August 2012 - Koha-devel - lists.koha-community.org

How to get CAS based Single Sign On working?
by Koustubha Kale 25 May '15

25 May '15

Hi All, I have a situation where we have a main website based on Joomla. This Joomla site uses CAS to authenticate users. We would like users ( who have already logged in to the Joomla site via CAS) to be able to follow a link to Koha site and be logged in automatically ( e.g. Single Sign On ). I have got simple CAS ( not proxy CAS ) working with Koha ( version 3.2.6 ) using the same CAS server as used by Joomla site. Still when I login to Joomla site with CAS, and then go to Koha and click on login and follow the CAS login link, I am asked to fill up CAS login information again. How can this be avoided by using CAS SSO? I have read through http://wiki.koha-community.org/wiki/SSO_CAS_improvements and am confused by the statement "Note: simple sso cas support (no proxy) has been done already by biblibre, is in testing and should be released in Koha 3.2" in there. Does this statement mean that SSO is possible with simple CAS? If yes how to set it up? If not, I have also read through bugs 6012, 5630 and the CAS Proxy examples. Frankly I haven't been able to figure out how I can achieve the above required functionality based on all the reading. Any help is much appreciated.. Another related query is, Is it possible to automatically create users in Koha once they have authenticated with CAS? Regards, Koustubha Kale Anant Corporation Contact Details : Address : 103, Armaan Residency, R. W Sawant Road, Nr. Golden Dyes Naka, Thane (w), Maharashtra, India, Pin : 400601. TeleFax : +91-22-21720108, +91-22-21720109 Mobile : +919820715876 Website : http://www.anantcorp.com Blog : http://www.anantcorp.com/blog/?author=2

3 3

koha-3.00.00 make fails
by Rick Welykochy 28 Dec '13

28 Dec '13

Hi all, I have downloaded http://download.koha.org/koha-3.00.00.tar.gz and began installing it on Debian/etch (in a Linux vserver environment). 0. FINE: cd koha-3.00.00 1. FINE: perl Makefile.PL 2. FAIL: make : : : Manifying blib/man3/C4::UploadedFile.3pm Manifying blib/man3/C4::BackgroundJob.3pm Manifying blib/man3/opac::opac-dictionary.3pm Manifying blib/man3/serials::member-search.3pm /usr/bin/perl "-Iblib/arch" "-Iblib/lib" fix-perl-path.PL blib /usr/bin/perl "-Iblib/arch" "-Iblib/lib" rewrite-config.PL blib/KOHA_CONF_DIR/koha-conf.xml Bad arg length for Socket::inet_ntoa, length is 0, should be 4 at rewrite-config.PL line 67. make: *** [blib/KOHA_CONF_DIR/koha-conf.xml] Error 255 Any ideas what went wrong and how to fix? Aside: there were a couple of other problems with this "out of the box" debian based install of Koha 3.0 which had to be fixed along the way, e.g. seven missing perl modules and libgd not found. I'll report on the entire process from unpacking to a running Koha/3.00.00 once I get to the end. cheers rickw -- ________________________________________________________________ Rick Welykochy || Praxis Services || Internet Driving Instructor The manual said Windows XP/SP2 or better so I installed Linux.

