| # | Time | Nick | Message |
|---|
| # | 09:07:40 | bshum has joined #evergreen |
| # | 11:39:42 | elene has joined #evergreen |
| # | 11:40:50 | elene | hello everyone |
| # | 11:41:12 | elene | i have a quick question about evergreen 2.0.1 installation |
| # | 11:41:18 | elene | would be great if someone could help |
| # | 11:42:58 | elene | I just installed 2.0.1 on Ubuntu 10.04 for testing purposes and it seems to load the OPAC ( http://opac2.library.ac.ge ) |
| # | 11:43:16 | elene | but when I try to log in from Staff Client as the administrator, I get the following error: |
| # | 11:43:22 | elene | method=open-ils.auth.authenticate.init |
| # | 11:43:22 | elene | params=["isiadmin"] |
| # | 11:43:23 | elene | THROWN: |
| # | 11:43:23 | elene | Network Failure: status = 200 |
| # | 11:43:23 | elene | service=open-ils.auth&method=open-ils.auth.authenticate.init¶m=%22isiadmin%22 |
| # | 11:43:23 | elene | STATUS: |
| # | 11:43:23 | elene | 200 |
| # | 11:43:56 | elene | Could anyone help me find out what the error means? |
| # | 11:43:58 | elene | thanks! |
| # | 11:44:21 | phasefx | elene: a status 200 as an error is very weird. hrmm |
| # | 11:44:49 | elene | hi phasefx, thanks for ur help |
| # | 11:45:04 | elene | would u happen to know a log file where I could take a look? |
| # | 11:46:40 | phasefx | have you restarted apache since starting/restarting opensrf? (oy, my isp is going up and down, so |
| # | 11:47:52 | elene | yes, I have tried restarting everything and everything seems to be loading without problems |
| # | 11:48:41 | elene | another strange thing is that when visiting http://opac2.library.ac.ge it seems to work, but the system will not allow me to choose languages from drop-down menu :) |
| # | 11:49:08 | phasefx | restarting apache last in the order of events? |
| # | 11:51:25 | elene | I am not sure what it means :) could you tell me the right order of restarting everything? |
| # | 11:52:14 | phasefx | http://evergreen-ils.org/dokuwiki/doku.php?id=troubleshooting:checking_for_errors |
| # | 11:55:55 | phasefx | apache is always restarted last, after an osrf_ctl.sh start_all or restart_all |
| # | 11:56:57 | elene | thank you. yes, I did try the steps outlined at the wiki article, except fishing for the errors |
| # | 11:57:07 | elene | as I am not familiar with error log files :) |
| # | 11:57:40 | elene | apart from checking error logs, everything seems to be working fine and no errors are reported when starting or stopping the services |
| # | 11:58:12 | elene | perhaps you could advise me which log file would be responsible for the error i mentioned? |
| # | 12:05:08 | phasefx_ | I'd grep for ERR and Fail with all the logs in /openils/var/log/. In this case, I'd expect to see something in gateway.log or activity.log (I forget which would be configured for a wiki install), and also in the Apache logs (probably /var/log/apache/) |
| # | 12:05:43 | elene | i see. thanks a lot. i will take a look now :) |
| # | 12:20:30 | elene | I deleted logs, restarted server and chcked osrfsys.log which includes this error message, many times. |
| # | 12:20:31 | elene | [2011-02-20 17:18:27] /usr/sbin/apache2 [ERR :23882:EX.pm:66:] Exception: OpenSR |
| # | 12:20:31 | elene | F::EX::Session 2011-02-20T17:18:27 OpenSRF::Utils::SettingsClient /usr/local/sha |
| # | 12:20:31 | elene | re/perl/5.10.1/OpenSRF/Utils/SettingsClient.pm:103 Session Error: router@private |
| # | 12:20:31 | elene | .localhost/opensrf.settings IS NOT CONNECTED TO THE NETWORK!!! |
| # | 12:21:26 | elene | gateway.log also has some errors: |
| # | 12:21:26 | elene | gateway 2011-02-20 17:18:27 [WARN:23882:transport_session.c:783:] Error 401 in I |
| # | 12:21:27 | elene | Q packet: not authorized |
| # | 12:21:51 | elene | and then these two errors: |
| # | 12:21:55 | elene | gateway 2011-02-20 17:18:27 [WARN:23882:transport_session.c:685:] Error connecti |
| # | 12:21:55 | elene | ng to jabber |
| # | 12:21:56 | elene | gateway 2011-02-20 17:18:27 [INFO:23882:transport_session.c:667:] Received <erro |
| # | 12:21:56 | elene | r> message with type auth and code 401 |
