On Sat, Dec 30, 2017 at 8:43 AM, Andres Freund <andres@anarazel.de> wrote:
> On 2017-12-29 23:06:10 +1300, Thomas Munro wrote:
>> On Wed, Dec 13, 2017 at 2:01 PM, Andres Freund <andres@anarazel.de> wrote:
>> > Use new overflow aware integer operations.
>>
>> frogmouth seems to have crashed in the vicinity of various number
>> processing tests -- could this commit be responsible? A wild guess as
>> there isn't much to go on in the build log:
>
> That seems not that likely an explanation, given there were a good
> number of runs without failure after this commit went in, before the
> animal started to fail due to PHJ.
Hmm, yeah, also the more recent failure is bombing in different
places, so I take that back. I was thrown by a couple of bits in the
earlier build log that seemed to evoke this commit:
INSERT INTO FLOAT4_TBL(f1) VALUES ('10e70');
! ERROR: value out of range: overflow
SELECT * FROM generate_series('+4567890123456789'::int8,
'+4567890123456799'::int8, 0);
! WARNING: terminating connection because of crash of another server process
--
Thomas Munro
http://www.enterprisedb.com