| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | lockhart(at)fourpalms(dot)org |
| Cc: | Ruslan A Dautkhanov <rusland(at)scn(dot)ru>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: date/time compatible problems in 7.2 |
| Date: | 2002-02-21 03:39:32 |
| Message-ID: | 4554.1014262772@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Thomas Lockhart <lockhart(at)fourpalms(dot)org> writes:
>> psql:/.../dbdump/.dbrestore.tmp:1624094: ERROR: copy: line 1, Bad
>> timestamp external representation 'Fri 25 Jan 23:59:59 2002 KRAT'
>> psql:/.../dbdump/.dbrestore.tmp:1624094: lost synchronization with
>> server, resetting connection
> Not sure why it is crashing. But "KRAT" is a time zone not recognized by
> the PostgreSQL date/time parser.
The "crash" is totally expected behavior after any error during a COPY
IN. There isn't any other way to recover except to reset the
connection. Yes, this sucks, it's broken, etc, but there's no way to
fix it except to redesign the frontend/backend COPY protocol :-(
Trust me, this *will* get changed next time we have occasion to make
incompatible changes in the FE/BE protocol. But I'm not sure that it's
a sufficient reason to create a protocol incompatibility all by itself.
As to the specific changes in datetime datatype behavior that cause
the error report, I bow to Thomas' superior knowledge...
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2002-02-21 03:47:58 | Re: SET NULL / SET NOT NULL |
| Previous Message | Rod Taylor | 2002-02-21 03:38:01 | Re: SET NULL / SET NOT NULL |