Bug #105207 for CHI-Driver-BerkeleyDB: Driver Close

Sat Jun 13 08:26:17 2015 NHORNE [...] cpan.org - Ticket #105206: Ticket created

Subject:

Driver Close

When does the driver close the database? I have a destructor called by END not DESTROY which fails because the database is "already closed". How do I force CHI to be closed/unloaded afterwards?

Sat Jun 13 08:26:26 2015 NHORNE [...] cpan.org - Ticket created

Subject:

Driver Close

When does the driver close the database? I have a destructor called by END not DESTROY which fails because the database is "already closed". How do I force CHI to be closed/unloaded afterwards?

Sat Jun 13 08:32:16 2015 swartz [...] pobox.com - Correspondence added

Subject:	Re: [rt.cpan.org #105207] Driver Close
Date:	Sat, 13 Jun 2015 05:32:00 -0700
To:	bug-CHI-Driver-BerkeleyDB [...] rt.cpan.org
From:	Jonathan Swartz <swartz [...] pobox.com>

When the CHI handle is destroyed, its dbh will be destroyed as well. You could try keeping a global reference to $cache->dbh somewhere, that might prevent it from being destroyed: $global_dbh = $cache->dbh On Jun 13, 2015, at 5:26 AM, Nigel Horne via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: Show quoted text

> Sat Jun 13 08:26:26 2015: Request 105207 was acted upon. > Transaction: Ticket created by NHORNE > Queue: CHI-Driver-BerkeleyDB > Subject: Driver Close > Broken in: 0.02 > Severity: Important > Owner: Nobody > Requestors: NHORNE@cpan.org > Status: new > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > > > > When does the driver close the database? I have a destructor called by END not DESTROY which fails because the database is "already closed". How do I force CHI to be closed/unloaded afterwards?

Sat Jun 13 08:32:16 2015 The RT System itself - Status changed from 'new' to 'open'

Sat Jun 13 13:31:20 2015 ether [...] cpan.org - Ticket #105206: Merged into ticket #105207

Sat Jun 13 13:31:20 2015 ether [...] cpan.org - Merged into ticket #105207

Mon Jun 15 12:41:22 2015 njh [...] bandsman.co.uk - Correspondence added

Subject:	Re: [rt.cpan.org #105207] Driver Close
Date:	Mon, 15 Jun 2015 12:24:37 -0400
To:	<bug-CHI-Driver-BerkeleyDB [...] rt.cpan.org>
From:	Nigel Horne <njh [...] bandsman.co.uk>

What causes the handle to be destroyed? It's certainly still in scope when my destructor is being called. -Nigel Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > > > When the CHI handle is destroyed, its dbh will be destroyed as well. You could try keeping a global reference to $cache->dbh somewhere, that might prevent it from being destroyed: > > $global_dbh = $cache->dbh > > On Jun 13, 2015, at 5:26 AM, Nigel Horne via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: >

Download smime.p7s
application/pkcs7-signature 4.1k

Message body not shown because it is not plain text.

Mon Jun 15 13:22:36 2015 swartz [...] pobox.com - Correspondence added

Subject:	Re: [rt.cpan.org #105207] Driver Close
Date:	Mon, 15 Jun 2015 10:22:07 -0700
To:	bug-CHI-Driver-BerkeleyDB [...] rt.cpan.org
From:	Jonathan Swartz <swartz [...] pobox.com>

I thought you said it was an END block. On Jun 15, 2015, at 9:41 AM, njh@bandsman.co.uk via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: Show quoted text

> Queue: CHI-Driver-BerkeleyDB > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > > > What causes the handle to be destroyed? It's certainly still in scope > when my destructor is being called. > > -Nigel

>> <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > >> >> When the CHI handle is destroyed, its dbh will be destroyed as well. You could try keeping a global reference to $cache->dbh somewhere, that might prevent it from being destroyed: >> >> $global_dbh = $cache->dbh >> >> On Jun 13, 2015, at 5:26 AM, Nigel Horne via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: >>

> > > <smime.p7s>

Mon Jun 15 13:31:58 2015 njh [...] bandsman.co.uk - Correspondence added

Subject:	Re: [rt.cpan.org #105207] Driver Close
Date:	Mon, 15 Jun 2015 13:27:04 -0400
To:	<bug-CHI-Driver-BerkeleyDB [...] rt.cpan.org>
From:	Nigel Horne <njh [...] bandsman.co.uk>

