Thread: pg_dumpall's subseq IMPORT ERRORS
I looked farther up the STDOUT/STDERR of the
psql import using the file from a pg_dumpall
There are tons of errors there besides encoding.
Sample errors:
psql:/tmp/cluster_pg_dumpall.txt:10: ERROR: role "airburst" does not exist
SET
psql:/tmp/cluster_pg_dumpall.txt:14: ERROR: relation "public.computers" does not existist
psql:/tmp/cluster_pg_dumpall.txt:19: ERROR: index "stats2_etime_index" does not exist
psql:/tmp/cluster_pg_dumpall.txt:58: ERROR: table "stats_before_spring04" does not exist
psql:/tmp/cluster_pg_dumpall.txt:76: ERROR: sequence "new_computers_id_seq" does not exist
psql:/tmp/cluster_pg_dumpall.txt:170: ERROR: language "plpgsql" does not exist
SET
psql:/tmp/cluster_pg_dumpall.txt:173: ERROR: function public.plpgsql_call_handler() does not exist
CREATE FUNCTION
SET
psql:/tmp/cluster_pg_dumpall.txt:191: NOTICE: using pg_pltemplate information instead of CREATE LANGUAGE parameters
psql:/tmp/cluster_pg_dumpall.txt:4382: ERROR: invalid byte sequence for encoding "UTF8": 0x91
HINT: This error can also happen if the byte sequence does not match the encoding expected by the server, which is controlled by "client_encoding".
CONTEXT: COPY brokenbin, line 443
Obviously this import is not a panacea!
I do know that the new DB was initialized w/ UNICODE and the old w/ SQL_ASCII.
I did a grep for "CREATED" or "CREATE D", they're not in the dump.
So,
A) How do I back out since I imported this into (I guess) the 'postgres' database instead of 'airburst'?
B) Once I get back there, what do I need to do besides
a) CREATE DATABASE (with SQL_ASCII encoding)
b) In that DB create the ROLE 'airburst'
c) ?
C) Then do I import the production version's dump?
Thanks again all,
Ralph Smith
=====================