old.kete.net.nz

rake zebra:load_initial_records
(in /home/kete/apps/kapkete/releases/20091015004405)
rake aborted!
uninitialized constant SITE_NAME

and with --trace:

kete@kete-prod:~/apps/kapkete/current$ rake zebra:load_initial_records --trace
(in /home/kete/apps/kapkete/releases/20091015004405)
** Invoke zebra:load_initial_records (first_time)
** Invoke environment (first_time)
** Execute environment
** Execute zebra:load_initial_records
rake aborted!
uninitialized constant SITE_NAME
/home/kete/apps/kapkete/releases/20091015004405/vendor/rails/activesupport/lib/active_support/dependencies.rb:443:in `load_missing_constant'
/home/kete/apps/kapkete/releases/20091015004405/vendor/rails/activesupport/lib/active_support/dependencies.rb:80:in `const_missing'
/home/kete/apps/kapkete/releases/20091015004405/vendor/rails/activesupport/lib/active_support/dependencies.rb:92:in `const_missing'
(erb):4

Zebra Issues

Looks like the cause was that that publicserver and privateserver values in zerba/conf/kete-zebra-servers.xml had the protocol and host, but were missing port numbers. A quick look at the zoom dbs through Kete, adding the ports to the file manually, and restarting zebra allowed the rebuild to continue.

SITE_NAME Issues

The SITE_NAME and SITE_URL values were slightly incorrect (site_name did not match the domain name, and site url was not needed in this case). Correcting these two, and restarting the application solved this issue.

My zebra/conf/kete-zebra-servers.xml file is actually a /zebradb/conf/kete-zebra-servers.xml.template file. I've copied that template file to the xml only file, but I'm not sure the instructions are still valid - i don't see a protocol and host listing for the private or public servers. Have I missed a step here? I'm getting the SITE_NAME error when I run the zebra:load_initial_records on the command line and a

There was a problem with the rebuild: Connect failed (10000) localhost:9901

error when I do a rebuild of the database via the website. I have checked and both ports 9901 and 9902 are open for me. I'm stumped - any help would be appreciated!

First, a question. Where were you at in the process just before you encountered this issue?

> My zebra/conf/kete-zebra-servers.xml file is actually a
> /zebradb/conf/kete-zebra-servers.xml.template file. I've copied that template
> file to the xml only file, but I'm not sure the instructions are still valid

There is a handy tool that will set up your zebra/conf/kete-zebra-servers.xml for you based on the template (but substituting in real values for variable names) called 'rake zebra:set_ports'.

HOWEVER, during a normal installation you shouldn't have to call this task directly. It should be called behind the scenes for you during the Configuration step.

That's why I'm asking where you are at in the installation. If you have done the installation through the configuration step, Zebra should be properly configured and the zebra/conf/kete-zebra-servers.xml file should exist. If it doesn't exist, something has indeed gone wrong.

A common cause of these type of issues is that some steps were run as root when they should have been run as the kete unix user and therefore permissions are causing issues. I would make sure that everything is owned by the kete user under your app's root directory.

I have the site up and running and went through all of the web-based config steps until I hit the last one that asked me to index the server to bring in the starter Kete content. That process seemed to time out and so now I have a message that says

500 Error!

Oops! Zebra failed to connect to get the information you requested.

This could be because Zebra is not started, or is currently offline for maintenance.

Searching for the error message I got when I tried to re-index the database (after doing a rake zebra:start with no errors) was when I stumbled upon this page.

Let me know if you need anything else!!

Oh - I did check and everything under my history directory (my app) looks like it's owned by the kete user I created in the beginning of this process, so that shouldn't be an issue? I *think* I moved between root and kete to install as directed...

Another preliminary question. Are you running another instance of Kete on this host? If so, it is important that each Kete site's Zebra's port numbers are different than any other Kete site's Zebra's port numbers. You can definitely have issues is you point one Kete site at another's Zebra.

Ok assuming that port number conflict is not the issue, for whatever reason, the Zebra configuration steps failed to run during the web based Configuration. I'll fill you in on how to work around this, but keep in mind that the fact that it failed may be an indication another problem somewhere. It could also possibly be that there is a bug in Kete for configuration.

