Somewhere after r90570 the test started to crash on the bot, but locally I was not able to reproduce the crash yet so I couldn't bisect the change which caused this regression.
Skipping test until further investigation.
This is definetely a sideefect since now that the test is skipped, the next test is failing.
Instead of skipping I marked the test as CRASH.
(In reply to comment #1)
> Instead of skipping I marked the test as CRASH.
The crash disappeared during the weekend.
I'm removing the test from test_expectations.txt: