| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Gunnar R|nning <gunnar(at)candleweb(dot)no> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Problem migrating dump to latest CVS snapshot. |
| Date: | 2001-03-23 01:41:48 |
| Message-ID: | 15436.985311708@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general pgsql-hackers |
Gunnar R|nning <gunnar(at)candleweb(dot)no> writes:
>> BTW, did your original data contain any fractional-second timestamps?
>> I'm wondering if the original value might have been something like
>> 2000-10-24 15:14:59.999
>> in which case sprintf's roundoff of the seconds field to %.2f format
>> would've been enough to do the damage.
> What do you mean by original value ? The value we have in the production
> database ? If so, that shows up as 2000-10-24 15:14:60.00+02 independent of
> what platform my client is running on. The production platform was as I
> mentioned Solaris 2.7.
If you still have the value stored in the original database, please try
select date_part('seconds', ...)
to see what that reports as the true seconds part of the value.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Vince Vielhaber | 2001-03-23 01:51:59 | Re: Extreme Newbie Questions |
| Previous Message | Vince Vielhaber | 2001-03-23 01:29:20 | Re: Call for platforms |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Limin Liu | 2001-03-23 02:00:10 | Re: [HACKERS] SPI example does not work for 7.1beta4 |
| Previous Message | Vince Vielhaber | 2001-03-23 01:29:20 | Re: Call for platforms |