It is. N Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > > > I thought you said it was an END block. > > On Jun 15, 2015, at 9:41 AM, njh@bandsman.co.uk via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: >

>> Queue: CHI-Driver-BerkeleyDB >> Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > >> >> What causes the handle to be destroyed? It's certainly still in scope >> when my destructor is being called. >> >> -Nigel

>>> <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > >>> >>> When the CHI handle is destroyed, its dbh will be destroyed as well. You could try keeping a global reference to $cache->dbh somewhere, that might prevent it from being destroyed: >>> >>> $global_dbh = $cache->dbh >>> >>> On Jun 13, 2015, at 5:26 AM, Nigel Horne via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: >>>

>> >> <smime.p7s>

-- Nigel Horne Conductor: Rockville Brass Band, Washington Metropolitan GSO @nigelhorne | fb/nigel.horne | bandsman.co.uk | concert-bands.co.uk | www.nigelhorne.com Unless it's for my eyes only, please use "reply all"

Download smime.p7s
application/pkcs7-signature 4.1k

Message body not shown because it is not plain text.

Mon Jun 15 13:41:06 2015 swartz [...] pobox.com - Correspondence added

Subject:	Re: [rt.cpan.org #105207] Driver Close
Date:	Mon, 15 Jun 2015 10:40:41 -0700
To:	bug-CHI-Driver-BerkeleyDB [...] rt.cpan.org
From:	Jonathan Swartz <swartz [...] pobox.com>

I thought that in an END block, everything gets destroyed. Maybe not. In that case, I don’t know…but CHI isn’t doing anything to explicitly close the database. On Jun 15, 2015, at 10:31 AM, njh@bandsman.co.uk via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: Show quoted text

> Queue: CHI-Driver-BerkeleyDB > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > > > It is. > > N

>> <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > >> >> I thought you said it was an END block. >> >> On Jun 15, 2015, at 9:41 AM, njh@bandsman.co.uk via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: >>

>>> Queue: CHI-Driver-BerkeleyDB >>> Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > >>> >>> What causes the handle to be destroyed? It's certainly still in scope >>> when my destructor is being called. >>> >>> -Nigel

>>>> <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > >>>> >>>> When the CHI handle is destroyed, its dbh will be destroyed as well. You could try keeping a global reference to $cache->dbh somewhere, that might prevent it from being destroyed: >>>> >>>> $global_dbh = $cache->dbh >>>> >>>> On Jun 13, 2015, at 5:26 AM, Nigel Horne via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: >>>>

>>> >>> <smime.p7s>

> > > -- > Nigel Horne > Conductor: Rockville Brass Band, Washington Metropolitan GSO > @nigelhorne | fb/nigel.horne | bandsman.co.uk | concert-bands.co.uk | www.nigelhorne.com > > Unless it's for my eyes only, please use "reply all" > > > > <smime.p7s>

Tue Jun 16 10:27:21 2015 NHORNE [...] cpan.org - Correspondence added

Show quoted text

> >>>> $global_dbh = $cache->dbh

I tried that, but got Can't locate object method "dbh" via package "CHI::Driver::BerkeleyDB__WITH__CHI::Driver::Role::Universal" -Nigel

Tue Jun 16 12:20:45 2015 swartz [...] pobox.com - Correspondence added

Subject:	Re: [rt.cpan.org #105207] Driver Close
Date:	Tue, 16 Jun 2015 09:20:35 -0700
To:	bug-CHI-Driver-BerkeleyDB [...] rt.cpan.org
From:	Jonathan Swartz <swartz [...] pobox.com>

Sorry, I meant $global_db = $cache->db On Jun 16, 2015, at 7:27 AM, Nigel Horne via RT <bug-CHI-Driver-BerkeleyDB@rt.cpan.org> wrote: Show quoted text

> Queue: CHI-Driver-BerkeleyDB > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=105207 > >

>>>>>> $global_dbh = $cache->dbh

> > I tried that, but got > > Can't locate object method "dbh" via package "CHI::Driver::BerkeleyDB__WITH__CHI::Driver::Role::Universal" > > -Nigel

Tue Jun 16 13:24:33 2015 NHORNE [...] cpan.org - Correspondence added

I believe that I now have a work around in place. It's not the same as yours, but is inspired by it. I would prefer not to have to do it, but it works for now. The fix is basically: my $cache = CHI->new(.....); foo($cache) Whereas I used to have, what I consider a neater option of: foo(CHI->new(....));