| # | 12:24:24 | elene | does this mean something went wrong with the installation? :) |
| # | 12:26:22 | bshum | elene: Indeed, that looks like something went wrong with your ejabberd |
| # | 12:27:28 | elene | I see. would you suggest reinstalling ejabberd, or deleting entire system and reinstalling server and evergreen from scratch? |
| # | 12:27:39 | bshum | When you installed OpenSRF, did you change all the parts of the /etc/ejabberd/ejabberd.cfg file? |
| # | 12:28:20 | bshum | Also (this came up recently) but did you use any special characters in the ejabberdctl registration part? |
| # | 12:28:30 | bshum | Other than numbers and letters |
| # | 12:28:59 | bshum | The changes to ejabberd.cfg are noted in step 6 of the OpenSRF 1.6 install |
| # | 12:29:13 | bshum | Sometimes people miss the other bullet items towards the end |
| # | 12:29:19 | elene | thanks, bshum! I think I followed the guide and changed all the parts. my evergreen is 2.0.1 |
| # | 12:29:21 | bshum | For max_stanza_size, maxrate, mod_offline |
| # | 12:30:25 | elene | I have max_stanza_size, 2000000 |
| # | 12:30:43 | bshum | Sounds like you got those good |
| # | 12:30:43 | elene | maxrate 500000 |
| # | 12:31:03 | bshum | Did you use any special characters when registering ejabberd users in step 8? |
| # | 12:31:31 | elene | let me quickly check step 8 |
| # | 12:31:35 | bshum | Sure. |
| # | 12:31:41 | elene | i gather the step is same in 1.6 and 2.0 guide |
| # | 12:32:03 | bshum | Step 8 of the OpenSRF install guide for 1.6.2 version of OpenSRF: http://www.open-ils.org/dokuwiki/doku.php?id=opensrf:1.6:install |
| # | 12:32:33 | bshum | The Evergreen install guide is something else. |
| # | 12:33:05 | elene | i see. i am checking now |
| # | 12:34:02 | elene | I see. yes, I did follow the step and registered all the four users |
| # | 12:34:10 | elene | and used simple passwords without special chars :) |
| # | 12:35:09 | bshum | Okay, good to check |
| # | 12:36:18 | elene | thanks. would you advise me to reinstall opensrf (if thats possible), or reinstall everything from scratch? :) |
| # | 12:37:54 | bshum | Well, if you reinstall opensrf then you have to reinstall Evergreen as well. Starting from scratch is certainly an option too, but a bit drastic. |
| # | 12:38:18 | bshum | I'm not sure exactly, but 401 seems like an authentication error. |
| # | 12:38:52 | bshum | Maybe your opensrf_core.xml file has the wrong passwords in it? |
| # | 12:39:52 | bshum | (unlikely, now I'm just guessing) |
| # | 12:40:00 | bshum hates Sundays |
| # | 12:40:14 | elene | loool |
| # | 12:40:26 | elene | is there a way for me to check the passwords? |
| # | 12:40:42 | elene | like try logging in from terminal somehow |
| # | 12:42:05 | bshum | Well, you would just open up the /openils/conf/opensrf_core.xml file using a text editor in the terminal and looking for the parts that define <passwd> or <password> |
| # | 12:42:38 | elene | great thanks i'll do just that |
| # | 12:50:48 | elene | I did. They all seem to be correct :( |
| # | 12:50:54 | elene | should I check router log files? |
| # | 12:51:41 | elene | router private log has this error in it: |
| # | 12:51:44 | elene | router 2011-02-20 17:18:05 [WARN:26609:transport_session.c:642:] Received <strea |
| # | 12:51:44 | elene | m:error> message from Jabber server |
| # | 12:51:44 | elene | router 2011-02-20 17:18:05 [WARN:26609:osrf_router.c:229:] Top level select call |
| # | 12:51:44 | elene | failed with errno 9: Bad file descriptor |
| # | 12:51:44 | elene | router 2011-02-20 17:18:05 [WARN:26609:socket_bundle.c:553:] _socket_send(): Err |
| # | 12:51:45 | elene | or sending data with return -1 |
| # | 12:51:45 | elene | router 2011-02-20 17:18:05 [WARN:26609:socket_bundle.c:554:] Last Sys Error: Bad |
| # | 12:51:45 | elene | file descriptor |
| # | 12:52:15 | elene | and router public log includes this error: |
| # | 12:52:17 | elene | router 2011-02-20 17:18:05 [WARN:26608:transport_session.c:642:] Received <strea |
| # | 12:52:17 | elene | m:error> message from Jabber server |
