Default Project

Mark de Reeper

started review CR-421

23 May
Proposed fix for OPENAM-24 - Identity Changes not propagating to policy decisions
Mark de Reeper

No it would be okay in its current form as all use of the timerPoll is in if blocks depending on the polling and cleanup flags.

sachiko

summarized and closed CR-419

22 May
No summary.
Peter Major

finished reviewing CR-419

22 May
Peter Major

If polling is disabled and cleanup is disabled, would this code just threw NPE otherwise? Should we try to fix that use case in any way?

Mark de Reeper

finished reviewing CR-419

21 May
sachiko

started review CR-419

21 May
OPENAM-1326: Deadlock in PeriodicRunnable (side effect of OPENSSO-5377)
Mark de Reeper

Yes if this is the same problem being solved for the other containers. Would obviously require a lot of re-testing but would probably simplify things long term.

omebold

replied to andi in CR-408

21 May

Imagine having an already packaged osgi bundle of the customer containing all the necessary activiti stuff. This is not too probable so it is rather some possible extension point than some high prio thing.

omebold

replied to andi in CR-408

21 May

Changed to the existing header.
I've not spent any time on implementing a generic way of finding the username in the remote client knowing that this part will be refactored as well.

Peter Major

Looks good to me.

Peter Major

finished reviewing CR-418

21 May
Kohei

replied to sachiko in CR-418

21 May

OK.
I uploaded a new diff file.

Peter Major

Should this be considered as a complete replacement for the whole container lifecycle-listener stuff?

iwakata

summarized and closed CR-396

21 May
No summary.
sachiko

finished reviewing CR-418

20 May
sachiko

Please update copyright section

Mark de Reeper

This should also be SystemTimerPool.getTimerPool().schedule(this, new Date(expectedTime));

andi

I might just use the existing header to carry the user name, e,g, X-OpenIDM-Username

Also note that when we go to the json resource 2.x the explicit knowledge required to find the user in the context goes away, and will be replaced by an explicit API.

andi

As discussed, as a future task it may make sense to consider if refactoring this integration to provide a plug-in for openidm variable access might make sense. (But location independence and storage back-end are higher prios)

andi

Do we need/use both embedded and local?

jason

Abandoned CR-419

18 May
Mark de Reeper

finished reviewing CR-411

18 May
Kohei

started review CR-418

18 May
sachiko

finished reviewing CR-415

17 May
sachiko

The rule about copyright notice is as follows (got clarifications from Steve for other CR)

Square brackets are to be removed when you see them in the source code.
So the standard copyright line should be like this:

> * Portions Copyrighted 2012 ForgeRock Inc

Where a file already has a ForgeRock copyright notice; if it says ForgeRock AS it should be updated to read ForgeRock Inc and the year updated if required.

Mark de Reeper

Please ignore the last three patches, Crucible generated errors when they were added so it looked like it hadn't worked, they are the same as original LoggingThread diff.

Mark de Reeper

started review CR-415

17 May
Proposed changes to allow J2EE Agent applications to recover from multiple reloading/redeployments
GuruAllan

Abandoned CR-417

17 May
alin

Abandoned CR-416

17 May