Re: dropdb --force

čt 21. 11. 2019 v 6:33 odesílatel vignesh C <vignesh21(at)gmail(dot)com> napsal:

> On Mon, Nov 18, 2019 at 6:30 PM Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com>
> wrote:
> >>
> >> I'll send this test today
> >
> >
> > here is it
> >
>
> Thanks for adding the test.
> Few comments:
> This function is same as in test/recovery/t/013_crash_restart.pl, we
> can add to a common file and use in both places:
> +# Pump until string is matched, or timeout occurs
> +sub pump_until
> +{
> + my ($proc, $stream, $untl) = @_;
> + $proc->pump_nb();
> + while (1)
> + {
> + last if $$stream =~ /$untl/;
> + if ($psql_timeout->is_expired)
> + {
>

yes, I know - probably it can be moved to testlib.pm. Unfortunately it is
perl, and I am not able to this correctly. More it use global object - timer

This can be Tests drop database with force option:
>

done

+#
> +# Tests killing session connected to dropped database
> +#
>
> This can be changed to check database foobar1 does not exist.
>

done

+# and there is not a database with this name
> +is($node->safe_psql('postgres', qq[SELECT EXISTS(SELECT * FROM
> pg_database WHERE datname='foobar1');]),
> + 'f', 'database foobar1 was removed');
>
> This can be changed to check the connections on foobar1 database
> +
> +# and it is connected to foobar1 database
> +is($node->safe_psql('postgres', qq[SELECT pid FROM pg_stat_activity
> WHERE datname='foobar1' AND pid = $pid;]),
> + $pid, 'database foobar1 is used');
>

done

> This can be changed to restart psql as the previous connection will be
> terminated by previous drop database.
> +
>

done

new patch attached

Regards

Pavel

+# restart psql processes, now that the crash cycle finished
> +($killme_stdin, $killme_stdout, $killme_stderr) = ('', '', '');
> +$killme->run();
>
> Regards,
> Vignesh
> EnterpriseDB: http://www.enterprisedb.com
>