CERN-BARC phone conference, 19/7/06
Agenda:
- Quattor work - status of wassh re-engineering (Sharma
)
- Lemon developments (Murthy, Sharma, Salim)
- CCTracker (Dinesh, Bill )
Reports:
Murthy:
Hello German,
Please find the status report below,
Oramon tests were held due to db problems and a bug in oramon. Both the problems
are solved now and tests are again conducted. Started working on encryption. As
decided earlier will use the rsa/dsa keys for encrypting the samples.
regards
murthy
Salim:
Hi German,
I've developed http-fetch-module for
lemon-xml-api-cpp using "libcurl". This library is by default available
in linux. So there is no need to install additional libraries for this
purpose. Next I'll integrating and test this module with the lemon-xml-api
developed so far.
Dinesh:
Dear German/Bill ,
CCTracker status:
1) Apache ant tasks to automate CCService :
(repeatable build process for development and updating web service)
Tasks created for converting WSDL to Java code, xml schema to castor objects.
Tasks added to netbeans project to compile, deploy/undeploy CCService to tomcat.
2) CCService server message logging implemented with Log4j.
3) Checked exceptions in CCService method (SOAP faults) tested and now working.
I will extend it to remaining methods of the web service.
Sharma:
Sensor-Exception
I have corrected the problems reported by
Dennis about sensor exception. Bug# 17020 has also been corrected. I have
also incorporated support for alarm state management (Bug # 17840) in sensor
exception. This is currently being tested here.
Wassh
I have seen the links sent by you about
the scripts fetching data using URL. Now I have idea about how to do it. I
will keep you posted about the doubts if there are any.
Discussion on test environments (German):
Dennis' instructions on
how to setup a test environment have been added, and I have prepared a stub
page for defining the sensor-exception tests, which needs to be expanded by
Sharma.
Discussing with Dennis
and Miro, achieving an automated testing environment for sensor-exception
(e.g. invocable via 'make test') seems rather difficult, so I think it's
sufficient to declare in the above-mentioned twiki page exactly what tests
should be run and what results we should expect.
On every new release of
sensor-exception, the BARC team will send a mail containing the actual
observed results for all tests.