Re: portal pinning - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: portal pinning
Date
Msg-id 9992839b-26e3-3223-67cb-7d41c9cb4069@2ndquadrant.com
Whole thread Raw
In response to Re: portal pinning  (Vladimir Sitnikov <sitnikov.vladimir@gmail.com>)
List pgsql-hackers
On 1/10/18 13:53, Vladimir Sitnikov wrote:
>> committed
> 
> I'm afraid it causes regressions for pgjdbc.
> Here's CI log: https://travis-ci.org/pgjdbc/pgjdbc/jobs/327327402
> 
> The errors are:
> testMetaData[typeName = REF_CURSOR, cursorType =
> 2,012](org.postgresql.test.jdbc2.RefCursorTest)  Time elapsed: 0.032 sec 
>  <<< ERROR! org.postgresql.util.PSQLException: ERROR: cannot drop pinned
> portal "<unnamed portal 1>"   
> 
> It looks like "refcursor" expressions are somehow broken.

Hmm, it seems like we are missing some test coverage in core for that.

I guess I'll have to revert this patch and go with the first approach of
putting it directly into PL/Perl and PL/Python.

-- 
Peter Eisentraut              http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services


pgsql-hackers by date:

Previous
From: Sergey Burladyan
Date:
Subject: Re: Why standby restores some WALs many times from archive?
Next
From: Andres Freund
Date:
Subject: Re: Dubious shortcut in ckpt_buforder_comparator()