| # | 12:52:17 | elene | router 2011-02-20 17:18:05 [WARN:26608:osrf_router.c:229:] Top level select call |
| # | 12:52:17 | elene | failed with errno 9: Bad file descriptor |
| # | 12:52:18 | elene | router 2011-02-20 17:18:05 [WARN:26608:socket_bundle.c:553:] _socket_send(): Err |
| # | 12:52:18 | elene | or sending data with return -1 |
| # | 12:52:18 | elene | router 2011-02-20 17:18:05 [WARN:26608:socket_bundle.c:554:] Last Sys Error: Bad |
| # | 12:52:19 | elene | file descriptor |
| # | 12:55:51 | phasefx | elene: if we fail to help you here today (very few folks around), you may want to send a copy of your logs/configs (with sensitive passwords scrubbed) to the OPEN-ILS-DEV mailing list |
| # | 12:56:39 | elene | thank you phasefx. I will that. |
| # | 12:57:04 | elene | phasefx, bshum, thanks very much for your help! |
| # | 12:57:08 | phasefx | minimal logs as suggested in the checking_for_errors document |
| # | 12:57:33 | elene | I see |
| # | 12:57:38 | elene | allright |
| # | 12:57:59 | phasefx | just mean, stop everything, erase the logs, then start stuff back up |
| # | 12:58:59 | phasefx | in this case, I bet you'll get errors with start_router (assuming you replace start_all with start_router, start_perl, and start_c), so could stop there |
| # | 13:03:28 | elene has quit IRC |
| # | 13:05:58 | bshum | Pfft, lesson learned... don't uninstall memcached from a drone until *after* Evergreen is all installed... |
| # | 13:06:17 | bshum | :( |
| # | 13:06:34 | _bott_ has quit IRC |
| # | 13:08:32 | elene has joined #evergreen |
| # | 13:09:18 | elene | hello again, I was disconnected, so did not have a chance to say goodbye :) |
| # | 13:09:33 | elene | so thanks again for your help and I will be posting the logs on the list |
| # | 13:09:59 | rsinger has quit IRC |
| # | 13:11:21 | elene has left #evergreen |
| # | 13:42:50 | jennam has quit IRC |
| # | 15:13:02 | bshum | Hmm |
| # | 15:13:18 | bshum | Just a guess, and if someone is around, let me know |
| # | 15:13:27 | bshum | But if we're using an NFS share for /openils/var |
| # | 15:13:37 | bshum | It seems like I should change the opensrf.xml file |
| # | 15:13:46 | bshum | So that the unix.pids go somewhere else |
| # | 15:13:59 | bshum | Otherwise, the NFS share locks up usage of to only one computer? |
| # | 15:23:22 | bshum | Cool, guess that does work. |
| # | 15:29:10 | jeff | bshum: yes, there was some discussion about that last week -- closely related, if not exactly the same scenario |
| # | 15:29:37 | bshum | jeff: Neat, I'll check back in the logs to see if I can find something about it. |
| # | 15:31:49 | bshum | Thanks! |
| # | 15:33:53 | jeff | 'welcome |
| # | 15:34:00 | jeff | how did the upgrade go? |
| # | 15:34:34 | bshum | jeff: Still underway; been a long weekend :) |
| # | 15:34:54 | bshum | Based on that last tidbit, I think I'll have our bricks rockin' and rolling shortly. |
| # | 15:35:16 | eeevil | bshum: db upgrade went smoothly? |
| # | 15:35:23 | jeff | excellent. continued good luck, and enjoy! |
| # | 15:35:24 | bshum | eeevil: As far as we can tell. |
| # | 15:35:41 | eeevil | :) |
| # | 15:35:41 | bshum | eeevil: There was one minor hiccup, but I think it turned out to be a bad sequence on our end. |
| # | 15:36:17 | bshum | eeevil: Bumping the sequence higher allowed us to upgrade to 2.0+ very smoothly. |
| # | 15:36:17 | eeevil | good good ... lots of folks will end up using that, so that it worked for you is great news |
| # | 15:36:38 | bshum | Next step, 9.0, right? :) |
| # | 15:36:52 | eeevil | bshum: mind making a note of that sequence on the wiki and the -dev list? |
| # | 15:37:08 | bshum | eeevil: Sure, I'd be happy to. For us, it was probably an accident. |
| # | 15:37:22 | bshum | We use action/trigger notices |
| # | 15:37:31 | bshum | And it looks like one of the new samples was bumping against ours |
| # | 15:37:39 | bshum | Because we hadn't done the sequence right |
| # | 15:37:51 | bshum | Direct DB editing fail :( |
