Re: Upgrade 96 -> 11 - Mailing list pgsql-general

From Adrian Klaver
Subject Re: Upgrade 96 -> 11
Date
Msg-id e06c290c-d6a3-73f6-4740-cc0260c79433@aklaver.com
Whole thread Raw
In response to Re: Upgrade 96 -> 11  (Adrian Klaver <adrian.klaver@aklaver.com>)
Responses Re: Upgrade 96 -> 11
List pgsql-general
On 9/2/19 5:52 PM, Adrian Klaver wrote:

>> It's still creating the schema elements when it fails, it hasn't 
>> started linking yet
> 
> Alright at least you still a working 9.6 cluster .
> 
> Not sure where to go from here. Like you I am not sure how it can CREATE 
> EXTENSION and not actually follow through on that. Especially with no 
> errors for that operation. I'm going to have to think on this. Hopefully 
> someone else has an idea on this and can chime in.

Aah. I don't have postgis installed, still:

pg_dump --binary-upgrade -s  -d production -U postgres > 
production_binary.sql

-- For binary upgrade, create an empty extension and insert objects into it
DROP EXTENSION IF EXISTS tablefunc;
SELECT pg_catalog.binary_upgrade_create_empty_extension('tablefunc', 
'public', true, '1.0', NULL, NULL, ARRAY[]::pg_catalog.text[]);


Try the above on your schema and see what you get.


> 
>>
>>      >
>>      > I have set PGBINOLD, PGBINNEW, PGDATAOLD, PGDATANEW correctly.
>>     

-- 
Adrian Klaver
adrian.klaver@aklaver.com



pgsql-general by date:

Previous
From: Michael Lewis
Date:
Subject: Re: literal vs dynamic partition constraint in plan execution
Next
From: "Kumar, Virendra"
Date:
Subject: Running a Simple Update Statement Fails, Second Time Suceeds.