One lesson from XL we got is that we need testing framework for cluster, so any cluster project should at least pass functional and performance testing.
+1. In early XC days, we focused a lot on adding newer features and supporting as many PG features as possible. That took its toll on the testing and QA. It was a mistake though my feeling was we tried to correct that to some extend with XL. We did a 9.5 merge, which of course was a big deal, but other than more time is being spent on improving stability and performance
XL was very easy to break and I'm wondering how many corner cases still exists.
Your team reported 2 or 3 major issues which I think we were able to fix quite quickly. But if there are more such issues which your team has recorded somewhere, I would request you to send them to the XL mailing list. I would definitely want to look at them and address them.