Re: [BUGS] BUG #11208: Refresh Materialized View Concurrently bug using user Postgres - Mailing list pgsql-hackers

From Kevin Grittner
Subject Re: [BUGS] BUG #11208: Refresh Materialized View Concurrently bug using user Postgres
Date
Msg-id 1409066131.57092.YahooMailNeo@web122304.mail.ne1.yahoo.com
Whole thread Raw
List pgsql-hackers
Kevin Grittner <kgrittn@ymail.com> wrote:
> Kevin Grittner <kgrittn@ymail.com> wrote:
>>  "bemanuel.pe@gmail.com" <bemanuel.pe@gmail.com> wrote:
>>
>>>  tjma_dw=> set role user_dw;
>>>
>>>  tjma_dw=> CREATE TABLE foo_data AS SELECT i, md5(random()::text) FROM
>>>  generate_series(1, 10) i;
>>>  SELECT 10
>>>  tjma_dw=> CREATE MATERIALIZED VIEW mv_foo AS SELECT * FROM foo_data;
>>>  SELECT 10
>>>  tjma_dw=> ALTER MATERIALIZED VIEW mv_foo OWNER TO user_dw;
>>>  ALTER MATERIALIZED VIEW
>>>  tjma_dw=> REFRESH MATERIALIZED VIEW mv_foo;
>>>  REFRESH MATERIALIZED VIEW
>>>  tjma_dw=> ALTER TABLE foo_data OWNER TO user_dw;
>>>  ALTER TABLE
>>>  tjma_dw=> REFRESH MATERIALIZED VIEW mv_foo;
>>>  REFRESH MATERIALIZED VIEW
>>>  tjma_dw=> create unique index on mv_foo (i);
>>>  CREATE INDEX
>>
>>>  /pgsql/pg94/bin/psql -Upostgres -p 5434 tjma_dw
>>
>>>  tjma_dw=# refresh materialized view CONCURRENTLY mv_foo;
>>>  ERROR:  permission denied for relation pg_temp_432971_2
>>>  CONTEXT:  SQL statement "DELETE FROM public.mv_foo mv WHERE ctid
>>>  OPERATOR(pg_catalog.=) ANY (SELECT diff.tid FROM  pg_temp_10.pg_temp_432971_2
>>>  diff WHERE diff.tid IS NOT NULL AND diff.newdata IS NULL)"
>>
>>  Yeah, that's a bug
>
> Attached is my proposed fix.  I will push it in a day or two if there
> are no objections.

Done.  I think we will have a third beta release; which should
include this fix.

The master branch needed to be adjusted from the initially posted
patch because of changes there.  That version is attached.

Thanks for testing the beta and for the report!

--
Kevin Grittner
EDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
Attachment

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Scaling shared buffer eviction
Next
From: Amit Kapila
Date:
Subject: Re: Scaling shared buffer eviction