First, get the port numbers and password for your Zebra configuration (what you added during the search engine configuration step). Here's how (from command line in app's root directory):

script/console # add 'production' after command if not in development mode
ZoomDb.all.each { |z| p "database: #{z.database_name} - port: #{z.port} - password: #{z.zoom_password}" } # note the results 
exit 

The public and private Zebra databases have the same password, so you only need to note that once, but it is important what port number corresponds to what database (public or private).

Now that you have the Zebra settings, here's what I would do from the command line in the app's root directory (assuming development environment, if production, add RAILS_ENV=production before each command):

rake zebra:stop
rake zebra:set_keteaccess ZEBRA_PASSWORD=your_zebra_password # run this, then check that zebradb/keteaccess matches, if so, proceed
rm zebradb/conf/kete-zebra-servers.xml
rake zebra:set_ports PUBLIC_PORT=your_public_port PRIVATE_PORT=your_private_port
rake zebra:init #clears the zebra databases, probably not necessary, but doing to make sure consistent data rake zebra:start # check log/zebra.log, it should end with two lines that start with "Adding dynamic listener" and each line should have either public or private port, assuming it does, proceed
rake zebra:load_initial_records

Restart the backgroundrb process as well. Now you should be able to go back to the site's /site/search/setup_rebuild webpage and rebuild the Zebra search databases.

Tags: Zebra, Troubleshooting, rebuild search

Thanks for this - I did end up going back and redoing the entire installation just to make sure I had the permissions right and all. This is the only thing running on this server - I fired up an Amazon instance just for it! After I did the full reinstall and web-based configuration I got the exact same issue. It's not connecting to itself, somehow. After reinstall, I ran through the instructions you have below and when I ran the load_initial_records command I got this:

 Error while adding record to public zebra instance (Connect failed (10000) localhost:9901).
 Error while adding record to private zebra instance (Connect failed (10000) localhost:9902)

Which is different - I didn't get that far last time - but those port numbers are correct. A bit farther up in the instructions (the previous one, I think, zebra:init) the log file said:

root@localhost:/home/kete/apps/history# RAILS_ENV=production rake zebra:init
(in /home/kete/apps/history)
21:23:45-12/12 zebraidx(28171) [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
21:23:45-12/12 zebraidx(28171) [log] config ../conf/zebra-public.cfg
21:23:45-12/12 zebraidx(28171) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
21:23:45-12/12 zebraidx(28171) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
21:23:45-12/12 zebraidx(28171) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
21:23:45-12/12 zebraidx(28171) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
21:23:45-12/12 zebraidx(28171) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
21:23:45-12/12 zebraidx(28171) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
21:23:45-12/12 zebraidx(28171) [log] enabling shadow spec=./shadow:4G
21:23:45-12/12 zebraidx(28171) [log] cache_fname = ./shadow/cache
21:23:45-12/12 zebraidx(28171) [log] zebra_stop: 0.01 0.00 0.00
21:23:45-12/12 zebraidx(28173) [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
21:23:45-12/12 zebraidx(28173) [log] config ../conf/zebra-public.cfg
21:23:45-12/12 zebraidx(28173) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
21:23:45-12/12 zebraidx(28173) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
21:23:45-12/12 zebraidx(28173) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
21:23:45-12/12 zebraidx(28173) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
21:23:45-12/12 zebraidx(28173) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
21:23:45-12/12 zebraidx(28173) [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
21:23:45-12/12 zebraidx(28173) [log] enabling shadow spec=./shadow:4G
21:23:45-12/12 zebraidx(28173) [log] cache_fname = ./shadow/cache
21:23:45-12/12 zebraidx(28173) [log] zebra_stop: 0.00 0.00 0.00

It looks like zebra is stopping automatically? Thanks

Tags: Zebra, Troubleshooting, rebuild search

So we know the problem is that Zebra is dying for some reason. I.e. that is actually your problem.

Is there a "syslog" for your Amazon instance? I would tail that log while running 'rake zebra:start' and see if anything suspicious crops up.

What version of Zebra are you running?

