[Bucardo-general] Eek - can't figure out how to fix this one....

[Karl Denninger]

So I have a database I want to replicate.

I set up Bucardo, and tell it about both database copies - the master
and the remote.  All appears ok.
I add all the tables and sequences to a herd.

I added the sync as well and started it.  All appears good.

Then I told the system to "onetimecopy" it to bring everything into
synchronization.

That's when I got this:

[Sun Aug  1 12:44:21 2010]  MCP Running validate_sync on "fapforum"
[Sun Aug  1 12:44:21 2010]  MCP Source database backend PID is 71331
[Sun Aug  1 12:44:21 2010]  MCP Connecting to target database
"replicated_fap"
[Sun Aug  1 12:44:22 2010]  MCP Target database backend PID is 86354
[Sun Aug  1 12:44:22 2010]  MCP   Inspecting source table
public.ignore_thread on database "fapforum"
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "inserted" (timestamp with time zone)
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "login" (text)
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "number" (integer)
[Sun Aug  1 12:44:22 2010]  MCP   Inspecting source table public.vote on
database "fapforum"
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "inserted" (timestamp with time zone)
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "login" (text)
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "number" (integer)
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "ordinal" (integer)
[Sun Aug  1 12:44:22 2010]  MCP     Checking column on target database
"replicated_fap": "voted" (integer)
[Sun Aug  1 12:44:22 2010]  MCP   Inspecting source table
public.post_thread on database "fapforum"
[Sun Aug  1 12:44:23 2010]  MCP     Checking column on target database
"replicated_fap": "blocked" (boolean)
[Sun Aug  1 12:44:23 2010]  MCP     Checking column on target database
"replicated_fap": "forum" (text)
[Sun Aug  1 12:44:23 2010]  MCP     Checking column on target database
"replicated_fap": "inserted" (timestamp with time zone)
[Sun Aug  1 12:44:23 2010]  MCP     Checking column on target database
"replicated_fap": "login" (text)
[Sun Aug  1 12:44:23 2010]  MCP     Checking column on target database
"replicated_fap": "number" (integer)
[Sun Aug  1 12:44:23 2010]  MCP     Checking column on target database
"replicated_fap": "ordinal" (integer)
[Sun Aug  1 12:44:23 2010]  MCP Warning: Source database for sync "fapforum"
has column "ordinal" of table "public.post_thread" with a DEFAULT of
"nextval
('public.post_thread_ordinal_seq'::regclass)", but target database
"replicated_fap" has a DEFAULT of
"nextval('post_thread_ordinal_seq'::regclass)"
[Sun Aug  1 12:44:23 2010]  MCP Validation of sync fapforum FAILED

And that's that - bang.

A look at both shows that I did not intentionally specify the "public"
schema in either case.  In any event trying to specify it doesn't do me
a bit of good, nor does recreating the default value without the
"public" tag in front.  I still get the same error - the system sees the
"public " schema prefix on the local copy, but not on the remote one,
whether I have it explicitly specified or not, and as a result it won't
validate and start the copy.

I'm stuck here..... this looks like an interesting alternative to SLONY,
but first I have to get it running :-)

-- Karl

-------------- next part --------------
A non-text attachment was scrubbed...
Name: karl.vcf
Type: text/x-vcard
Size: 124 bytes
Desc: not available
Url : https://mail.endcrypt.com/pipermail/bucardo-general/attachments/20100801/cce97bc3/attachment.vcf