Compilation of HDB++ Configurator GUI
|
|
---|---|
Dear Tango Community, In order to understand better how to build TANGO & HDB++ from the sources, I have just created a script to automatically build and install TANGO and HDB++. https://github.com/sblanchet/tango-install It compiles without hiccup most of the programs except HDB++ CONFIGURATOR GUI. I have just cloned the github repository https://github.com/tango-controls/hdbpp-configurator/ and then run maven inside. (I have no experience with maven, so I think that I have missed something)
As an alternative, I have also tried to run the precompiled jar file from https://www.tango-controls.org http://www.tango-controls.org/community/project-docs/hdbplusplus/hdbplusplus-doc/configuration-gui/ Direct link: https://bintray.com/tango-controls/maven/download_file?fi le_path=org%2Ftango%2Fhdb%2Fhdbpp-configurator%2F2.2%2Fhdbpp-configurator-2.2.jar Then I have followed this documentation to start up the application. http://www.esrf.eu/computing/cs/tango/tango_doc/tools_doc/hdb++-configurator/introduction.html#start-up The application runs but when subscribing to a new variable I get an error message "Event properties wrongly configured"
I will really appreciate a hint to get out of this pitfall. Best regards, — Sebastien
Sebastien
|
|
|
---|---|
Hi Sebastien, HDB++ is fully event-based. It fully relies on Tango archive events, so when you try to add an attribute to the list of attributes archived in HDB++, the HDB++ ConfigurationManager will check whether the Tango attribute you want to add is correctly configured to send archive events. There are different ways for a tango device server to send archive events. The easiest one is to enable the polling on this attribute (the attribute will be read every x ms) and to set up archive event properties (absolute change archive threshold, relative change archive threshold and/or the archive period) for this attribute. If the configuration manager detects an inconsistency in the attribute configuration (polling not started or no archive event property set), it will report the error you are seeing. Another way for a device server to send archive events is to push archive events by code directly when needed. In this case, you don't necessary need to have the polling enabled for this attribute. Depending on how is configured your device server, you might still need to configure correctly the archive event properties for this attribute. If your device server is pushing events by code, you can click on the "Event pushed by code" checkbox in the HDB++ configurator dialog box which appears when you want to add an attribute to an archiver. We will have a look at the MAVEN issue. Hoping this helps, Reynald
Rosenberg's Law: Software is easy to make, except when you want it to do something new.
Corollary: The only software that's worth making is software that does something new. |
|
|
---|---|
Hi again, To verify whether the attribute you want to add is correctly configured to send archive events, I think the HDB++ configuration manager is actually trying to subscribe for archive events coming from this attribute before to add it to the system. In your case, it might be that your device server managing your attribute was not running when you tried to add the attribute? Or it might be your HDB++ configuration manager device server was not running? The ConnectionFailed in the stack trace seem to indicate that something which was supposed to run was not running. Cheers, Reynald
Rosenberg's Law: Software is easy to make, except when you want it to do something new.
Corollary: The only software that's worth making is software that does something new. |
|
|
---|---|
Hi Sebastien I have updated the pom.xml and I think it's better now. I have added the bintray reference in pom.xml file. It was in my $HOME/.m2/settings.xml file before. Regards Pascal |
|
|
---|---|
Hi Reynald and Pascal, With the new commit b8d9c6111699a47e2e744f9d4e2da0342ceadc31 from Pascal, I can build HDB Configurator without error. When running the jar file for the first time, I have got an error because I had an IPv6 address for localhost in my /etc/hosts After removing the IPv6 address, the error message disappears.After that, I have applied Reynald's instructions and finaly I have successfully subscribed to sys/tg_test/1/double_scalar (see the screenshot). It seems to work, because I see the new records in the mysql database. Now I will focus on the HDB Viewer to plot the data. Thank for your help.
Sebastien
|