Hi
I’m testing Camunda batch process with Junit4 all is going well and fast when we try with the Postgres database, but as soon we change to Oracle to execute the same tests as Postgres did, those tests execute well, but it takes longer than expected. Postgres takes around 3 minutes to execute 226 tests, but Oracle can take between 12 minutes and 1 hour and even more to execute the same 226 tests.
Do you know if we have to know some special considerations to configure with Oracle database?
Is it normal behavior between Camunda batch and Oracle?
This is our Postgres configuration:
spring:
profiles: dev-postgres
jpa:
hibernate:
ddl-auto: validate
database: POSTGRESQL
database-platform: org.hibernate.dialect.PostgreSQL9Dialect
properties:
hibernate:
ejb:
interceptor: com.opessoftware.library.jpa.util.HibernateCustomInterceptor
datasource:
driverClassName: org.postgresql.Driver
hikari:
pool-name: fatca-custom-pool
maximum-pool-size: 50
jdbc-url: ${jdbcUrl:jdbc:postgresql://localhost:5432}/fatca?reWriteBatchedInserts=true
username: postgres
password: *********
opessoftware:
databasename: fatca
auth:
datasource:
url: ${jdbcUrl:jdbc:postgresql://localhost:5432}/auth
username: postgres
password: *********
fatca:
engine:
datasource:
url: ${jdbcUrl:jdbc:postgresql://localhost:5432}/fatca-camunda?reWriteBatchedInserts=true
username: postgres
password: *********
batchsize: 5000
camunda:
bpm:
application:
isDeleteUponUndeploy: false
isScanForProcessDefinitions: false
isDeployChangedOnly: true
isResumePreviousVersions: true
resumePreviousBy: a value
job-execution:
enabled: true
metrics:
enabled: false
db-reporter-activate: false
generic-properties:
properties:
batchJobsPerSeed: 20
invocationsPerBatchJob: 5
database:
type: postgres
schema-update: true
jdbc-batch-processing: true
history-level: full
This is our Oracle configuration:
spring:
profiles: dev-oracle
jpa:
hibernate:
ddl-auto: none
database: oracle
properties:
hibernate:
dialect: org.hibernate.dialect.Oracle12cDialect
ejb:
interceptor: com.opessoftware.library.jpa.util.HibernateCustomInterceptorOracle
datasource:
driverClassName: oracle.jdbc.OracleDriver
hikari:
pool-name: fatca-custom-pool
maximum-pool-size: 50
jdbc-url: ${jdbcUrl}
username: fatca
password: ********
opessoftware:
databasename: fatca
auth:
datasource:
url: ${jdbcUrl}
username: auth
password: ********
fatca:
engine:
datasource:
url: ${jdbcUrl}
username: "\"fatca-camunda\""
password: ********
batchsize: 5000
camunda:
bpm:
application:
isDeleteUponUndeploy: false
isScanForProcessDefinitions: false
isDeployChangedOnly: true
isResumePreviousVersions: true
resumePreviousBy: a value
job-execution:
enabled: true
metrics:
enabled: false
db-reporter-activate: false
generic-properties:
properties:
batchJobsPerSeed: 20
invocationsPerBatchJob: 5
database:
type: oracle
schema-update: true
jdbc-batch-processing: true
history-level: full
Let us know if we doing something wrong in our configuration.
We are using:
- Our own configure engine with SpringProcessEngineConfiguration.
- Camunda 7.14
- Camunda batch 1.5.1
Thanks in advance!
Juan.