Michael Paquier <michael@paquier.xyz> writes:
> This animal has been unstable for some time and this is likely just a
> symptom masking a bigger issue, but leafhopper has reported an
> intriguing failure yesterday on HEAD:
> TRAP: failed Assert("outerstartsel <= outerendsel"), File:
> "costsize.c", Line: 3678, PID: 823960
> I don't recall seeing something like that. Does that ring a bell for
> anybody?
We've had trouble with that assertion before, eg [1][2][3]. But the
previous reports have generally had to do with unexpected NaN or
Inf values in the calculation. Not sure why leafhopper and only
leafhopper would be showing that.
regards, tom lane
[1] https://www.postgresql.org/message-id/7270.1302902842@sss.pgh.pa.us
[2] https://www.postgresql.org/message-id/19320.1323039341@sss.pgh.pa.us
[3] https://www.postgresql.org/message-id/E1kUT0Q-000057-AW@gemulon.postgresql.org
