Session test#4594
Conversation
There was a problem hiding this comment.
That's a nasty test bug. So is the reason why this wasn't needing in Ice 3.6 that Glacier is doing a better job of cleaning up objects than before?
There was a problem hiding this comment.
The disconnect method was never invoking the del method nothing related to ice 3.6
There was a problem hiding this comment.
|
The tests should now turn back to green. |
|
Adding breaking label so the PR can be tested tomorrow with both ice 3.5 and ice 3.6 on breaking. now that https://ci.openmicroscopy.org/view/Breaking/job/OMERO-DEV-breaking-build is green cf. https://ci.openmicroscopy.org/view/Breaking/job/OMERO-DEV-breaking-build/637/ |
|
Tests all pass on breaking! Can't see from CI that both Ice 3.5 and Ice 3.6 are covered -- do we switch something and then look again tomorrow? snoopycrimecop/openmicroscopy@develop/breaking/trigger...snoopycrimecop:develop/merge/trigger shows a merge conflict but perhaps I'm looking at the wrong branches. |
|
(Ah, @sbesson tells me that the develop one is from the devspace.) |
|
|
OMERO-push includes, Failure of |
|
@mtbc: that's the one I was mentioning this am, not related to ice but more set-up |
|
@mtbc I have exactly the same issue across all running devspaces, not sure what permissions has to be change and where exactly. If you have any ideas, happy to test |
|
Might be interesting to check which directories are being assumed to be on the same partition. |
|
It turned out that is not a problem of directories or permissions, just lack of right ports build opened. CLI and Insight requires 4063 and 4064. devspace should be green (blue) tomorrow http://10.0.51.117:8080/job/OMERO-test-integration/ |
5 participants
Add test to check if the
_del_method actually deletes the sessionThis is not the case with ice version < 3.6
cf. https://trac.openmicroscopy.org/ome/ticket/12921
I have fixed a bug in the disconnect method. We will need to watch if some tests failed (wrong assumption in the tests)