It appears that students are no longer being presented with peer submissions to grade.
How many students do you have in your test workshop and what do you
have the over allocation set to and the number of expected student
submissions? I'm currently running a test workshop with four
students. I initially set it up so that each student had to assess one
other student, and the over allocation was set to 1, and no one ever
got anything to assess.
I changed the number to assess from 1 to 2, and then one student got 2
submissions to assess, two got 1 to assess, and the last got no new ones to
assess. In the end this led to four new assessments, which were allocated evenly, one per submission. Growing suspicious, I upped the number to assess from 2 to 3, and indeed got additional assessment assignments (1, 2, 1, and 0 respectively). Notice that this means that the fourth student has had
no submissions to assess, while the first and second have 3 and the third as 2.
All this suggests that there are at least two different bugs in the workshop code. First, it looks like the number of required assessments is in fact off by one: Requiring 1 got none, requiring 2 got an average of 1, and requiring 3 got an average of 2. This may be related to the fact that in my test workshop I've required the assessment of one instructor submission and self-assessment, so it's possible that the code in (incorrectly) counting either of these in its count of assessments assigned/done.
Second, it looks like the overallocation code is broken in some way, since a distribution of 3, 3, 2, 0 shouldn't be allowed if the requirement is 2 with an overallocation of 1.
It's possible that these problems only show up with very small workshops, or maybe they're just easier to see when you only have four students.