| # | 15:38:18 | eeevil | bshum: that, (9.0) thankfully, should be pretty easy ... if you have a machine with lots of drive space for testing, I'd be interested to know if pg_upgrade (binary/in-place major version upgrade) works well |
| # | 15:38:39 | bshum | eeevil: It's already on my to-do list for next week. :D |
| # | 15:38:42 | eeevil | rad |
| # | 16:07:27 | dbs | bshum: been working on (ugh) physical stuff but discussion you and I had re NFS and openils/var was on Feb 15 |
| # | 16:07:52 | dbs has nice pink bits of insulation decorating his scalp |
| # | 16:08:14 | bshum | Oh gosh |
| # | 16:08:19 | bshum | That really was *me* wasn't it? :S |
| # | 16:08:57 | bshum | Hmm |
| # | 16:09:03 | bshum | That makes perfect sense. |
| # | 16:09:13 | bshum | I just didn't realize it applied to the opensrf.xml config files directly. |
| # | 16:09:27 | bshum | It seemed that the oils_brick example would have taken care of it. |
| # | 16:14:00 | bshum | Thanks dbs |
| # | 16:14:33 | dbs | bshum: happy to be of some help on the weekend o' stress :) |
| # | 16:18:30 | bshum | Oh shoot, you DID say it. |
| # | 16:18:40 | bshum | 2011-02-15T11:23:00 <dbs> /openils/var/run doesn't work if /openils = NFS |
| # | 16:18:55 | bshum | Go figure, I just completely missed the meaning behind that |
| # | 16:19:21 | dbs | bshum: I probably didn't express it clearly enough |
| # | 16:19:39 | bshum | Well, no worries, it's all coming together! |
| # | 16:19:45 | dbs | supa! |
| # | 16:32:41 | bshum1 has joined #evergreen |
| # | 16:32:47 | bshum has quit IRC |
| # | 16:32:49 | bshum1 is now known as bshum |
| # | 16:51:41 | elene has joined #evergreen |
| # | 16:53:24 | elene | hi everyone. i am in the process of installing opensrf 1.6.2 and seem to have a problem registering ejabberd users (section 8 i of the guide) |
| # | 16:53:46 | elene | I issue the command |
| # | 16:53:55 | elene | ejabberdctl register router private.localhost testpassword |
| # | 16:54:09 | elene | and I get an error |
| # | 16:54:10 | elene | Can't register user router@private.localhost at node ejabberd@opac2: not_allowed |
| # | 16:54:55 | elene | my ejabberd.cfg includes this line: {auth_method, internal}. |
| # | 16:55:14 | elene | Would anyone have a quick hint, or should I post my question to the mailing list? |
| # | 16:55:15 | elene | Thanks |
| # | 16:57:25 | dbs | elene: what does your "hosts" line in ejabberd.cfg say? |
| # | 16:57:45 | elene | hi dan, many thanks for answering me |
| # | 16:57:51 | elene | i will check quickly |
| # | 16:58:13 | elene | it says: {hosts, ["localhost"]}. |
| # | 16:58:17 | dbs | ah |
| # | 16:58:34 | dbs | you're getting ahead of yourself, then; you need to edit it first to say: {hosts, ["localhost", "public.localhost", "private.localhost"]}. |
| # | 16:58:47 | dbs | and then you'll need to restart ejabberd |
| # | 16:59:03 | elene | oops i seem to have missed that step :) |
| # | 16:59:11 | elene | thank you so much :) |
| # | 16:59:14 | jeff | dbs++ |
| # | 17:00:14 | dbs | Might as well make sure you've got all of step 6 on http://evergreen-ils.org/dokuwiki/doku.php?id=opensrf:1.6:install done before moving on :) |
| # | 17:00:32 | elene | i see. thank you. |
| # | 17:01:03 | dbs goes back into the basement to get some more labour done |
| # | 17:07:33 | dbs | but not before checking out http://jshint.com/ for possible integration into the buildbot |
| # | 17:13:37 | jeff | what, no "xulrunner" Environment setting? :) |
| # | 17:53:15 | bshum has quit IRC |
| # | 18:50:27 | elene | im afraid i have one more question :) |
| # | 18:50:45 | elene | i seem to have finished EG 2.0.1 installation |
| # | 18:51:02 | elene | but when i issue autogen |
| # | 18:51:16 | elene | i get this error |
| # | 18:51:24 | elene | Updating Evergreen organization tree and IDL using '/openils/conf/opensrf_core.xml' |
| # | 18:51:24 | elene | Updating fieldmapper |
| # | 18:51:24 | elene | Empty filename at /openils/lib/perl5/OpenILS/Utils/Fieldmapper.pm line 197 |
| # | 18:52:12 | elene | please note that this is the first time i am running the autogen script in this installation |
| # | 18:52:17 | elene | thanks |