Tags: Zebra, Troubleshooting, rebuild search

I'm not sure exactly how to find the zebra version, but the zebra.cfg file shows the module path as idzebra-2.0, which looks like the path for the newer version, not the 1.3 version, right? As for the syslog - nothing is written when I run the rake zebra:start to that log. The apache error log, though shows:

root@localhost:/home/kete/apps/history# rake zebra:start & tail -f /var/log/apache2/error.log
[2] 8264
        from /opt/ruby-enterprise-1.8.7-2012.02/lib/ruby/gems/1.8/gems/passenger-3.0.18/lib/phusion_passenger/spawn_manager.rb:275:in `handle_spawn_application'
        from /opt/ruby-enterprise-1.8.7-2012.02/lib/ruby/gems/1.8/gems/passenger-3.0.18/lib/phusion_passenger/abstract_server.rb:357:in `__send__'
        from /opt/ruby-enterprise-1.8.7-2012.02/lib/ruby/gems/1.8/gems/passenger-3.0.18/lib/phusion_passenger/abstract_server.rb:357:in `server_main_loop'
        from /opt/ruby-enterprise-1.8.7-2012.02/lib/ruby/gems/1.8/gems/passenger-3.0.18/lib/phusion_passenger/abstract_server.rb:206:in `start_synchronously'
        from /opt/ruby-enterprise-1.8.7-2012.02/lib/ruby/gems/1.8/gems/passenger-3.0.18/helper-scripts/passenger-spawn-server:99
/home/kete/apps/history/vendor/rails/railties/lib/rails/gem_dependency.rb:119:Warning: Gem::Dependency#version_requirements is deprecated and will be removed on or after August 2010.  Use #requirement
/home/kete/apps/history/vendor/rails/railties/lib/rails/gem_dependency.rb:119:Warning: Gem::Dependency#version_requirements is deprecated and will be removed on or after August 2010.  Use #requirement
/home/kete/apps/history/vendor/rails/railties/lib/rails/gem_dependency.rb:119:Warning: Gem::Dependency#version_requirements is deprecated and will be removed on or after August 2010.  Use #requirement
/home/kete/apps/history/vendor/rails/railties/lib/rails/gem_dependency.rb:119:Warning: Gem::Dependency#version_requirements is deprecated and will be removed on or after August 2010.  Use #requirement
/home/kete/apps/history/vendor/rails/railties/lib/rails/gem_dependency.rb:119:Warning: Gem::Dependency#version_requirements is deprecated and will be removed on or after August 2010.  Use #requirement
(in /home/kete/apps/history)

With no mention of starting zebra that I can see. Those errors above *may* be related to the fact that I tried to use an IP as the site name yesterday when I was trying different things. I ended up having to get into the database to change the value to get Rails working again. Anyway, I'm just not sure enough of what I'm seeing in the logs to know where to go next.

Tags: Zebra, Troubleshooting, rebuild search

Is the Amazon instance Debian or Ubuntu based? You can run "dpkg --list idzebra*" to get a list of the Zebra packages. Might also be good to do dpkg --list yaz*.

In the past I have had an issue with Zebra installation where, atl least for the Zebra start up process, "localhost" was not resolving to the host machine as it should. If I recall correctly, that didn't stop Zebra from running though, you just couldn't connect to it.

rake zebra:start is just a wrapper for a call to the zebrasrv command. If you drop down to that level, we may be able to isloate things more.

First though, can you post the zebradb/conf/kete-zebra-servers.xml please? It would be good to rule that out as the cause.

The Apache warnings can be ignored in this case.

Tags: Zebra, Troubleshooting, rebuild search

This is on a Debian Squeeze Instance. The kete-zebra-servers.xml file is below:

<?xml version="1.0" encoding="UTF-8"?>

<!-- /home/walter/projects/kete-collection/zebradb/conf/kete-zebra-servers.xml $

<yazgfs>

<!-- [scheme:]host[:port][/databaseName] -->
        <!-- scheme: tcp, ssl, unix, http, sru? -->
<!-- Includes the skeletons of unix socket in addition to the fully configured $
     unix socket wasn't found to be any faster, but keeping around for future r$
<!-- private is limited to ONLY localhost for security, since passwords are pas$
<listen id="privateserver" >tcp:localhost:9902</listen>
<!-- <listen id="privateserver-unix" >unix:/home/kete/apps/history/tmp/sockets/$
<!-- this should catch with the @ wildcard, both localhost and any outside ip i$
<listen id="publicserver" >tcp:localhost:9901</listen>
<!-- <listen id="publicserver-unix" >unix:/home/kete/apps/history/tmp/sockets/z$

