Re: [HACKERS] pgbench regression test failure - Mailing list pgsql-hackers

From Fabien COELHO
Subject Re: [HACKERS] pgbench regression test failure
Date
Msg-id alpine.DEB.2.20.1711131918090.18461@lancre
Whole thread Raw
In response to Re: [HACKERS] pgbench regression test failure  (Steve Singer <steve@ssinger.info>)
Responses Re: [HACKERS] pgbench regression test failure  (Steve Singer <steve@ssinger.info>)
Re: [HACKERS] pgbench regression test failure  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Hello Steve,

>                printf("number of transactions actually processed: " INT64_FORMAT "/%d\n",
> -                          total->cnt - total->skipped, nxacts * nclients);
> +                          total->cnt, nxacts * nclients);
>
> I think you want ntx instead of total->cnt here.

Indeed... and this is also what my git branch contains... I just sent the 
wrong version, sorry:-( The same fix is also needed in the else branch.

Here is the hopefully right version, which passes tests here.

-- 
Fabien.
Attachment

pgsql-hackers by date:

Previous
From: David Steele
Date:
Subject: Re: [HACKERS] pg audit requirements
Next
From: Fabien COELHO
Date:
Subject: Re: [HACKERS] pgbench: Skipping the creating primary keys afterinitialization