reg_pg_stefanz@perfexpert.ch writes:
> it run into an issue with: ERROR: no known snapshots
Hi! This has been reported before, but I'm embarrassed to say that
the previous report seems to have fallen between the cracks.
I've created a patch [1] that resolves your example as presented.
If you're in a position to apply the patch and see if it resolves
your real non-simplified case, that would be very helpful.
Also, this fix in principle will create a small performance
penalty for FOR-loops in non-atomic contexts such as DO loops.
It'd be interesting to know if the penalty is noticeable in
your usage.
regards, tom lane
[1] https://www.postgresql.org/message-id/1032535.1620789957%40sss.pgh.pa.us