Update 1.2 to 1.3
The following sections explain which adjustments must be made to migrate from Camunda Cloud 1.2.x to 1.3.2 for each component of the system.
Server​
Zeebe​
A critical issue which may lead to data loss was identified in 1.3.0 and 1.3.1. This issue is related to the new assignee and candidate group feature introduced in 1.3.0, and only affects users which make use of it. However, when updating, it's still recommended that you skip versions 1.3.0 and 1.3.1 and update directly from 1.2.9 to 1.3.2.
Please refer to the release notes for more.
With Zeebe 1.3.0, we removed the monitoring specific configuration from the broker and the gateway. Instead, the monitoring configuration is now completely handled via the Spring Boot management server. Read this Spring Boot documentation for more information on how to configure the server.
For the broker, this replaces the following configuration settings:
zeebe:
broker:
network:
monitoringApi:
host: 0.0.0.0
port: 9600
For the gateway, the following was removed:
zeebe:
gateway:
monitoring:
enabled: true
host: 0.0.0.0
port: 9600
To configure the monitoring URL for the broker and gateway, you can instead use the server.port
and server.address
properties. Given how Spring loads its configuration, it's not possible to configure this directly in the YAML configuration. However, you can use environment variables - respectively SERVER_PORT
and SERVER_ADDRESS
- or system properties - respectively -Dserver.port=
or -Dserver.address=
- to configure them.
To disable monitoring the monitoring server, configure the broker or gateway with the following:
spring:
main:
web-application-type: none
Operate​
A critical issue was found on Operate data importer which may lead to incidents not being imported to Operate. This issue is affecting versions 1.3.0
, 1.3.1
, 1.3.2
and 1.3.3
.
We strongly recommend to skip affected versions and make sure you are running version 1.3.4
if updating from 1.2.x
.
Please refer to the release notes for more.
With Operate 1.3, a new feature was introduced to propagate incidents from called instances to calling instances. This feature is only available for instances started after the version 1.3 update. Older instances will still be shown as active even though incidents in called instances may exist.
Tasklist​
Because of internal changes in user data processing update to Tasklist 1.3 will erase all information about task assignments. End users would need to claim their tasks again.