2024-01-23

Attending: Dom, J-M, Frances, Khaled, Melissa, Kevin, Chris, David G, Andreas, David S, Petr, June, Kyle, Will, Seb, Maadhav, Erin

Start: 2:00 pm UK

Accepting minutes from last meeting

(2-3 minutes each)

IDR
- Focusing on issue leading to server being not responsive
  - slow to load chuncks (secondary effects)
  - Will looked at alternative to access NGFF data on s3 (geesefs vs goofys) but no real improvement
  - Petr - testing with
    - idr-next: idr0090 NGFF data on s3, with e.g. 100 images in figure is quickest route to e.g. 502 Bad Gateway (timeouts).
    - idr-next: On idr0138 (NGFF data but on filesystem not s3, regular imported with chunks, rather than mkngff). Server recovers better.
    - idr-next: With idr0002 data (not NGFF) much harder to break web
    - IDR production: With idr0002 data on production IDR, everything is OK
  - Will - we’ve both noticed that IDR is faster than idr-next. Is that expected?
  - Chris - Glencoe uses different stack, microservices etc.
OMERO
- working on last remaining ansible roles (used mainly by IDR)
- Andreas: cleanup of python2 compatibility code.
  - JM: team to help with testing
Bio-Formats
- Melissa: 7.2.0 work in progress
- David: ZarrReader: checking for any potential perf improvements
NGFF
- Will/Petr: missing blosc library prevents NGFF import on merge-ci
GerBI
- Discussed annotation scripts at Friday’s web meeting, how to release etc.
Glencoe
- Chris: few small things in progress. Python 3.12 support, Java 17
- Erin: Webknosses integration presentation tomorrow. Will present at next week’s meeting
Community

(5 min. max; tech. Discussion should be highlighted to relevant people and rescheduled)

(20-25 minutes plus 15 minutes questions max)