Re: [HACKERS] Parallel worker error - Mailing list pgsql-hackers

From Tom Lane
Subject Re: [HACKERS] Parallel worker error
Date
Msg-id 24580.1504094103@sss.pgh.pa.us
Whole thread Raw
In response to Re: [HACKERS] Parallel worker error  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
I wrote:
> I can duplicate this in HEAD and v10, but not 9.6, so I've added it
> as an open issue for v10.  No idea what broke it.

Oh, scratch that: the issue exists in 9.6, it's just that the particular
test query you're using here does not generate a parallelized plan in
9.6, as shown by "explain".  With a different query that does get
parallelized, 9.6 fails too:

regression=#  select sum(ten) from tenk1;
ERROR:  role "testuser1" does not exist
CONTEXT:  parallel worker

Still, I think it's reasonable to characterize this as "must fix for v10".
        regards, tom lane



pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: [HACKERS] [BUGS] [postgresql 10 beta3] unrecognized node type: 90
Next
From: Robert Haas
Date:
Subject: Re: [HACKERS] Re: Poor cost estimate with interaction between tablecorrelation and partial indexes