2020 March, 11 Tech Committee Meeting notes
Date
Mar 11, 2020
8am PST / 4pm CEST
Meeting Link
Attendees
@Sebastian Brudziński
@Paulina Buzderewicz
@Klaudia Pałkowska (Deactivated)
@Oskar Hinc (Unlicensed)
@Wesley Brown
@Chongsun Ahn (Unlicensed)
?
Discussion items
Time | Item | Who | Notes |
|---|---|---|---|
5m | Agenda and action item review | @Sebastian Brudziński |
|
10m | Does any code required by the reporting stack still live outside of the OpenLMIS org? | @Sebastian Brudziński | E.g. https://github.com/onaio/superset-patchup
|
10m | Current status of the translations and how the build process for this works | @Wesley Brown |
|
15m | Invalid token issues | @Klaudia Pałkowska (Deactivated) |
|
Notes
Does any code required by the reporting stack still live outside of the OpenLMIS org?
We have superset-patchup repository but we are not using it currently: https://github.com/openlmis/superset-patchup
The ref-distro is set up to use the superset-patchup from the ONA repository instead of our own
@Paulina Buzderewicz contacted @Daniel Serkowski (Unlicensed) who worked on the Angola implementation to ask why we moved back from our own repository to ONA's repository
One idea is to fork the current ONA's repository containing the most recent changes and move it to our's
superset-patchup is likely the only outstanding repo that we don't own for reporting stack
Chongsun is planning to fix the problem with auth to superset as soon as this is moved
Current status of the translations and how the build process for this works
The recent demo was held in French and we discovered that some parts were not translated (e.g. system messages)
The English messages are pushed to Transifex with each build
The translated messages are included in the docker images that are built
Do we have that in our Wiki?
We have this: Translations
Invalid token issues
We have started seeing some invalid token issues both in core and Malawi and we don't know where they are coming from
Looked into this issue a little bit in performance tests: the invalid token shows up there for a few of the requests, but the token is actually valid
Did we try increasing the logging level for the Spring OAuth?
Yes, but we didn't notice anything
Could we try increasing the logging level for Spring OAuth in Malawi?
Are they client-level tokens or service-level tokens?
We think it's a service-level token issue
Most likely not related to token expiration