| # | 18:55:19 | elene | this is how my Fieldmapper.pm file looks like at that section: |
| # | 18:55:46 | elene | 194 # parse the IDL ... |
| # | 18:55:46 | elene | 195 my $parser = XML::LibXML->new(); |
| # | 18:55:46 | elene | 196 my $file = $args{IDL} || OpenSRF::Utils::SettingsClient->new->config_value( 'IDL' ); |
| # | 19:04:14 | elene | I have seen Dan's reply on mailing list on similar question at http://list.georgialibraries.org/pipermail/open-ils-general/2010-February/002544.html |
| # | 19:05:23 | elene | suggesting that the problem lies in having old opensrf_core.xml and opensrf.xml files, and that opensrf.xml. file should include an <IDL> element. |
| # | 19:06:09 | elene | however, taking a look at the installation guides of eg 2.0.1. and opensrf 1.6.x, I can not file any reference to <IDL> element |
| # | 19:34:05 | jeff | elene: the <IDL> element isn't mentioned in the install docs, it's just that the OpenSRF tarball includes an example opensrf config that does not include that element, while the Evergreen tarball includes an example opensrf config that DOES contain that element. |
| # | 19:36:00 | elene | thank you, jeff. I skipped copying the EG example files |
| # | 19:36:21 | elene | as I though I had already done it when installing OpenSRF |
| # | 19:37:28 | elene | I have now overwritten the files and reentered passwords in opensrf_core.xml |
| # | 19:37:45 | elene | now, when issuing autogen |
| # | 19:37:55 | elene | it does seem to work a bit |
| # | 19:38:09 | elene | Updating Evergreen organization tree and IDL using '/openils/conf/opensrf_core.xml' |
| # | 19:38:09 | elene | Updating fieldmapper |
| # | 19:38:09 | elene | Updating web_fieldmapper |
| # | 19:39:15 | elene | Jeff, could you kindly advise me where I could read about how I should configure opensrf.xml file? thanks. |
| # | 19:46:29 | dbs | elene: the eg_db_config.pl script generally does all of the configuration of opensrf.xml you need for a basic Evergreen installation script |
| # | 19:49:09 | elene | thank you, dan. i have copied the xml files again from eg, and rerung the eg_db_config.pl script again |
| # | 19:49:27 | elene | but when running autogen, I still get this error |
| # | 19:49:45 | elene | Updating Evergreen organization tree and IDL using '/openils/conf/opensrf_core.xml'Updating fieldmapper |
| # | 19:49:45 | elene | Updating web_fieldmapper |
| # | 19:49:45 | elene | Updating OrgTree |
| # | 19:50:27 | elene | So I thought it was because my opensrf.xml was not configured correctly :) |
| # | 20:02:22 | dbs | I ... don't see an error there |
| # | 20:05:54 | elene | oops |
| # | 20:06:10 | elene | this is the error i am getting when issuing autogen? |
| # | 20:06:21 | elene | .Exception: OpenSRF::EX::Session 2011-02-21T00:48:51 OpenSRF::Transport /usr/local/share/perl/5.10.1/OpenSRF/Transport.pm:92 Session Error: mailto:router@private.localhost/open-ils.cstore IS NOT CONNECTED TO THE NETWORK!!! |
| # | 20:17:09 | elene | ejabberd status: |
| # | 20:17:18 | elene | ejabberdctl status |
| # | 20:17:18 | elene | The node ejabberd@opac2 is started with status: started |
| # | 20:17:18 | elene | ejabberd 2.1.2 is running in that node |
| # | 20:17:59 | dbs | elene, |
| # | 20:18:32 | dbs | elene: sorry - hmm - how did you start the OpenSRF services? Did you use the "-l" flag for osrf_ctl.sh? |
| # | 20:18:43 | elene | yes i did |
| # | 20:19:08 | elene | and i access the domain by: opac2.library.ac.ge |
| # | 20:19:15 | elene | it has a fqdn |
| # | 20:19:43 | elene | but i have not entered it anywhere, except in /etc/hosts file |
| # | 20:19:56 | dbs | yeah, that should be okay |
| # | 20:20:03 | elene | good thanks |
| # | 20:20:19 | dbs | So we need to figure out why open-ils.cstore isn't starting (or staying connected) |
| # | 20:20:29 | elene | i see |
| # | 20:21:02 | dbs | this would probably be a good time to post your /openils/var/log/osrfsys.log somewhere |
| # | 20:21:13 | elene | i see just a second |
| # | 20:24:23 | dbs | @later tell eeevil Hey, so RFC3339 requires a colon in the timezone offset between hour and minute; I guess that's why Dojo requires it (and ECMAScript 5 provides ISO parsing based on RFC3339). My curiosity is finally satisfied :) |
