From: Tim Brody <tdb2 AT ecs.soton.ac.uk>
Date: Wed, 11 Mar 2009 10:39:08 +0000
| Threading: | ↑ [EP-tech] mod_perl not quite V2 from g.knights AT qut.edu.au • This Message |
http://mailman.ecs.soton.ac.uk/mailman/listinfo/eprints-tech
*** EPrints community wiki - http://wiki.eprints.org/
With 1.99 I think you have to explicitly tell EPrints to look for
mod_perl 2 with (in SystemSettings):
'apache' => '2',
There are also compatibility issues between CGI.pm and 1.99, so you
require a specific version of CGI.pm.
If you can, I recommend using mod_perl 2!
All the best,
Tim.
On Wed, 2009-03-11 at 14:09 +1000, Guy Knights wrote:
> http://mailman.ecs.soton.ac.uk/mailman/listinfo/eprints-tech
> *** EPrints community wiki - http://wiki.eprints.org/
> Hi,
>
> Just a bump - is anyone able to give me any feedback about this mod_perl ↵
query?
>
> Any advice/info on this would be appreciated.
>
> Thanks,
> Guy
>
> Guy Knights
> Computer Systems Officer
> Library Systems, QUT
> Phone: (07) 3138 3910
> Email: g.knights AT qut.edu.au
>
> -----Original Message-----
> From: eprints-tech-bounces AT ecs.soton.ac.uk [mailto:eprints-tech-bounces ↵
AT ecs.soton.ac.uk] On Behalf Of Guy Knights
> Sent: Wednesday, 25 February 2009 4:03 PM
> To: eprints-tech AT ecs.soton.ac.uk
> Subject: [EP-tech] mod_perl not quite V2
>
> X
> http://mailman.ecs.soton.ac.uk/mailman/listinfo/eprints-tech
> *** EPrints community wiki - http://wiki.eprints.org/
> Our eprints server is hosted with a central team at QUT, and we asked ↵
during setup for mod_perl 2 to be installed. However, I've been browsing ↵
through the Eprints doco and noticed that the install process for Redhat EL 4 ↵
mentions that the version of mod_perl 2 that comes with EL 4 is not exactly ↵
version 2, but rather beta 1.99. I did a test on our server, and yes, it's ↵
actually version 1.9916 that is installed.
>
> My question is, does this have any negative effect on the operation of ↵
EPrints? Our server runs well enough and having the incorrect version of ↵
mod_perl doesn't appear to have affected the stability of the system; however ↵
we do get errors in the logs that I suspect *could* be linked to the ↵
discrepancy. For example, if we run 'refresh_views' and then request a browse ↵
view, the system will take some time to generate the view and it will display ↵
ok in the browser. However, in the httpd log we get the following error (which ↵
only occurs at the time the view is generated):
>
> ------------------------------------------------------------------
> ---------------- EPrints System Error ----------------------------
> ------------------------------------------------------------------
> CGI Query object not defined!
> ------------------------------------------------------------------
> EPrints System Error inducing stack dump
> at /usr/local/eprints/perl_lib/EPrints.pm line 91
> EPrints::abort('CGI Query object not defined!') called at ↵
/usr/local/eprints/perl_lib/EPrints/Session.pm line 3080
> EPrints::Session::param('EPrints::Session=HASH(0xb2dbbb0)') called ↵
at /usr/local/eprints/perl_lib/EPrints/Session.pm line 795
> EPrints::Session::get_full_url('EPrints::Session=HASH(0xb2dbbb0)') ↵
called at /usr/local/eprints/archives/quteprints/cfg/cfg.d/dynamic_template.pl ↵
line 67
> ↵
EPrints::Config::quteprints::__ANON__('EPrints::Session=HASH(0xb2dbbb0)', ↵
'HASH(0xb801e3c)') called at /usr/local/eprints/perl_lib/EPrints/Repository.pm ↵
line 1164
> eval {...} called at ↵
/usr/local/eprints/perl_lib/EPrints/Repository.pm line 1164
> EPrints::Repository::call('EPrints::Repository=HASH(0xa1001e8)', ↵
'ARRAY(0xb78cb40)', 'EPrints::Session=HASH(0xb2dbbb0)', 'HASH(0xb801e3c)') ↵
called at /usr/local/eprints/perl_lib/EPrints/Session.pm line 2625
> EPrints::Session::prepare_page('EPrints::Session=HASH(0xb2dbbb0)', ↵
'HASH(0xb801e3c)', 'page_id', 'browseindex') called at ↵
/usr/local/eprints/perl_lib/EPrints/Session.pm line 2535
> ↵
EPrints::Session::write_static_page('EPrints::Session=HASH(0xb2dbbb0)', ↵
'/usr/local/eprints/archives/quteprints/html/en/view/series/Ad...', ↵
'HASH(0xb801e3c)', 'browseindex') called at ↵
/usr/local/eprints/perl_lib/EPrints/Update/Views.pm line 410
> ↵
EPrints::Update::Views::update_view_menu('EPrints::Session=HASH(0xb2dbbb0)', ↵
'HASH(0x8c926f8)', 'en', 'ARRAY(0xacb66e0)') called at ↵
/usr/local/eprints/perl_lib/EPrints/Update/Views.pm line 133
> ↵
EPrints::Update::Views::update_view_file('EPrints::Session=HASH(0xb2dbbb0)', ↵
'en', '/view/series/Advances_in_Tourism_Research_series/index.html', ↵
'/view/series/Advances_in_Tourism_Research_series/') called at ↵
/usr/local/eprints/perl_lib/EPrints/Apache/Rewrite.pm line 224
> ↵
EPrints::Apache::Rewrite::handler('Apache::RequestRec=SCALAR(0xb450f98)') ↵
called at -e line 0
> eval {...} called at -e line 0
>
> I don't know whether the above could be resolved by an upgrade to an ↵
actual version of mod_perl 2, but I'm interested to find out whether there are ↵
any potential side-effects to running a beta 1.99 version like the one we have ↵
installed.
>
> Thanks,
> Guy
>
> Guy Knights
> Computer Systems Officer
> Library Systems, QUT
> Phone: (07) 3138 3910
> Email: g.knights AT qut.edu.au
>
>
>
>