4 5

Re: [Koha-devel] koha-zebra-daemon not starting
by Paul 24 Jan '13

24 Jan '13

I've tried to keep this to a minimum, there is a 'conclusion' further down: At 11:47 AM 8/20/2012 -0300, Tomas and Barry wrote: >On Sun, Aug 19, 2012 at 4:36 PM, Paul <paul.a(a)aandc.org> wrote: > > Koha 3.8.3 on Ubuntu 12.04 LTS server (new install, full upgrade) >[snip] > > But neither OPAC nor staff search functions are operational. > > > > The only anomaly that I've found so far is that > > /etc/init.d/koha-zebra-daemon when run from the command line reports > failure > > at line 73 (and does not create any error logs), but ps aux reports the > > daemon as running (well, "S+" -- sleep forground.) > > > > Anyone got any thoughts please? > >You need to answer what method r u using for install. Pls see my earlier email [Mon, 20 Aug 2012 09:48:19 -0400] >If using >tarball, then you might be running the zebra daemon using de 'koha' >user and 'koha' group. Do they exist? If this was the case, then this >should create the missing dir: > >mkdir -p /var/lock/koha/zebradb/biblios/ ; chown -R koha:koha >/var/lock/koha/zebradb/ Yes - tarball, but the dir was already (correctly) there: paul@server:/var/lock/koha/zebradb$ ls -l total 0 drwxr-xr-x 2 koha koha 40 Aug 20 11:33 authorities drwxr-xr-x 2 koha koha 40 Aug 20 11:33 biblios >Also (just in case): >chown -R koha:koha /var/log/koha Again, correctly there: paul@server:/var/log$ ls -l | grep koha drwxr-xr-x 2 koha koha 4096 Aug 19 11:19 koha >Also, every time you run rebuild_zebra.pl you should do it like: > >su - koha -c "KOHA_CONF=/etc/koha/koha-conf.xml >PERL5LIB=/usr/share/koha/lib >/usr/share/koha/bin/migration_tools/rebuild_zebra.pl -b -r -v -x" What's the '-c'? (I don't think I've ever used this construct) and I don't think there's any difference, but I 'su koha' first, then 'cd usr/share/koha', then KOHA_CONF=/etc/koha/koha-conf.xml PERL5LIB=/usr/share/koha/lib ./bin/migration_tools/rebuild_zebra.pl -a -r -v and KOHA_CONF=/etc/koha/koha-conf.xml PERL5LIB=/usr/share/koha/lib ./bin/migration_tools/rebuild_zebra.pl -b -r -v -x and Barry Cannon wrote: >Shouldn't running "zebrasrv -f /etc/koha/koha-conf.xml" (or whatever your >path is) as the index owner give you a good indication of what is going on >(it should at least get you pointed in the right direction as it should >dump out the error to console) Interesting !!! or ;=} -- I tried that: koha@server:/$ zebrasrv -f /etc/koha/koha-conf.xml 11:51:06-20/08 [log] zebra_start 2.0.44 419ad759807269fdfa379799a051ed3a551c6541 11:51:06-20/08 [log] config /etc/koha/zebradb/zebra-biblios-dom.cfg 11:51:06-20/08 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so [snip more filters for zebra-biblios-dom and zebra-authorities-dom 11:51:06-20/08 [server] Adding dynamic listener on unix:/var/run/koha/zebradb/bibliosocket id=1 11:51:06-20/08 [server] Adding dynamic listener on unix:/var/run/koha/zebradb/authoritysocket id=2 at which point that terminal just hangs on an ongoing op, but shows that the sockets were not previously opened. So, re-indexing again: for authorities: 25901................................ Records exported: 25933 ==================== REINDEXING zebra ==================== skipping biblios ==================== CLEANING ==================== and for biblios: 37301...........................................................................37401.................... Records exported: 37417 ==================== REINDEXING zebra ==================== 12:04:12-20/08 zebraidx(2841) [warn] Unknown register type: ==================== CLEANING ==================== Which corresponds to the ps aux | grep koha during the re-indexing: koha 2738 0.0 0.1 119084 4212 pts/2 S+ 12:01 0:00 su - koha -c KOHA_CONF=/etc/koha/koha-conf.xml PERL5LIB=/usr/share/koha/lib /usr/share/koha/bin/migration_tools/rebuild_zebra.pl -b -r -v -x koha 2747 73.8 1.4 259028 56272 pts/2 S+ 12:01 2:07 /usr/bin/perl /usr/share/koha/bin/migration_tools/rebuild_zebra.pl -b -r -v -x koha 2841 96.1 1.6 233556 64240 pts/2 Rl+ 12:04 0:27 zebraidx -c /etc/koha/zebradb/zebra-biblios-dom.cfg -v none,fatal,warn -g marcxml -d biblios update /tmp/q2UPWJgDrk/biblio and now we've got the search in OPAC and staff-admin finding the 37301 biblios. And I still don't know what the "[warn] Unknown register type: " might mean. Some hangover from 3.6.1 to 3.8.3? BUT ... the incremental cron (shows up in /var/log/syslog as working every 1 minute), does _NOT_ add new records. Conclusion: I'll keep going at this (half an hour here and there, it's our "tourist season", busiest time of the year) but somewhere, can't remember, there's something about Ubuntu 12.04 not renewing sockets, .LOKs, etc on reboot. Maybe I'm way off track, but it's the best avenue I have at the moment. Thoughts, please? Best - Paul

