Re: pg_dump fatal error

I believe this error can occur when the input to pg_restore terminates or
is truncated unexpectedly. I'm going to guess a failure in the SSH piece.

Cheers,
Steve

On Thu, Feb 11, 2016 at 10:33 AM, drum(dot)lucas(at)gmail(dot)com <drum(dot)lucas(at)gmail(dot)com
> wrote:

> ERROR: unexpected message type 0x58 during COPY from stdin
> On 12 February 2016 at 07:30, drum(dot)lucas(at)gmail(dot)com <drum(dot)lucas(at)gmail(dot)com>
> wrote:
>
>> Hi guys
>>
>> I'm doing a copy (pg_dump) from a slave server to a test server.
>> The DB size is 1,7 TB, and I get the error at 1,4 TB.
>>
>> Command:
>>
>>> ssh postgres(at)servidorslave01 "/usr/pgsql-9.2/bin/pg_dump
>>> --exclude-table-data='junk.*' --format=custom db_live"
>>> | /usr/pgsql-9.2/bin/pg_restore --dbname=uat_temp --exit-on-error
>>
>>
>> Error:
>>
>>> |postgres|pg_restore| CONTEXT: COPY ja_feedlog, line 1
>>> STATEMENT: COPY ja_feedlog (id, clientid, mobiuserid, customerid,
>>> invoiceid, description, gtime, jobid, log_type, serialised_data,
>>> push_status, requestorid, the_geom, admin_read, visitid) FROM stdin;
>>
>>
>> LOG: could not send data to client: Broken pipe
>>> FATAL: connection to client lost
>>
>>
>> What can cause that kind of error? I've tried more than once, actually.
>>
>> Thanks
>> Lucas
>>
>
>