Huh, looks like the query won't hang if I don't specify * as columns, but doing any kind binding spec, either in the from spec, or with where, will end up with zero results, even though it just found same thing with no spec and limit

there wasn't for this one - we've been doing regular snapshot releases (nearly continuous deployment) when there haven't been breaking changes

over the last few weeks we've mostly been working on performance and bug-fixes, so there hasn't been much if anything user-facing

that said, happy to write some notifications now that we're starting to get people using it for longer-term spikes (previously it's mostly just been people playing with XTQL, short-lived in-memory databases etc, and giving feedback)

Yeah, I started testing out v2 as a local disk based cache to see how things work out

(time
   (do
    [(xtdb/query @xtdb/node '(-> (from :data [id])
                                 (limit 1)))
     (xtdb/query @xtdb/node '(-> (from :data [id])
                                 (where (= id "06420618721502"))))]))

; (out) "Elapsed time: 52.128218 msecs"
[[{:id "06420618721502"}] []]

yeah, these should still work fine - nothing immediately springs to mind. I wonder whether the other issues you're seeing are causing us to get the object-store in an inconsistent state :thinking_face:

Hm. This last one was supposedly finished fine, and was queryable before. Even when the transactions threw exceptions it was possible to restart a node and do queries just fine, without all the data though