6 17

"Damaged" values
by Paul 09 Jan '13

09 Jan '13

In our "Authorized values" > "Damaged", we have added a term "Normal wear" to the Koha defaults 0, 1 and 2: 0 Not damaged Not damaged 1 Damaged Some damage noted 2 Unknown Condition not stated 3 Normal wear Normal wear However, in both staff interface and OPAC, "normal wear" items are appearing as damaged I assume (maybe incorrectly) that the value of "damaged" is read by detail.pl starting line 196: if ($item->{damaged}) { $item->{itemdamagedloop} = GetAuthorisedValues($authvalcode_items_damaged, $item->{damaged}) if $authvalcode_items_damaged; } and then passed to detail.tt start line 310: [% IF ( itemloo.damaged ) %] [% IF ( itemloo.itemdamagedloop ) %] [% FOREACH itemdamagedloo IN itemloo.itemdamagedloop %] [% IF ( itemdamagedloo.selected ) %] <span class="dmg">[% itemdamagedloo.lib %]</span> [% END %] [% END %] [% ELSE %] <span class="dmg">Damaged</span> [% END %] [% END %] I have no experience with this type of 'loo' and 'loop' coding, but it seems to me that "normal wear" is not being picked up, and therefore the ELSE default "Damaged" is being used. Could some kind person please point me to the logic of these loops? Either a brief explanation here (preferred) or a pointer to a tutorial (I have googled with no success.) Or is there something more intrinsic|relevant to the use of authorized values for "damaged"? Bugs [3637] and 5797 may be germane. Many thanks, Paul --- Maritime heritage and history, preservation and conservation, research and education through the written word and the arts. <http://UltraMarine.ca>, <http://AandC.org> and <http://MarDoc.ca>

2 2

Re: [Koha-devel] combining and minification of javascript and css
by Paul 26 Sep '12

26 Sep '12

At 08:36 PM 4/7/2012 -0400, Chris Nighswonger wrote: >On Sat, Apr 7, 2012 at 6:02 PM, Dobrica Pavlinusic <dpavlin(a)rot13.org> wrote: >I would like to address this one as part of performance work. >[snip] >Minification alone doesn't buy us much: > >OPAC search: >minify js 334974 -> 311553 93.01% >minify css 148277 -> 133128 89.78% > >intranet search: >minify js 382102 -> 362172 94.78% >minify css 134754 -> 123771 91.85% > >But combining JS and CSS files together saves us round-trips to server: > >OPAC 32 -> 20 requests >intranet 27 -> 13 requests Has anyone tried using "page speed v1.12" (the google add-on to bugzilla) which although designed for chrome will work with firefox? Google seem to be quite keen on the 'minify' principle (see <http://code.google.com/p/minify/>.) It has started pushing me towards looking at various combinations of 'minify' (in itself not a great saver, as stated above, as it seems just to remove white space and comments then unify logic on a single line basis), 'combination' (which might have limits with the variety of somewhat older browsers) and x-cache which has advantages, but only in a stable environment -- try changing a favicon ;={ First thoughts from my end seem to point towards combining (css files separately from js at this point in time) into what's needed to display the page ("instant gui gratification") and what can be slightly delayed ("page functions") and getting the download server==>client in the correct order. Best - Paul

3 2

Record indexer integration (Zebra)
by Tomas Cohen Arazi 20 Sep '12

20 Sep '12

An effort is being made to update the INSTALL.* files and to aid our users in running Koha almost auto-magically (read .deb install method, copy and paste instructions for src install in debian/ubuntu, etc). A missing piece (from my point of view) is the integration of the record indexer daemon into Koha (currently the default is to reindex -z every 5', and there are links to using Koha::Contrib::Tamil tools for having a *daemon* to do it). I vote for doing an effort to have Koha::Contrib::Tamil integrated into Koha. Once it is in the source tree this will: - Avoid Koha version-specific incompatibilities (read ease of maintenance). - Provide a simple way (SysV) to run the daemon. - Make it possible for the daemon to evolve (not sure, but I read that jcamins had some patches to adapt the Koha::Contrib::Tamil to the koha-common package multi-instance schema). Even if we move to a SolR/Zebra/Whatever: if we support Zebra it will be needed. As spoken with FrédericD, this implementation of the indexer lacks some features (?) as record sanitization, or ISO record indexing instead of MARCXML, but maybe we conclude it is not needed anymore there. Or it could be reintroduced once is in the source tree. As I see it, once it is integrated in the Koha source its easier to patch and make it evolve. Regards To+

