Time/Place

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

• Time: 11:00am Eastern Daylight Time US (UTC-4)
• Audio/Video Conference Link: https://lyrasis.zoom.us/j/396037251
  
  • Meeting ID: 396 037 251
  • Find your local number: https://lyrasis.zoom.us/u/ad6Xb7q3ia

  • Join fedora-project.slack.com on the "tech" channel

  • Self-register at: http://slack.fcrepo.org/

Attendees 

1. Danny Bernstein
2. David Wilcox 
3. Jared Whiklo 
4. Peter Winckles 
5. Daniel Lamb 
6. Ben Pennell 
7. Thomas Bernhart  
8. Calvin Xu 
9. Michael Ritter  

Agenda

1. Announcements
   1. 2021-04 Fedora 6 Sprint 
   2. North American User Group Debrief
2. Performance issues needing attention
   1. Short-lived transactions related issues
      1. 
      2. 
      3. 
   2. Migration Utils slow-down
      1. Do we know the problem is neither memory nor IO bound?
      2. Are there JVM tunings that we haven't tried?
      3. Is a heap dump likely to help us?
      4. Status of instrumenting migration-utils with micrometer:
         
         
3. Pilots / Testing
4. Pre release short list of bug fixes and improvements
   
   1. 
   2. 
   3. 
   4. 
   5.  
   6. 
5. Marmotta is retired: fcrepo-camel-toolbox/fcrepo-ldpath depends on it.  What's our plan?
6. Volunteers to be the maintainer on https://github.com/fcrepo-exts/migration-utils
7. Any other new tickets/issues to be considered for the release
8. Your topic

Tickets

1. In Review

   
   

2. Please squash a bug!

   
   

3. Tickets resolved this week:

   
   

4. Tickets created this week:

   
   

Notes

1. Announcements
2. Performance Issues
   1. Short live transactions
      1. More of an impact on reads than writes, e.g. getting the root resource of your repository with 1m+ items
      2. Ordering adds more latency
      3. Ongoing work for short lived transactions
         1. Long lived transactions perform full join
         2. Short transactions perform a simpler query
         3. Could be expanded to writes
      4. Search Index
         1. Current work is for a synchronous update
         2. Question of how much of a performance impact there is when updating the index
         3. Possibility of adding config to update async through event bus from initial discussion
   2. Long running transactions in bad states
      1. If changes are already committed to the ocfl layer, changes can't be rolled back
      2. Might be able to attempt to rollback, then mark transaction as failed
      3. Can also prevent the transaction from being committed at all
   3. Migration utils
      1. For pid list: need to stop iterating the F3 repository once we've processed all items
      2. metrics: Try and capture bytes/sec

Actions

• Look through migration util commits since December to see if there are any potential causes of performance drop off.