Shibboleth Developer's Meeting, 2015-04-17
10:00 Central US / 11:00 Eastern US / 16:00 UK
Calls are normally the 1st and 3rd Fridays of each month. Next call would be Friday 2015-05-01. Any reason to deviate from this?
60 to 90 minute call window.
This week's call will use the Lync system at OSU. To participate, call:
- +1 (614) 688-1800 (please use if possible)
- +1 (800) 678-6114 (use only if you're charged for the 614 number)
The Conference ID is: 738127#
International participants should be able to access the 800 number without charge through Skype.
Nothing for this week.
- Web server now up, Confluence/JIRA moved, running latest versions with embedded Tomcats
- Postgres and data directories are backed up in cron scripts, saving a few days worth and mirrored via sftp
- Mail coming from admin.shibboleth.net for the moment
- Next steps...
- IDP-594 Some progress on localStorage
Picked up localStorage again. I don't think sub flows are a good idea because the conversation scope is stored in the HTTP session. Don’t really want to deal with input/output attribute mapping. Instead, we'll likely have two view-states, one for reading and optionally one for writing if localStorage is supported, falling back to cookies and the servlet filter for writes if not.
To transfer the serialized storage service context map between the client (browser) and the server (the IdP storage service) in a view-state, we can use (a) Web Flow/MVC model binding, or we can (b) evaluate an expression in the transition-state of the view-state. The advantage to model binding (a) is less XML, whereas the advantage to evaluating an expression (b) is explicit XML configuration and control.
It looks like the existing ServletRequestScopedStorageService can be retrofitted to support localStorage. I started that work and have a mock-up. My main concern at this point, as I’ve mentioned previously on the last two dev calls, is changing the scope of the ServletRequestScopedStorageService to session from singleton, and the aop-proxy I think we’ll need, at least to avoid deployers making mistakes when injecting a session scoped bean into a singleton (bean with a longer life).
My plan is to get the mock-up working, including detection for localStorage support and cookie fallback, then committing that for further work and review.
- INFRA-117 Work on service moves.
- V2 EOL plan discussion
- Schedule moving services
- When can Nexus be moved ? Jenkins would follow.
- IDP-691 Replacing java.util.Timer
- IDP-692 Replacing c3p0 with DBCP
- IDP-696 Deprecate Java 7