Investigate if an ECC data table analysis table in the Public schema can preempt a second attempt of an analysis. See this email exchange:
Yes, I saw that it was very fast. I believe this is another edge-case issue: my working hypothesis is that the original congruence analysis of your data table from Wednesday was still in the database when you just reran the update, which caused it to pre-empt performing another congruence analysis. I'll have to run additional tests and review the code to see if this is possible.