<!-- PRIVATE KETE RECORDS -->
<server id="privateserver"  listenref="privateserver">
 <directory>./private</directory>
    <config>../conf/zebra-private.cfg</config>
    <!-- location of CQL-to-PQF file
         see: http://www.indexdata.com/yaz/doc/tools.tkl#tools.cql.map -->
</server>

<!-- PRIVATE KETE SERVER UNIX SOCKET
<server id="privateserver-unix"  listenref="privateserver-unix">
    <directory>./private</directory>
    <config>../conf/zebra-private.cfg</config>
</server> -->

<!-- PUBLIC KETE SERVER -->
<server id="publicserver"  listenref="publicserver">
    <directory>./public</directory>
    <config>../conf/zebra-public.cfg</config>
    <!-- location of CQL-to-PQF file
         see: http://www.indexdata.com/yaz/doc/tools.tkl#tools.cql.map -->
</server>

<!-- PUBLIC KETE SERVER UNIX SOCKET
<server id="publicserver-unix"  listenref="publicserver-unix">
    <directory>./public</directory>
    <config>../conf/zebra-public.cfg</config>
</server>  -->
</yazgfs>

Well - the results of the list idzebra command shows an issue, I think?

 dpkg --list idzebra
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name           Version        Description
+++-==============-==============-============================================
un  idzebra        <none>         (no description available)

No version listed - so maybe I screwed up the installation? Yaz seems to be ok:

dpkg --list yaz
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name           Version        Description
+++-==============-==============-============================================
ii  yaz            4.2.47-1indexd utilities for the Z39.50 toolkit

I'll hold off on doing anything until I hear back. Thank you SOOOOO much for all your help here!

Tags: Zebra, Troubleshooting, rebuild search

>  dpkg --list idzebra
> Desired=Unknown/Install/Remove/Purge/Hold
> | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
> |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
> ||/ Name          
> Version        Description
> +++-==============-==============-============================================
> un  idzebra       
> <none>         (no description
> available)
>
> No version listed - so maybe I screwed up the installation? Yaz seems to be ok:

Try it with the wildcard: dpkg --list idzebra*

Note the asterisk.

I don't think you could get as far as you have without having Zebra installed at all.

Tags: Zebra, Troubleshooting, rebuild search

Ah! A better response this time!

dpkg --list idzebra*
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name           Version        Description
+++-==============-==============-============================================
un  idzebra        <none>         (no description available)
ii  idzebra-2.0    2.0.53-1indexd virtual package with most components
ii  idzebra-2.0-co 2.0.53-1indexd common files
ii  idzebra-2.0-do 2.0.53-1indexd documentation
ii  idzebra-2.0-ut 2.0.53-1indexd utility programs including indexer and serve

Thanks!

Tags: Zebra, Troubleshooting, rebuild search

Do you know how to use IRC? It might make sense to work through this together there.

Essentially the next step I would do is 'cd zebradb' and then 'zebrasrv -f conf/kete-zebra-servers.xml' and see what the standard output says. It might not give us anything new, but perhaps it will provide some clues that weren't obvious with just looking at the log before.

Tags: Zebra, Troubleshooting, rebuild search

I do, but it's been a while since I've used it... I have mIRC loaded on my machine and an interview scheduled in an hour - but until then, let me know what server to connect to and I'll give it a shot!