| # | 20:24:23 | pinesol | dbs: The operation succeeded. |
| # | 20:27:28 | elene | ok, i pasted the last lines of the osrfsys.log |
| # | 20:27:28 | elene | http://paste.lisp.org/display/119857 |
| # | 20:30:03 | dbs | Hmm, "Could not open TCP socket to Jabber server: IO::Socket::INET: connect: Connection refused" sounds like ejabberd wasn't running for a while? It isn't clear what order things are happening in. Can you: |
| # | 20:30:24 | dbs | 1) Shut down the opensrf services with: "osrf_ctl.sh -l -a stop_all" |
| # | 20:30:36 | elene | ok |
| # | 20:31:00 | elene | i did |
| # | 20:31:08 | elene | osrf_ctl.sh -l -a stop_all |
| # | 20:31:08 | dbs | 2) Clean out any process ID files (PID files) with the command: "rm -f /openils/var/run/*pid" |
| # | 20:31:08 | elene | Stopping OpenSRF C process 7316... |
| # | 20:31:08 | elene | kill: 209: No such process |
| # | 20:31:27 | elene | i did |
| # | 20:31:45 | dbs | 3) Remove any log files with the command "rm -f /openils/var/log/*.log" |
| # | 20:32:03 | elene | i did |
| # | 20:32:24 | dbs | 4) Start the opensrf router with "osrf_ctl.sh -l -a start_router" |
| # | 20:32:44 | dbs | then paste the contents of the new /openils/var/log/osrfsys.log file |
| # | 20:32:46 | elene | i did |
| # | 20:33:15 | elene | there is no error log |
| # | 20:33:53 | elene | log directory contains only router-public.log and router-private.log |
| # | 20:34:34 | elene | should I try osrf_ctl.sh -l -a start_all and see if logs are created? |
| # | 20:35:10 | dbs | no |
| # | 20:35:18 | dbs | we're just going step by step |
| # | 20:35:25 | elene | i see |
| # | 20:35:36 | dbs | okay, now start the perl services: "osrf_ctl.sh -l -a start_perl" |
| # | 20:35:49 | dbs | and paste the contents of the new /openils/var/log/osrfsys.log file |
| # | 20:37:23 | elene | ok this did generate a huge osrfsys.log |
| # | 20:37:23 | elene | http://paste.lisp.org/display/119859 |
| # | 20:38:30 | elene | and also, lots of open-ils*unix.log files were created in log dir |
| # | 20:38:49 | elene | they all have this error in them: |
| # | 20:38:49 | elene | 2011/02/21-01:36:24 OpenSRF::UnixServer (type OpenSRF) starting! pid(17814) |
| # | 20:38:49 | elene | Binding to UNIX socket file /openils/var/lock/open-ils.ingest-unix.sock using SOCK_STREAM |
| # | 20:39:09 | elene | 2011/02/21-01:36:24 OpenSRF::UnixServer (type OpenSRF) starting! pid(17814) |
| # | 20:39:18 | elene | Binding to UNIX socket file /openils/var/lock/open-ils.ingest-unix.sock using SOCK_STREAM |
| # | 20:39:20 | dbs | those aren't errors, they're just INFO |
| # | 20:39:30 | dbs | but "router@private.localhost/opensrf.settings IS NOT CONNECTED TO THE NETWORK!!!" is definitely a sign of a problem |
| # | 20:39:38 | elene | i see |
| # | 20:40:12 | elene | i have a fresh ubuntu 10.04 and tried to follow the guides step by step |
| # | 20:40:47 | dbs | that's okay |
| # | 20:40:54 | elene | though i may have done something wrong :( |
| # | 20:41:19 | dbs | might be a problem with the password for the router@private.localhost ejabberd user not being specified correctly in opensrf_core.xml |
| # | 20:41:40 | dbs | I would suggest double-checking opensrf_core.xml |
| # | 20:41:45 | elene | ok, i will reregister the user now now and reenter the pwd |
| # | 20:42:47 | dbs | if you try to reregister the user, don't be surprised if ejabberd says "Hey, this user has already been registered!" |
| # | 20:45:03 | elene | i checked and the password seems to be correct. I tried reregistering the user, but it tells me that User router@private.localhost already registered at node ejabberd@opac2 |
| # | 20:45:37 | elene | should I delete the user and try reregistering? |
| # | 20:45:41 | dbs | no |
| # | 20:46:00 | dbs | might be a good time to try the settings-tester.pl script |
| # | 20:46:53 | elene | ok, as root or as opensrf? |
| # | 20:46:54 | dbs | as the "opensrf" user, run "perl Open-ILS/src/support-scripts/settings-tester.pl > testing.out" and paste the contents of testing.out ? |
