영 기술을 정리합니다. gooper@gooper.com로 문의 주세요.


Cloudera CDH/CDP Cloudera의 API를 이용하여 impala의 실행되었던 쿼리 확인하는 예시

총관리자 2018.05.03 18:29 조회 수 : 5003

curl -u 'xxxx':'xxxx' 'http://CM_server.domain.com:7180/api/v11/clusters/cluster/services/impala/impalaQueries?from=2017-10... statement RLIKE ".*fawzea.*"' >>f.json


http://CM-server.domain.com:7180/api/v19/clusters/cluster/services/impala/impalaQueries?statment%20RLIKE%20%22.*test.*%22




*참고 : https://cloudera.github.io/cm_api/apidocs/v19/index.html

Introduction

This document describes the Cloudera Manager REST API. All requests and responses are presented in Javascript Object Notation (JSON).

The API resources listed below follow standard Create-Read-Update-Delete (CRUD) semantics where the HTTP request path defines the entity to be acted on and the HTTP method expresses the type of action to perform.

HTTP MethodOperation
POSTCreate entries
GETRead entries
PUTUpdate or edit entries
DELETEDelete entries

All collections in the API use plural names, 'users', instead of the singular, 'user'. To address a specific user in the system, expand the URL path to include the user identifier. For example, '/users/foo' identifies user 'foo' and '/users/bar' identifies user 'bar'.

CollectionPOST (create)GET (read)PUT (update)DELETE (delete)
/usersCreate a new userList all users in the systemBulk update all usersDelete all users
/users/fooerrorRead information about user 'foo'If user 'foo' exists, update their information; otherwise, error.Delete user 'foo'

Keep in mind that not all collections support all operations. For example, events in the system are read-only; you cannot create new events with the API.

You can list the entries in a collection using one of two views: 'summary' or 'full'. The default 'summary' view provides the core information about each entry. The 'full' view is more heavyweight and provides a fully expanded view of each entry. The view is controlled by a query parameter called 'view' e.g. 'GET /users?view=full'.

HTTP requests should set the "Content-Type" header to "application/json" for most calls, unless specified otherwise.

Authentication

The Cloudera Manager API uses HTTP basic access authentication. It accepts the same user credentials as the web interface. Different users may have different levels of access, as defined by their roles. (See the user management API calls for more.) With every authenticated request, the server returns a session cookie, which can be subsequently used for authentication.

Error Code

The API returns HTTP code in the 400s and 500s upon errors:

  • 400 — Invalid request parameters; Malformed requests
  • 401 — Authentication failure
  • 403 — Authorization failure
  • 404 — Object not found
  • 500 — Internal API error
  • 503 — Response temporarily unavailable; caller should retry later

See /tools/echoError for a discussion on the body of an error response.

API Usage Tutorial

Tutorial link.

API Version

The API can be queried for the latest version it supports. By making a GET request to the '/version' URL, a string will be returned with the highest supported version (for example, "v2").

Note that servers that only support API v1 do not support this call. In that case, a 404 error will be returned, and can be used to detect that the server only supports v1.

The version should be prepended to the path of the REST resources listed below. The documentation of the different resources lists the version which added the particular functionality.

REST Resources

This API supports a Representational State Transfer (REST) model for accessing a set of resources through a fixed set of operations. The following resources are accessible through the RESTful model:

There is also a WADL document describing the REST API.

Data Model

All endpoints act on a common set of data. The Cloudera Manager API uses JSON.

Data Elements

이 게시물을
이 글의 추천인 목록 목록
번호 제목 날짜 조회 수
187 oozie 4.1 설치 - maven을 이용한 source compile on hadoop 2.5.2 with postgresql 9.3 2015.04.30 4144
186 cloudera에서 spark-shell를 실행했을때 default master는 spark.master=yarn-client임 2018.06.20 4137
185 It is indirectly referenced from required .class files 오류 발생시 조치방법 2017.03.09 4136
184 Failed to write to server: (no server available): 2022.01.17 4135
183 DB별 JDBC 드라이버 2015.10.02 4133
182 [Sentry]HDFS의 ACL을 Sentry와 연동후 테스트 2020.06.02 4130
181 source, sink를 직접 구현하여 사용하는 예시 2019.05.30 4128
180 [kudu]테이블 drop이 안되고 timeout이 걸리는 경우 조치 방법 2020.06.08 4123
179 HUE를 사용할 사용자를 추가 하는 절차 2018.05.29 4112
178 HDFS에서 quota 설정 방법및 확인 방법 2022.03.30 4107
177 [Impala TLS/SSL이슈]RangerAdminRESTClient.java:151] Failed to get response, Error is: TrustManager is not specified 2023.02.02 4104
176 Cloudera설치중 실패로 여러번 설치하는 과정에 "Running in non-interactive mode, and data appears to exist in Storage Directory /dfs/nn. Not formatting." 오류가 발생시 조치하는 방법 2018.03.29 4094
175 start-all.sh로 spark데몬 기동시 "JAVA_HOME is not set"오류 발생시 조치사항 2016.08.01 4091
174 hive metastore ERD file 2018.09.20 4090
173 sqoop에서 oracle관련 작업할때 테이블명, 사용자명, DB명은 모두 대문자로 사용할것 2014.05.15 4088
172 권한회수 및 권한부여 명령 몇가지 2017.11.16 4084
171 like검색한 결과를 기준으로 집계를 수행하는 java 소스 2016.12.19 4084
170 우분투 16.04LTS에 Zeppelin 0.7.3설치 2018.04.18 4081
169 [AD(LADP)] CDP1.7에서 AD및 Kerberos를 연동해도 각 노드에 os account, os group은 생성되어야 하지만 SSSD서비스를 이용하면 직접 생성될 필요가 없다. 2022.06.10 4080
168 [CentOS 7.4]Hadoop NFS gateway기동시 Cannot connect to port 2049 오류 발생시 확인/조치 2022.03.02 4080
쓰기 태그
위로