Cron Job Questions Koha 3 with Zebra on Ubuntu 7.10
When a new item is added to the catalog it is not showing up until the rebuild_zebra.pl is run. The staff doing the cataloging would like to be able to have access to the records immediately. 1. What cron jobs or other procedures should be running, and how often should they run, to enable the records to update right away? 2. And a related question.... What cron jobs do you recommend running (and how often) in a "typical" koha w/ Zebra installation. Thanks, Bruce Raymond
When a new item is added to the catalog it is not showing up until the rebuild_zebra.pl is run. The staff doing the cataloging would like to be able to have access to the records immediately.
1. What cron jobs or other procedures should be running, and how often should they run, to enable the records to update right away?
2. And a related question…. What cron jobs do you recommend running (and how often) in a “typical” koha w/ Zebra installation.
In your Koha root directory, take a look to misc/cronjobs/crontab.example. It's a self-explaining file. Specifically for 1: rebuild_zebra.pl -b -a -z >/dev/null -- Frédéric
"Bruce Raymond" <braymond@regisjesuit.com> wrote:
When a new item is added to the catalog it is not showing up until the rebuild_zebra.pl is run. The staff doing the cataloging would like to be able to have access to the records immediately.
1. What cron jobs or other procedures should be running, and how often should they run, to enable the records to update right away?
I think the zebraqueue_daemon.pl should be running to enable the records to update rightaway. It is usually started by koha-zebraqueue-ctl.sh and requires the "daemon" package to be installed on debian and ubuntu. Hope that helps, -- MJ Ray (slef) Webmaster for hire, statistician and online shop builder for a small worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/ (Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237
Do note that the current zebraqueue_daemon script is only suitable for (relatively) low-circulation libraries. This daemon process updates the zebra index as soon as a record is modified in your database, but is limited to updating a single record at a time. Since the record is modified with every circ transaction (so that it holds current status information), running this daemon process is not possible if you are (or may be) performing more than 30 transactions/minute (on run-of-the-mill hardware). So, unless you are a single (or few)-site, low-transaction library ( << 30 transactions/minute ), the rebuild_zebra -z cron script should be run rather than the zebraqueue_daemon. Obviously, there's room to experiment here; beefing up the hardware could change things significantly, and I'm throwing numbers off the top of my head, not reading from a spec sheet. ( and to be clear, a transaction is 1 checkout or return / 1 patron , not n checkouts or returns / patron ). HTH Ryan On Wed, Aug 27, 2008 at 6:40 AM, MJ Ray <mjr@phonecoop.coop> wrote:
"Bruce Raymond" <braymond@regisjesuit.com> wrote:
When a new item is added to the catalog it is not showing up until the rebuild_zebra.pl is run. The staff doing the cataloging would like to be able to have access to the records immediately.
1. What cron jobs or other procedures should be running, and how often should they run, to enable the records to update right away?
I think the zebraqueue_daemon.pl should be running to enable the records to update rightaway. It is usually started by koha-zebraqueue-ctl.sh and requires the "daemon" package to be installed on debian and ubuntu.
Hope that helps, -- MJ Ray (slef) Webmaster for hire, statistician and online shop builder for a small worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/ (Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237 _______________________________________________ Koha mailing list Koha@lists.katipo.co.nz http://lists.katipo.co.nz/mailman/listinfo/koha
-- Ryan Higgins LibLime * Open-Source Solutions for Libraries Featuring KohaZOOM ILS 888-564-2457 x704
Make sure you run the Cron job under your koha user account! Also make sure you run an scripts (like rebuild_zebra.pl) as user koha!! If you are using zebra indexing: zebrasrv appears to run under user koha. That means any lock files it needs to acquire will also need to be owned by user koha. If you happen to run rebuild_zebra.pl as root, or as some other user, zebrasrv will get hung up on the lock file permissions. This will be mostly transparent to you except that the search function in koha will fail (gulp). I made this mistake and found out about the problem by: 1. Doing a ps -ef | grep zebra To find out where the error and log files for zebrasrv are stored (in /var/log/koha for me) 2. Looking in koha-zebradaemon-output.log for the error messages. 3. Going to the zebra locks direcory (/var/lock/koha/zebradb for me) and changing the ownership of the various files like this sudo chown -R koha * 4. Verify that koha owns all lock files by doing an ls -lR This fixed my problem. I then made sure I added the cron job for rebuild_zebra.pl in user koha's crontab: su koha crontab -e As an aside (and as of today's date) I am added the rebuild_zebra cron job because there is a reported problem with zebraqueue_daemon. I have been advised not to run zebraqueue_daemon due to a memory leak. The work around is to use the above mentioned cron job. Hope this helps someone else, Pete. -- View this message in context: http://koha.1045719.n5.nabble.com/Cron-Job-Questions-Koha-3-with-Zebra-on-Ub... Sent from the Koha - Discuss mailing list archive at Nabble.com.
participants (5)
-
Bruce Raymond -
Frederic Demians -
MJ Ray -
Peter Huerter -
Ryan Higgins