6 12

Re: [Koha-devel] 3.8.4 interim [was: koha-zebra-daemon not starting]
by Paul 15 Sep '12

15 Sep '12

At 04:41 PM 8/22/2012 -0300, Tomas Cohen Arazi wrote: [snip] > > The tarball has always worked extremely well on i386 machines -- I just > get the impression (not proven) that Ubuntu 12.04 gets its knickers in a > twist on AMD64. Â But with a little time and effort workarounds can be > found. Â Probably nothing wrong, just new quirks, with either Ubuntu or > Koha, but I do not need to be under pressure for the production upgrade. > >is not a 12.04 problem, is a problem with the specific file you're using >with dpkg, which lists some files not available anymore on 12.04. >As Mark said that file has already been patched and its waiting for >inclusion on next release. Thanks for the reply. But, as you say, "its waiting for inclusion on next release" so it's broken in this release. >That said, you can do a clean install using the instructions and ommiting >that --set-selections step. And installing the dependencies by hand. As >Mark also said. More or less what I've done -- but it would not be rocket science to edit install_misc/ubuntu.packages to avoid a certain amount of grief (and before anyone asks "well why don't you do it?" I will as soon as I'm certain that my experience is reproducible and bullet-proof. It may involve looking at the upstream .pl file to see how the list is invoked.) >I also use the tarball in our deployment (38 instances on a single BIG >server) without issues. I'm only trying a test install on relatively BIG server (64-bit, 6 core Intel i7-980X, solid state raided drives, 16Gb ram), mileage varies... > Besides that i386 packages pull which I reported and has been solved. When, where, how? I had to 'sudo apt-get purge .*:i386' for about 50 dependencies in the pull in koha-3.08.04.tar.gz >Anyway, I think we should go back to what your original problem was.Â Is >your incremental indexing working? That was 3.8.3, now trashed, but it never worked -- I'll try it on a 3.8.4 install as soon as I'm away from this machine. >Have you properly set your sax parser? [snip] Yes. Best - P. --- Maritime heritage and history, preservation and conservation, research and education through the written word and the arts. <http://NavalMarineArchive.com> and <http://UltraMarine.ca>

13 42

Overdue Notices
by glaws 14 Sep '12

14 Sep '12

I have a question about overdue notices in Koha 3.8.3. Not sure if this is a bug or not... When a patron has overdue items, the "First Overdue Notice" will generate a list of those items. Perfect. When other overdue items subsequently become overdue within an undetermined but limited amount of time, a "First Overdue Notice" is generated for those items, but if the first set of items have not been resolved (by returning them), they are repeated in the second "First Overdue Notice". See attached .pdf. Desired/correct functionality? -- Greg Lawson Network Administrator Rolling Hills Consolidated Library 1912 N. Belt Highway St. Joseph, MO 64506

3 3

Syspref format (995$x=foo)
by Jonathan Druart 30 Aug '12

30 Aug '12

Hi all, As a result of a QA comment from Marcel (see Bug 8307), I come back to you. I introduce a syspref which can be filled with the form: xxx$y=v where xxx is a field, y a subfield and v a value. Does that make sense for you ? Does someone have a better suggestion ? Thank you for your attention. Regards, Jonathan

3 2

Coding guidelines for command-line scripts
by Owen Leonard 30 Aug '12

30 Aug '12

We have the beginnings of some coding guidelines for command-line scripts, but it doesn't look like they are firm: http://wiki.koha-community.org/wiki/Coding_Guidelines#Command-line_argument… I was testing Bug 8674, "Need a delete biblios script" (http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=8674) and was frustrated that there was no help option, but was unsure whether that should block the script from getting a signoff. Would everyone agree that new command-line scripts must accept "--help" (and include relevant help, of course) ? Should it also be the standard that command-line scripts should not perform any action unless an argument is passed? Anything else? -- Owen -- Web Developer Athens County Public Libraries http://www.myacpl.org

7 7