You are here: TWiki > Accounting Web>GratiaRelease2012 (12 Oct 2016, KyleGross)
This list contains the tasks that could or should be addressed in FY2012 and beyond

  1. Investigate if and why the email/text report the "Fermilab" VO as inactive?
  2. Better support for pilot job
  3. Coping with truncation / rotation of scraped log files. (See details)
  4. Incorporate storage reports (and all UNL reports) into Gratia proper.
  5. Service probes, emails, display [OSG Metrics]
  6. Architectural Documentation [TBD]
  7. Improve on the gratia collector test suite to make it more 'portable' and 'configurable'
  8. Reorganize replication scheduling [Chris Green] (See details)
  9. Add support for SCRUM batch system (to be contributed if needed by Jeff porter)
  10. Ensure clean and clear separation of code modules between "Gratia, the job accounting system" and "Gratia, the data transport". (See details) [Philippe]
  11. Implement posix storage probe. [OSG Metrics].
  12. Update report to support the Grid, Number of Core and application exit code field in BIRT [Philippe]
  13. Update probe so they no longer require to be run as root.
  14. Add Hourly summary table

On going tasks:

  1. Release Management
  2. Regular security and code clean up including the email reports.
    • Increase robustness
  3. Run (and react to) pylint on the python scripts.

The following are outstanding known issues that should be addressed have not been scheduled yet (see for analysis):

  1. Do the development / re-factoring work necessary to enable a collector to run multiple concurrent RecordProcessor threads.
  2. Enable multiple collectors to interact with one DB, or multiple DBs using master-master replication or other clustering techniques.
  3. Reports to group by resource_group on request.
  4. Log entries may be in UTC or local time depending on service start order.
  5. Automatic maintenance of Site -- Probe and FQAN/VO -- printable VO name translation via interrogation of OIM.
  6. Gratia is not known to handle non-standard characters safely. Neither the probes nor the collector have been verified to handle non-standard characters safely: components should be verified / updated to handle UTF-8.
  7. Gratia probes cannot deal with HTTP proxies. Standard proxy library does not do the job because we
    • Send multiple HTTP envelopes per connection; and
    • Need to be able to do the same over HTTPS.
  8. In Gratia email/text reports, make sure to exclude a VO reporting group name that shows up as empty

The following tasks have been completed:


The following are currently not planned on being addressed but would be good improvements.

  • Handling on non-latin characters.
  • Automatic table optimization (schedulable, abortable, triggerable) [Insert will be disabled during optimization]
  • Automatic sync of VONameCorrection?? table and info OIM.
  • Support change of DN (and match to same Common Name)
  • Test (and fix) multi-thread insert in Innodb
  • Add compression capability to Collector to Collector connection.
  • Ability to upload only summarized record in Collector to Collector replication

  • Add work-around to 2 md5 checksum being equal for non-duplicate records.

  • Persistent selection/exclusion lists for all charts: Listboxes or any other web entry fields for specifying VO's, sites, date range, etc. should maintain the selected values so they can be used across multiple reports.
  • The ui should provide for selecting data for fqan, (wildcarded with regular expression), DN
  • In Reports Date range selection specification should contain the last day, week, month, 6 months and year periods
  • Report displaying the run time profile (distribution) of jobs by Site and VO. Chart or xls format is acceptable. (RR21)

See an old list of priorites and the old list of enhancement requests See also Karthik's old list Work and desiderata for FY 2010 Work and desiderata for FY 2011

-- PhilippeCanal - 31 Aug 2011

Topic revision: r3 - 12 Oct 2016 - 15:23:09 - KyleGross
Hello, TWikiGuest


TWiki | Report Bugs | Privacy Policy

This site is powered by the TWiki collaboration platformCopyright by the contributing authors. All material on this collaboration platform is the property of the contributing authors..