2016-11-29 Tuesday Team Meeting
Dundee:Petr, J-m, Mark, Roger, Jason, Graeme, Balaji, June, Simone, Gus, Helen, Dom, Ola, Chris, David Walker Remote: Sebastien, Josh, Eleanor, Andreas, Melissa, David, Emil, Will (14:15+)
Dundee:Petr, J-m, Mark, Roger, Jason, Graeme, Balaji, June, Simone, Gus, Helen, Dom, Ola, Chris, David Walker Remote: Sebastien, Josh, Eleanor, Andreas, Melissa, David, Emil, Will (14:15+)
Agenda - 2:00pm Start
Accepting minutes from last meeting
Project Timelines (2-3 minutes each)
-
Spaces
-
Files/Formats/OME Files (Sebastien)
- Full rebuild of C++ binaries due to libtiff secvuln 1 First time; good test. Went well.
-
Bio-Formats 5.3.0: closing last milestone
- Tiling: need final set of fixes & unit tests (samples, etc)
- Options: good to have final review and get it in tomorrow to start driving use cases
-
OME-Files 0.3.0 to focus on extending current support for heterogeneous metadata
- Related to upcoming MULTIMOT meeting (Dec 12/13)
- Discussing Roger’s proposal at the meeting tomorrow
- Roger: have fixes for transform issues
-
Metadata (Josh)
- Jason: IDR paper out
- More below.
-
OMERO 5.3.0
-
5.3.0-m6 available (OMERO+jpegxr)
- Jason: write to the lists? Was originally requested on IRC (#ome)
- Seb: would need to associate all the caveats
- Jason: another group in Australia
- J-m: could contact a few directly.
- Chris: danger of confusion
- Work on testlib to make integration tests possible in other repositories
- Discussion Thursday 10 UK on proposed deprecations (sharing, etc)
- Plan for upgrading nightshade to CentOS 7 so C6 can be dropped
-
Java 1.8 bump will require testing for MATLAB. Do workarounds exist?
- Chris: recognizing other ecosystems.
-
5.3.0-m6 available (OMERO+jpegxr)
-
Outreach
- neubias deadline end of the week
- mmc2017 Abstract submission opened
-
Files/Formats/OME Files (Sebastien)
-
Other releases/upgrades: 14:17 UK
-
Figure
- Number of cleanup PRs (other web apps as well)
-
Learning
- Holiday down time
-
Sysadmin
- Cowfish getting RAM’ed
- Suggestion to have another OpenStack training dedicated to production (features like snapshots etc). Potentially in the context of nightshade.
- Website redevelopment/redesign (Liza)
- Smuggler (Montpellier)
-
viewer
- Rename issue short list prefix OMERO.: discovery, vision, view, phase, imageviewer (IV)
-
Figure
-
Glencoe Update (Chris)
- David Walker: PhD student involved with the NPSC project.
- Trying to finish a few issues before beginning scalability work.
-
Elasticsearch experiment: https://github.com/glencoesoftware/omero-es
- Not yet highly installable
- Has pointed to a number of undocumented features
- Final contractual details before commencing Read Only Phase II, Service Routing work
-
Spaces
AOB (5 mins max - technical discussions should be highlighted to relevant people and rescheduled)
walkthrough of the IDR by Josh
-
The Mega-scenario
- Deployment
-
Infrastructure repo too complex for newcomers. Not a good starting point but contains deployment sources
- IDR/deployment as the entry point
-
Spinning up IDR deploys 8-9 instances at once
- Single proxy (nginx + ssh), 2 servers (1 for analysis) with 2 databases
- Deployment of IDR 0.3.1 will involve spinning the whole stack and switching IP addresses. No downtime
- Identical stack copy at UoD. Good place for experimentation
-
EuroBioImaging encountered issue with caching. PhenoImageShare also walked images and collecting thumbnails
- Need to keep system up and running for others
- 0.3.2: rendering. Requires tools to check everything has been updated.
-
IDR landing page (Jekyll)
- Includes virtual analysis environments. IDR notebooks capture and make IDR work available. Includes IPNBs to reproduce figures in paper.
- Walk through of mapr
- Future - search (combining searches e.g. images with certain gene plus phenotype, images in screenX with no annotation), import/background annotating, making use of feature data
- Jason: “mini-IDRs” for testing. J-m: similar to request from training.
-
The Mega-scenario