Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • Double-click to start the setup.
  • Follow the setup dialogs.
  • Under HTTP-Port, you can configure the service-manager services port. This port has no special significance but should only be changed if the default port (7281) is already being used.
  • Under "Enter yuuvis® RAD core-service parameter", enter the IP address of the server that hosts the core-service.  
    Also enter the systemuser API key that you noted during the core-service setup.
  • Under "Enter database parameter" at "database server" enter the IP address under which the database used for yuuvis® RAD (core-service) can be reached. You may need to modify the port (1433 is generally correct for MSSQL databases).
    Also enter the database name of the database used for the yuuvis® RAD (core-service), as well as the credentials of the database user designated for yuuvis® RAD into the corresponding fields.
  • Under "Enter enaio rendition-plus parameter", enter the IP address and the port of the server that hosts yuuvis® RAD rendition-plus. Rendition-Plus does not yet have to be installed. If a different port is selected during installation of Rendition-Plus, you must enter it instead of the default 8090.
  • Complete the setup but do not start the service yet.
  • Open the file "<service-manager>\config\application-prod.yml" and add the following line to the end of itUnder "IP Filter for dms-sidecar", enter the IP addresses and/or IP address ranges that should get access to the dms-sidecar. The dms-sidecar provides access to the core-service REST-WS API and thus access to it should be limited to the necessary servers only. The value is a regex expression, refer to this page for more details.
  • Complete the setup but do not start the service yet.
  • Open the file "<service-manager>\config\application-prod.yml" and add the following line to the end of it: feign.hystrix.enabled: false
    Also, change the value of "connection-time-ms" to 60000 and of "socket-timeout-ms" to 300000, to match the following example:

    feign.hystrix.enabled: false

    service:
       connection-timeout-ms: 60000
       socket-timeout-ms: 300000


  • If the Elasticsearch installation is on a different machine than the service-manager installation, edit the <service-manager>\config\application-es.yml file and replace the 127.0.0.1 IP with the IP of the Elasticsearch machine in the following lines:

    spring.data.elasticsearch.cluster-nodes: '127.0.0.1:9300'
    spring.elasticsearch.rest.uris: '127.0.0.1:9200'


  • If you are setting up a productive system and expect to have a very high amount of documents (i.e. > 50 Mio. documents) and/or data in elasticsearch (many metadata fields and/or big fulltexts) consider increasing the number of shards from 1 (the default) to a more appropriate one. If you have more than one node in your elasticsearch cluster, use at least the number of nodes as the number of shards. We do not recommend to go higher than 3 or 4 times of the number of nodes as this would rather lower the performance again. See also https://www.elastic.co/guide/en/elasticsearch/reference/current/size-your-shards.html for more details.
    To increase the number of shards, edit the file <service-manager>\config\index-prod.yml and add the line shown below. This must be done before the first start of the service-manager (index-service) to take effekt (otherwise a reindex using the index-service's swagger-ui endpoint is necessary).

    osfts.number-of-shards: 3


  • If the core-service installation is on a different machine than the service-manager installation, edit the file <service-manager>\config\application-mq.yml and replace the 127.0.0.1 IP with the IP of the service-manager machine in the following lines:

    spring.activemq.broker-url: 'tcp://127.0.0.1:61616'
    spring.activemq.broker-url-stomp: 'stomp://127.0.0.1:61777'

    If there are multiple service-manager installations on different machines, choose IP of the installation that hosts the messaging-service.

    Make sure that the server hosting the core-service can access the messaging-service using this IP (i.e. check firewall rules etc.).
  • If the structure-service and the discovery-service are on different machines and/or the discovery-service is running on a port different than the standard (7261), edit the file <service-manager>\config\servicewatcher-sw.yml and add the "env" parameter with the corresponding values (host and/or port of the discovery-service) to the section of the structure-service as follows:

    - name: structureservice
      type: executable
      profiles: prod,red,es,mq
      instances: 1
      port: 7461-7469
      path: ${appBase}/structureservice/structure-service.exe
      env:
        EUREKA_HOST: 10.10.1.2
        EUREKA_PORT: 9261


  • Modify the number of shards for the enaiored (metadata) index (optional):
    • The default number of shards for the enaiored index is 1. If you have multiple Elasticsearch nodes and/or several ten millions of objects, you should increase the number of shards (to at least the number of Elasticsearch nodes) for better distribution and performance of the data. 
    • Open the file <service-manager>\config\index-prod.yml
    • Increase the value of the property "osfts.number-of-shards" to the desired shard count.
  • Change the rollover settings for the logfiles (optional):

    • Open the file "<service-manager>\config\servicewatcher-sw.yml" and complete the logging section as shown below.

      logging:
         file: ./../../logs/services.log
         config: ./config/logback.xml
         maxDays: 15
         totalSizeCap: 3GB

      The parameter maxDays determines for how many days a log file should be saved.
      The parameter totalSizeCap limits the total size of all log files to the specified value.

  • Change the memory assigned to individual microservices (optional):
    • In the file <service-manager>\config\servicewatcher-sw.yml, modify the memory allocation and instances for the individual services. Note: If a service is allocated more than 1.5 GB of memory, you must set the parameter "arch: x64" for this service to activate the 64-bit support.
  • Define gateway authentication method and custom endpoints (optional):

    • In the "<service-manager>\config\gateway-prod.yml" file, you can adjust the authentication mechanism for the services by setting the "authentication.filter.*" properties as follows:

      • Basic-authentication:\ntlm: false, form: true, basic: true
      • NTLM (SSO) authentication: ntlm: true, form: false, basic: false

    • You can also define additional (external) endpoints as follows:

      routing.endpoints


  • Modify the data directories (optional):
    • In the <service-manager>\bin directory, run the enaio_red_service-managerw.exe file.
    • In the "Java" tab, under Java Options, set the temp directory with the JVM option Djava.io.tmpdir. You can use the options –Dcatalina.base and –Dcatalina.home to set the Tomcat work directories.
    • In the directory <service-manager>\config, open the application-prod.yml file and modify enaio.data.path. The relative path to the .jar files is <service-manager>\apps\<*service>. If you use './../..', you end up in the service-manager root directory.
  • Modify the log directory (optional):
    • In the <service-manager>\bin directory, run the enaio_red_service-managerw.exe file.
    • In the "Logging" tab, modify the log path and, if desired, set up or change the redirects for stdout and stderr.
    • In the "Startup" tab, modify the "--logging.file" argument. The relative path to the .jar files is under <service-manager>\apps\<*service>. Use './../..' to point to the service-manager root directory.
    • In the <service-manager>\config, open the servicewatcher-sw.yml file and in the "logging" section, modify "file". The relative path to the .jar files is under <service-manager>\apps\<*service>. Use './../..' to point to the service-manager root directory.
  • Modify the service user (optional):
    • In the <service-manager>\bin directory, run "enaio_red_service-managerw.exe".
    • (Optional) In the "Log On" tab, enter the technical (Windows) user.
  • Start the service.
  • You can see an overview of the running services at http://<ip>:7273. This is the Spring Boot Admin service. It can be used to permanently monitor the services.
  • It can take several minutes for all the services to start successfully.
  • "Connection refused" exceptions appearing in the service-manager log file in the first few minutes are nothing to worry about.

...