| # | 20:47:52 | elene | ok |
| # | 20:50:59 | elene | dbshttp://paste.lisp.org/display/119861 |
| # | 20:52:13 | elene | no errors :| |
| # | 20:52:16 | dbs | elene: weird. seems fine |
| # | 20:52:36 | elene | what if i just rebooted the server again? |
| # | 20:52:45 | elene | perhaps that could help? |
| # | 20:53:19 | elene | or perhaps i could paste the contents of the xml files? |
| # | 20:53:20 | dbs | nho |
| # | 20:53:23 | dbs | no no no |
| # | 20:53:32 | dbs | don't reboot the server |
| # | 20:53:38 | elene | ok |
| # | 20:54:42 | elene | should i paste opensrf.xml and opensrf_core.xml files? the passwords there are just generic testing passwords |
| # | 20:54:47 | dbs | Let's just go ahead and try starting up the OpenSRF C services: "osrf_ctl.sh -l -a start_c" |
| # | 20:54:52 | elene | ok |
| # | 20:55:09 | elene | ok i think it started |
| # | 20:55:15 | elene | should i check the logs now? |
| # | 20:55:27 | dbs | okay - yes, then open up osrfsys.log and jump to 2011-02-21 01:36:26 |
| # | 20:55:43 | dbs | and see if there's anything interesting after that |
| # | 20:56:24 | dbs | There will probably be many log entries for cstore, including some "errors" |
| # | 20:57:00 | dbs | maybe the best bet is just to paste the results of "grep ERR /openils/var/log/osrfsys.log" |
| # | 20:58:44 | elene | ok |
| # | 21:00:23 | elene | http://paste.lisp.org/display/119862 |
| # | 21:00:50 | elene | i think this is the same error we saw last time |
| # | 21:02:05 | dbs | yep, same timestamp. Okay, that's reasonably good |
| # | 21:02:22 | dbs | does "ps wax | grep open-ils.cstore" show anything? |
| # | 21:02:23 | elene | really? |
| # | 21:02:41 | elene | ps wax | grep open-ils.cstore |
| # | 21:02:41 | elene | 6039 ? S 0:00 OpenSRF Listener [open-ils.cstore] |
| # | 21:02:41 | elene | 6047 ? S 0:00 OpenSRF Drone [open-ils.cstore] |
| # | 21:02:53 | elene | oops, too long to paste here |
| # | 21:03:21 | elene | http://paste.lisp.org/display/119863 |
| # | 21:03:24 | elene | here it is |
| # | 21:04:59 | dbs | good |
| # | 21:05:10 | dbs | so what does "autogen.sh -u" do now? |
| # | 21:06:06 | elene | http://paste.lisp.org/display/119864 |
| # | 21:06:28 | elene | i think this is something new |
| # | 21:07:09 | dbs | hmm - okay, can you paste the last bits of osrfsys.log again then? |
| # | 21:07:21 | elene | ok |
| # | 21:08:14 | dbs | (that error basically says the ->gather() method call failed) |
| # | 21:09:29 | elene | ok here it comes |
| # | 21:09:30 | elene | http://paste.lisp.org/display/119865 |
| # | 21:11:03 | elene | hmm, is something wrong with the db scheme? |
| # | 21:11:49 | dbs | yeah, "ERROR: schema "config" does not exist" is bad |
| # | 21:12:41 | dbs | it sounds like the "eg_db_config.pl" command was either issued without the "--create-schema" option, or that it failed |
| # | 21:13:55 | elene | i see |
| # | 21:14:00 | elene | should i rerun it? |
| # | 21:14:14 | dbs | you should be able to "osrf_ctl.sh -l -a stop_all" and the "eg_db_config.pl --create-schema" to try creating it again, then "osrf_ctl.sh -l -a start_all; autogen.sh -u" to try again |
| # | 21:14:46 | elene | ok |
| # | 21:15:25 | elene | wait |
| # | 21:15:33 | elene | i accidentally ran |
| # | 21:15:40 | elene | ./autogen.sh -c /openils/conf/opensrf_core.xml -u |
| # | 21:15:44 | elene | and it said: |
| # | 21:15:53 | elene | autogen finished running, now run ./configure |
| # | 21:15:59 | elene has left #evergreen |
| # | 21:16:20 | elene has joined #evergreen |
| # | 21:16:31 | elene | does this change anything? |
| # | 21:18:07 | elene | this is what it said |
| # | 21:18:08 | elene | http://paste.lisp.org/display/119866 |
| # | 21:19:01 | dbs | that's a different (confusing) autogen.sh script |
| # | 21:19:13 | dbs | so no, it doesn't change anything |
| # | 21:20:50 | elene | i see. i ran the commands you suggested |
| # | 21:21:05 | elene | and create schema brought up lots of erros |
| # | 21:21:08 | elene | errors |
| # | 21:21:52 | elene | http://paste.lisp.org/display/119867 |
