Versions Compared

Key

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

...

  1. Connectathon
    1. Portugal
    2. Austria
    3. EC
    4. Update will follow!
  2. Relaxation topics:
    1. Decision on how to implement and to deal with message signature (https://openncp.atlassian.net/wiki/display/ncp/Open+discussion)
      1. Decision was taken during technical committee of 18/02 to implement digital signature
        1. Discussion regarding the comment published by Natasha on the discussion page
        2. Objective of implementation is to have "integrity"
      2. What are the next steps?
        1. Massimiliano Masi: are we using the framework for ssl communication?
        2. Task list, analysis
        3. Issue to be created in JIRA
        4. Next steps, activities, to be discussed during Task force security on  
    2. National connector specifications: National Connector Specifications
      1. Feedback received on the page during the security task force
        1. Next steps:
          1. Send the mail to the OpenNCP community for consultation - wait for comment from Soeren Bittins
          2. Then we can start the implementation
      •  S will create an issue in JIRA and organize it in Epic
  3. Cache mechanism
    1. See /wiki/spaces/ncp/pages/77266962
    2. S: the use of the tool might expose to network? Proposed to use a simpler solution?
    3. Massimiliano Masi: we discussed on multiple instances...since apparently there is no need of database...Massi analysed the different cache mechanisms. After testing, they all behave the same. Implemented and created 5 singleton instances...REDIS and REM cache installed on VM. 2 scenariis were analyzed.
    4. Values cached: all info stored in TSL
      1. Kostas Karkaletsis: Frequency: one a year
      2. Questioning the need of distributed cache...Alternative could be to keep the data local & invalidate?  => New version every...max time by node = 10 sec
        1. Discussions on publicString getProperty(String name) {
        2. scheduled.time.between.failed.logs.handling.minutes
        3. João Cunha:  I think it works like this (I remember now): if that property is not found, a default value of 60 is use. João Cunha: but needs confirmation

      3. What we could do is to look at epsos properties & check whether they need to be synchronized with others
      4. Natasha Carl: Micro caching - see comment on page: https://openncp.atlassian.net/wiki/pages/viewpage.action?spaceKey=ncp&title=Open+discussion
      5. Joao Cunha: How would we load properties?
        1. we would not resolve the performance issue
    5. Next steps:
      1. Look on all the properties that we need => With Kostas Karkaletsis and Joao Cunha
        1. Joao Cunha will share the list
      2. Whether to go via distributed cache or REDIS
      3. Continue the page, keeping Natasha Carl in the loop of the discussions
      4. SMP/SML 3pm
  4. JIRA Issues review - No time to go through the issues, this will be discussed during the next Technical Committee meeting
  5. AOB
    1. Massimiliano Masi: Topic to be put at the agenda of the next meeting: Usage of a toolkit on real time monitoring
  6. Next Technical Committee meeting proposed date  1pm

...