Comment 12 for bug 1930742

The relocation failure on install is not a regression vs 44.1 in -updates though, anyway. We should only fail on regressions, not make fixing additional issues requisites for verifying new issues.

But oh well, Steve fixed it, but I'd like to make sure we actually test the same things and don't come up with new tests each time that then fail but are not actually regressions compared to -updates.