Hey MJ ... just testing the new installer a bit and I've got a couple questions: rewrite-config.PL doesn't seem to do anything, could you explain how it works? I'd like to change the prefix to /home/jmf/koha when I do 'make' I get an error: mkdir /usr/lib/cgi-bin/koha: Permission denied at /usr/share/perl/5.8/ExtUtils/Install.pm line 443 make: *** [pm_to_blib] Error 13 My understanding of how make works is that it compiles in the current dir, and make install is what moves the files into their final directories ... of course, I could be wrong ... let me know. Thanks, -- Joshua Ferraro SUPPORT FOR OPEN-SOURCE SOFTWARE President, Technology migration, training, maintenance, support LibLime Featuring Koha Open-Source ILS jmf@liblime.com |Full Demos at http://liblime.com/koha |1(888)KohaILS ----- "MJ Ray" <mjr@phonecoop.coop> wrote:
Paul POULAIN <paul.poulain@free.fr> wrote:
MJ Ray a écrit :
The installer does not yet configure zebra. See the KNOWN BUGS bit of my announcement. That is the next-but-one thing on my TODO.
I think that the install setup should be : - install Koha, apache part - run webinstaller for librarian questions - at the end of the webinstaller, ask "are you a large library, that
will run zebra, or a small one that will run Koha 3.0 without zebra"
(explain what it means exactly here)
Here's where I'm confused: I thought that Zebra was the best thing since sliced bread and anyone who has enough server permissions to use it should be using it, because Koha-Z is faster than Koha-NZ on the same server. Is it?
If the user choose "NoZebra" => set the systempref accordingly If the user choose "Zebra" => just say "ok guy, you'll have to ask
your sysadm to install zebra, say him to look at wiki.koha.org/blabla Note that you won't be able to search in your catalogue until he has
done the stuff. Other functions are running immediatly" Then the sysadm runs what is needed to run zebra. I can't imagine even a second that setting zebra can be done through a script.
While some koha sysadmins will probably want to tweak Zebra, can't we ship a generally-useful default? What's the complication?
rebuild_zebra.pl will do most of the stuff, but the library will have to set zebra as a service, and that's different on every system (Linux,
Unix, Mac OSX, Windows)
What does this do (misc/migration_tools/rebuild_zebra.pl has no perldoc) and should we call it from some CGI (the web installer?) if required?
I'll add a note to README.txt about the zebra service.
Most of Zebra's configuration can now be set up using a set of standard conventions, I don't think we should just relegate that setup to
local sysadmin, it's too complicated a process. We need to provide
with some sound defaults IMO. For instance, I think official Koha 3.0 should have both global and database-specific configuration files,
"Joshua M. Ferraro" <jmf@liblime.com> wrote: the them probably
the following:
global: default.idx explain.att gils.att numeric.chr tagsetm.tag word-phrase-utf.chr explain.abs explain.tag kohalis sort-string-utf.chr usmarc.mar
database-specific: bib1.att record.abs
I think have added all of these and some more from zoom_basic to koha.git - I'll post patches shortly. Please fix as needed. I also see usmarc.mar there - what about other MARC flavours?
Of these, the only one that MUST be configured for each system is kohalis, that stores the Zebra authentication info.
I also need to update koha-conf.xml and the zebra-*.cfg files with system configuration details. Anything else?
Thanks, -- MJ Ray http://mjr.towers.org.uk/email.html tel:+44-844-4437-237 - Webmaster-developer, statistician, sysadmin, online shop builder, consumer and workers co-operative member http://www.ttllp.co.uk/ - Writing on koha, debian, sat TV, Kewstoke http://mjr.towers.org.uk/