Yes, Fridolyn is correct, the problem is when launching the cronjob manually using rebuild_zebra.pl -v -b -r I added one record manually into the catalog, and that one is showing up. But the ones that were loaded by taking from the primary instance (we duplicated the database) don't come up in the searches. Is there a "correct" or prescribed method of duplicating a database that we should be following? We are going to try Fridolyn's method of dealing with the variables. Thanks to all for your help. On 4/12/2011 6:12 AM, Fridolyn SOMERS wrote:
On Tue, Apr 12, 2011 at 11:38 AM, Tomas Cohen Arazi <tomascohen@gmail.com <mailto:tomascohen@gmail.com>> wrote:
2011/4/12 Fridolyn SOMERS <fridolyn.somers@gmail.com <mailto:fridolyn.somers@gmail.com>>: > Hie, > > There is one problem with multi instances that you maybe didn't notice : > > The environment variables KOHA_CONF and PERL5LIB are the same for each > instance > (they are redefined in cronjobs and apache conf). > They are mandatory for lauching perl scripts manually.
This is acknowledged in the instructions when creating the cronjob files.
To+
Yes I agree for cronjobs. But I understood the problem was when lauching a script manually. For example, recreating Zebra database : /perl /usr/share/koha/misc/migration_tools/rebuild_zebra.pl <http://rebuild_zebra.pl> -b -v -r/
In this case, the script looks for C4::Context using the PERL5LIB env var.
Regards,
-- Fridolyn SOMERS ICT engineer PROGILONE - Lyon - France fridolyn.somers@gmail.com <mailto:fridolyn.somers@gmail.com>
-- Linda Culberson lculber@mdah.state.ms.us Archives and Records Services Division Ms. Dept. of Archives & History P. O. Box 571 Jackson, MS 39205-0571 Telephone: 601/576-6873 Facsimile: 601/576-6824