Tags: Zebra, Troubleshooting, rebuild search

And while we're waiting - the output from your suggestion is:

zebrasrv -f conf/kete-zebra-servers.xml
20:07:52-13/12 [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
20:07:52-13/12 [log] config ../conf/zebra-private.cfg
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
20:07:52-13/12 [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
20:07:52-13/12 [log] config ../conf/zebra-public.cfg
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
20:07:52-13/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
20:07:52-13/12 [server] Adding dynamic listener on tcp:localhost:9902 id=1
20:07:52-13/12 [fatal] Failed to bind to tcp:localhost:9902 [Address already in use]
20:07:52-13/12 [server] Adding dynamic listener on tcp:localhost:9901 id=2
20:07:52-13/12 [fatal] Failed to bind to tcp:localhost:9901 [Address already in use]
20:07:52-13/12 [server] Adding dynamic listener on tcp:@:9999 id=0
20:07:52-13/12 [fatal] Failed to bind to tcp:@:9999 [Address already in use]

Tags: Zebra, Troubleshooting, rebuild search

Interesting. Do a 'rake zebra:stop' and rerun the same command please.

Tags: Zebra, Troubleshooting, rebuild search

You'll need to stop the zebrasrv process itself first, of course.

After run you run 'rake zebra:stop', confirm that no other Zebra instances are running by doing 'ps auxw | grep zebra'. You shouldn't get ay processes listed. If you do, you'll want to kill them.

Tags: Zebra, Troubleshooting, rebuild search

> I do, but it's been a while since I've used it... I have mIRC loaded on my
> machine and an interview scheduled in an hour - but until then, let me know what
> server to connect to and I'll give it a shot!

I've probably blown my capacity for helping out on this for the day. Let's schedule something for tomorrow. When works for you? Later in the day (your time, I assuming you are in US) works best for me.

Tags: Zebra, Troubleshooting, rebuild search

I'm pretty much blown out for the day as well. Tomorrow afternoon (1-3 my time) I'm going to be booked, but maybe later? 4pm CST? I'll run those commands tomorrow morning and post them for you then.

Tags: Zebra, Troubleshooting, rebuild search

4 pm CST tomorrow should work for me. I'll meet you in #kete channel on the irc.katipo.co.nz IRC server.

Talk to you then.

Tags: Zebra, Troubleshooting, rebuild search

Ok - rake zebra:stop gives me an error message about bad ps syntax, perhaps a bogus '-'
running the ps command right after that shows two zebrasrv processes running. I killed them manually, reran the ps command and confirmed they were stopped. I then re-ran the command and got:

zebrasrv -f conf/kete-zebra-servers.xml
14:12:59-14/12 [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
14:12:59-14/12 [log] config ../conf/zebra-private.cfg
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
14:12:59-14/12 [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
14:12:59-14/12 [log] config ../conf/zebra-public.cfg
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
14:12:59-14/12 [server] Adding dynamic listener on tcp:localhost:9902 id=1
14:12:59-14/12 [server] Adding dynamic listener on tcp:localhost:9901 id=2

This looked promising, so I re-loaded my history.nekls.org page to see if the 500 error would go away. It didn't, but I did get this in the logs.

14:13:09-14/12 zebrasrv(1) [session] Session - OK 1 tcp:::1 26401
14:13:09-14/12 zebrasrv(1) [request] Auth idPass kete -
14:13:09-14/12 zebrasrv(1) [request] Init OK - ID:81 Name:ZOOM-C/YAZ Version:3.0.53 735402374c0243bc385b4b1a7598ff5214109c44
14:13:09-14/12 zebrasrv(1) [request] Search public ERROR 109 1 1+0 RPN @attrset Bib-1 @attr 5=3 @attr 1=12 :Topic:
14:13:17-14/12 zebrasrv(2) [session] Session - OK 2 tcp:::1 26403
14:13:17-14/12 zebrasrv(2) [request] Auth idPass kete -
14:13:17-14/12 zebrasrv(2) [request] Init OK - ID:81 Name:ZOOM-C/YAZ Version:3.0.53 735402374c0243bc385b4b1a7598ff5214109c44
14:13:17-14/12 zebrasrv(2) [request] Search public ERROR 109 1 1+0 RPN @attrset Bib-1 @attr 5=3 @attr 1=12 :Topic:

I will now go log into the site and try to re-run the search index, see if that does anything...
Logging into the site produced these lines in the zebrasrv log:
14:15:54-14/12 zebrasrv(1) [session] Connection closed by client
14:15:54-14/12 zebrasrv(1) [session] Connection closed - end of session
14:15:54-14/12 zebrasrv(2) [session] Connection closed by client
14:15:54-14/12 zebrasrv(2) [session] Connection closed - end of session
14:16:10-14/12 zebrasrv(3) [session] Connection closed by client
14:16:10-14/12 zebrasrv(3) [session] Connection closed - end of session

Logging into the site and attempting to rebuild search databases, however, didn't write anything to the log. There is a connection when I'm not logged in but it fails when I am? I'm all kinds of confused. I should be back by 3:30 this afternoon, so 4pm my time in the IRC channel is doable!

