Time/Place

This meeting is a hybrid teleconference and IRC chat. Anyone is welcome to join...here's the info:

Attendees 

Agenda

  1. Updates on 4.7.4 release testing (4.7.4 JIRA tickets)?

    1. Release planned for Aug 7th

    2. Documented backwards-compatibility testing has been performed
    3. MySQL? PostgreSQL?
  2. Fedora alignment to API sprints
    1. Delta document - Additions would be helpful
    2. Spec editorial seeks feedback on impact of changes
  3. Updates from Peter Eichman re:Recent fedora-tech message
  4. Performance and PairTree discussions on #fcrepo channel of Project-Hydra slack
  5. Announcement: Andrew Woods will be out for the next two Thursdays (8/3 and 8/10)
    1. Volunteers to host those meetings?
  6. ...

  7. Status of "in-flight" tickets

    type key summary assignee reporter priority status resolution created updated due

    Unable to locate Jira server for this macro. It may be due to Application Link configuration.

Ticket Summaries

  1. Please squash a bug!

    key summary type created updated due assignee reporter priority status resolution

    Unable to locate Jira server for this macro. It may be due to Application Link configuration.

  2. Tickets resolved this week:

    key summary type created updated due assignee reporter priority status resolution

    Unable to locate Jira server for this macro. It may be due to Application Link configuration.

  3. Tickets created this week:

    key summary type created updated due assignee reporter priority status resolution

    Unable to locate Jira server for this macro. It may be due to Application Link configuration.

Minutes
4.7.4 Release Testing
  • Hydra testing appears to be done, CurationConcerns deemed not to be on critical path
  • Islandora testing complete
  • Build testing on all platforms
  • Backup/Restore testing complete, Andrew Woods will fill this section out
  • Jared Whiklo handling Postgres testing
  • Should the root node be versionable?  Discussion favors disallowing versioning of the root node and returning a 400 series error indicating the method is disallowed.  No use case has been seen so far for versioning of the root node.
  • Future plans for scripting testing backwards compatibility
    • Spin up previous 4.7.x version
    • Import data using an existing data set
    • Run tests
    • Spin down fcrepo and spin up the 4.7.4-RC
    • Run tests
    • Ensure 4.7.4-RC is capable of restarting 

We appear to be on schedule for 4.7.4 release on 8 August 2017, however Andrew Woods will not be available during this week.  Jared Whiklo is the release manager with support from Danny Bernstein.  Our original release date was planned for 1 August, but was postponed to RC2.  However due to key outages 4.7.4 may release on August 1st or 2nd.  Andrew Woods will email the list to announce the proposal to move to August 1st.

API Alignment Sprints and Delta Document 

Folks are currently signing up for planned sprints.  Further support for for the Delta Document (Agenda Item 2A) would be appreciated.  This delta document will be used to create the tickets for the planned sprints.

Editorial comments are appreciated with regard to use of features within the spec.  There are areas where removal of details (such as acl:accessToClass) would make it easier to conform to WebACL and no one seems to be using these, however this needs to be confirmed with the larger community.  However we need to determine an engagement method for informing the community of changes.  Possible method is to proposal initial removal from the specification on the Tech Call and if approved by the Tech Call announce potential larger to the larger community.  Specific questions regarding removal from the spec to be proposed and placed on the Tech Call Agendas where the initial discussion will take place. 

For the Delta Document, the most useful approach appears to be taking a portion of the specification and reading through it.  Then on the Delta Document note the differences between this and the current implementation.  If you have concerns over your knowledge over Fedora and what it is doing, feel free to reach out to Andrew Woods to work through a section together.  

Check Delta Document pages for notes with regard to someone having committed to perform work on a section (to avoid duplicate work).

Performance and PairTree

For performance reasons it has been suggested you create a three or four level deep path hierarchy of PairTree elements.  Spreading your resources out in a relatively broad tree so that a request to a container will only find a small number of children.  Fedora will automatically skip over these PairTree parents to gather the children directly, subverting the performance gains.  However removing the logic that skips over the PairTree elements will return this performance.  Samvera was having issues indexing large repositories when all children all returned (skipping PairTree elements and gathering directly).  Not skipping these PairTrees allows for more limited scope of returns.  Phuong Dinh and Chris Colvard are investigating the feasibility of indexing a Samvera repo with PairTree elements being returned.  

Interest in community feedback on useful usage of PairTrees and if it works.  

Andrews-less Operations

Jared Whiklo will lead the meeting on 3 August

Danny Bernstein will lead the meeting on 10 August



  • No labels

All content on the LYRASIS Wiki is licensed under the CC BY (Attribution) license, unless otherwise noted.