Versions Compared

Key

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

...

Expand
titleMeeting 02 - 2016-12-07

Meeting 02 - 2016-12-07

  • 12pm ET

 

  1. https://github.com/fcrepo4-labs/fcrepo-import-export/pull/48

    Should we support mapping repository paths, not just host/port

    Esmé Cowles - This ticket came out of integration test for round tripping, export and import into a different directory. 

    We had some allowance for different hostname ports, but base URI and everything else must be the same. At least the base URI could be modified in case of changed deployment.

    While doing that Hydra already has various sub-directories, so that could be great. But it brings up issues. How do you handle if you export a resource, but not any linked resources.

    Is it a requirement to be able to export a resource/tree of resources and then import it into a different location in a repository. This means we are changing the resource.

    There is more complication, but it would be nice to do it. If we run into unsolvable problems, then we can stop. But baseURI as a minimum. 

    Verification tooling would have to be updated as it not designed to handle this circumstance. If there is a need for this, then we can correct this in the verification tooling. 

    Does this resolve the Phase 1 #6 wish. "Support import into an existing empty container."

    We have been assuming that the container is going to be the same, host/port is a hard one to adjust for, changing base URI in Tomcat is somewhat easier. 

    Don’t want to get sidetracked on this issue and not achieve the sprint goals.

    Base Path is generally “/fcrepo/rest”

    Ingesting in to a Hydra instance under /dev and then after verifying moving it to /prod.

    Esmé Cowles will make a ticket or two and point to pull/48 as a first stab and move onto integration tests.

    Final question “What would we want inter-resource relationships to look like.” when ingesting to a new location. Principle was what would it look like if you exported the entire repository and import the entire repository. We haven’t talk through all the possibilities, it would be good to make sure we agree on this.

     

  2. Validation

    https://github.com/fcrepo4-labs/fcrepo-import-export-verify/pull/7

    Hope the PR is complete, Josh has not tested but feels it is almost there. The latest commit just changed the default logger level.

    Should be able to test the latest version this afternoon and merge.

    It would be great if that issue could be closed, and the config file as changed a bit and that needs to be dealt with.

    Won’t have working tool until the update to deal with new config file options. Should be a quick fix (possibly completed today).

    Still haven’t sorted out the issues around verifying RDF, filtering out server managed triples.

    Can Nick help with verifying PRs or with updating the code to test the config, Nick will take

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2332
    .

    If you exported and then tried to verify against a Fedora instance that might have new objects you will not get any errors. The system is not designed to check both ways.

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2322
    .

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2307
    is around verifying a Fedora to another Fedora too.

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2329
    this is trying to check something, it finds an error and breaks off. What we want is for it to deal with the error and move on. 

    Next person that runs in the “Too many open files in system” should document the steps in a new JIRA ticket.

    Nick has done lubm and the 10k indirects and never run into this. Josh ran the one-click on his machine and he may have had other applications which could have used up some file handles.

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2329
    generally the error returned from Fedora is a 404 not a “Too many open files in system”. Remove that error from the ticket so as not to confuse the issue.

    Josh will test PR-7 and that will close

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2284
    .

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2190
    &
    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2309
    is done once Nick completes
    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2332
    .

    Josh will next take

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2303
    .

    Probably need to walk through all the verification tickets and close tickets that might have been resolved by existing work.

    Jira
    serverDuraSpace JIRA
    columnssummary,type,assignee,reporter,priority,status,resolution
    maximumIssues20
    jqlQueryfilter=13801
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5

     

  3. Logging

    https://github.com/fcrepo4-labs/fcrepo-import-export/pull/49

    There is different logging on the application side,

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2169
    and
    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2170
    is about a different kind of log which is more of an audit log.

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2237
    is a different issue but has some discussion around the log format in the notes.

     

  4. Esmé Cowles is creating a ticket for setting Digest header when importing checksum for binary

     

  5. Danny Bernstein is moving onto

    Jira
    serverDuraSpace JIRA
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
    keyFCREPO-2296
    .

  6. Andrew Woods suggests we all add more words in IRC Standup notes so people can understand where everyone is at.

Expand
titleDebrief - 2016-12-16

Only a few tickets left inprogress and in review

Jira
serverDuraSpace JIRA
columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
keyFCREPO-2365

Jira
serverDuraSpace JIRA
columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
serverIdc815ca92-fd23-34c2-8fe3-956808caf8c5
keyFCREPO-2328
- leaving for later.

 

Verification has some errors showing up, was kind of hard as import-export tool was constantly updating.

Perhaps not have both tools worked at same time would be good, but having lots of people around was helpful.

 

 — Good —

IRC standups +++

3 java developers.

For some on their first sprint attending meetings and learning the processes was helpful.

Nice having for testers to have developers available to ask questions.

Nice to not be a Java developer so they could use the tool more like an end user.

4 new people jump in successfully

2 people doing user testing

 

— Bad —

code reviews could take some time

people in and out during sprint/travel ++

LoC BagIt stuff

JIRA made it hard to see what was going on (maybe an embedded jira chart on the sprint page).

Felt like there wasn’t a lot to work on (for reviewing)

Using (clicking on) a JIRA label was handy once you figure it out. 

Nice to have Andrew around

Working on import/export and verification at the same time might be more pain than its worth.

 

—Questions —

With import/export and verification would it have useful to have matching tags between the two products to synchronize. — dev was too fast at this point and there is no release for either. 

 

— Next Steps —

Some stuff to get us to Stage 3

Lots of stakeholder feedback

Ready for a 0.1 release and have people try it out.

 

Leaving Bag profiles tickets open for now and have stakeholders verify that they are ready.

 

Standups

IRC Template

No Format
[Import/Export Standup]
Finished yesterday: 
  {ticket titles and associated JIRA links}
  {and/or brief textual description}
Working on today:
  {ticket titles and associated JIRA links}
  {and/or brief textual description}
Blockers:
  {brief textual description}

...