Thanks,

Tags: Zebra, Troubleshooting, rebuild search

Just FYI - Walter was wonderful and fixed my problem (being a bit too liberal with the use of Root - I had started my zebra & backgroundrb scripts with it and it was causing some issues. Now the site is up, the content is there and I am forever grateful to Walter for his patience with me!

Problem solved!

Tags: Zebra, Troubleshooting, rebuild search

Ok - rake zebra:stop gives me an error message about bad ps syntax, perhaps a bogus '-'
running the ps command right after that shows two zebrasrv processes running. I killed them manually, reran the ps command and confirmed they were stopped. I then re-ran the command and got:

zebrasrv -f conf/kete-zebra-servers.xml
14:12:59-14/12 [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
14:12:59-14/12 [log] config ../conf/zebra-private.cfg
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
14:12:59-14/12 [log] zebra_start 2.0.53 46289c8b713538e12f49160e8e6eb79e131e8a20
14:12:59-14/12 [log] config ../conf/zebra-public.cfg
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-alvis.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-dom.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-regx.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-grs-marc.so
14:12:59-14/12 [log] Loaded filter module /usr/lib/idzebra-2.0/modules/mod-text.so
14:12:59-14/12 [server] Adding dynamic listener on tcp:localhost:9902 id=1
14:12:59-14/12 [server] Adding dynamic listener on tcp:localhost:9901 id=2

This looked promising, so I re-loaded my history.nekls.org page to see if the 500 error would go away. It didn't, but I did get this in the logs.

14:13:09-14/12 zebrasrv(1) [session] Session - OK 1 tcp:::1 26401
14:13:09-14/12 zebrasrv(1) [request] Auth idPass kete -
14:13:09-14/12 zebrasrv(1) [request] Init OK - ID:81 Name:ZOOM-C/YAZ Version:3.0.53 735402374c0243bc385b4b1a7598ff5214109c44
14:13:09-14/12 zebrasrv(1) [request] Search public ERROR 109 1 1+0 RPN @attrset Bib-1 @attr 5=3 @attr 1=12 :Topic:
14:13:17-14/12 zebrasrv(2) [session] Session - OK 2 tcp:::1 26403
14:13:17-14/12 zebrasrv(2) [request] Auth idPass kete -
14:13:17-14/12 zebrasrv(2) [request] Init OK - ID:81 Name:ZOOM-C/YAZ Version:3.0.53 735402374c0243bc385b4b1a7598ff5214109c44
14:13:17-14/12 zebrasrv(2) [request] Search public ERROR 109 1 1+0 RPN @attrset Bib-1 @attr 5=3 @attr 1=12 :Topic:

I will now go log into the site and try to re-run the search index, see if that does anything...
Logging into the site produced these lines in the zebrasrv log:
14:15:54-14/12 zebrasrv(1) [session] Connection closed by client
14:15:54-14/12 zebrasrv(1) [session] Connection closed - end of session
14:15:54-14/12 zebrasrv(2) [session] Connection closed by client
14:15:54-14/12 zebrasrv(2) [session] Connection closed - end of session
14:16:10-14/12 zebrasrv(3) [session] Connection closed by client
14:16:10-14/12 zebrasrv(3) [session] Connection closed - end of session

Logging into the site and attempting to rebuild search databases, however, didn't write anything to the log. There is a connection when I'm not logged in but it fails when I am? I'm all kinds of confused. I should be back by 3:30 this afternoon, so 4pm my time in the IRC channel is doable!

Thanks,

Tags: Zebra, Troubleshooting, rebuild search