| # | 21:22:19 | dbs | Looks like you missed some of the Evergreen installation commands |
| # | 21:22:37 | elene | i see :( |
| # | 21:23:11 | elene | i did run this command: |
| # | 21:23:11 | elene | eg_db_config.pl |
| # | 21:23:20 | elene | i mean |
| # | 21:23:24 | elene | perl Open-ILS/src/support-scripts/eg_db_config.pl --update-config \ |
| # | 21:23:24 | elene | --service all --create-schema --create-offline \ |
| # | 21:23:24 | elene | --user <user> --password <password> --hostname <hostname> --port <port> \ |
| # | 21:23:25 | dbs | See step 2. II. in http://evergreen-ils.org/dokuwiki/doku.php?id=server:2.0:install |
| # | 21:23:44 | dbs | did you run "make -f Open-ILS/src/extras/Makefile.install install_pgsql_server_debs_84" as root when you were going through these steps? |
| # | 21:24:23 | dbs | and if so, then you also needed to run the commands in 3. "Issue the following commands on your PostgreSQL server, adjusting the path for the contrib repository..." |
| # | 21:24:26 | elene | yes, i believe i did |
| # | 21:25:06 | elene | yes, i also substituted the distro with ubuntu-lucid |
| # | 21:25:54 | dbs | the "ERROR: language "plpgsql" does not exist" tells me that either you didn't install the postgresql libs, or you didn't run the commands like "createlang plpgsql evergreen" |
| # | 21:26:16 | elene | oops, i did not adjust the path |
| # | 21:26:41 | elene | i left /usr/share/postgresql/8.4/contrib/ |
| # | 21:26:54 | elene | but i guess this might not have been wrong |
| # | 21:27:11 | elene | i see, it WAS wrong |
| # | 21:28:13 | elene | strangely, i can not find contrib dir. it is neither /usr/local/share/contrib/ nor |
| # | 21:30:50 | elene | ok i checked |
| # | 21:31:05 | dbs | http://packages.ubuntu.com/lucid/amd64/postgresql-contrib-8.4/filelist suggests that it should be /usr/share/postgresql/8.4/contrib/ |
| # | 21:31:14 | elene | yes it is |
| # | 21:31:26 | elene | so i did not have to adjust the path |
| # | 21:32:05 | dbs | so when you become the "postgres" user and run "createlang plpgsql evergreen" -- what happens? |
| # | 21:32:16 | elene | su let me see |
| # | 21:33:03 | elene | postgres@opac2:/root$ createlang plpgsql evergreen |
| # | 21:33:04 | elene | could not change directory to "/root" |
| # | 21:33:04 | elene | createlang: language "plpgsql" is already installed in database "evergreen" |
| # | 21:34:08 | dbs | weird. that doesn't seem to match up at all with what the error message said |
| # | 21:34:50 | elene | i probably did something wrong, otherwise other users would have complained of the same issues |
| # | 21:35:13 | dbs | your database name is "evergreen", right? |
| # | 21:35:29 | elene | yes, i did not change the command |
| # | 21:35:49 | elene | i ran createdb -T template0 -E UNICODE evergreen |
| # | 21:36:35 | elene | i think there was an automated installer script for very early version of evergreen. does something like this still exist for 2.0 or 1.6? |
| # | 21:36:54 | dbs | yes... |
| # | 21:37:27 | elene | perhaps i would have better luck with that :) |
| # | 21:37:47 | dbs | http://svn.open-ils.org/trac/ILS-Contrib/browser/ESI-Examples/trunk/tools/eg_dev_squeeze_installer - for Debian Squeeze |
| # | 21:38:44 | elene | great. i gather this will not work on ubuntu |
| # | 21:39:00 | dbs | probably not without some customization |
| # | 21:40:02 | dbs | I'm really confused as to why on the one hand PostgreSQL is saying that plpgsql is already installed in the "evergreen" database, yet when you try to create the schema it fails because plpgsql is not installed |
| # | 21:40:33 | dbs | is your database on the same server as the rest of evergreen? |
| # | 21:40:39 | elene | yes |
| # | 21:41:07 | elene | and i used fresh ubuntu installation on VPS |
| # | 21:45:43 | jeff | two postgresql installs, psql / createlang using one port while another is being specified for eg_db_config.pl? |
| # | 21:50:29 | dbs | jeff: nope |
| # | 21:50:40 | jeff | drat. |
| # | 21:50:49 | jeff | that's at least a known situation to deal with. |
| # | 21:51:01 | dbs | tsearch2 wasn't installed |
| # | 21:51:09 | dbs | and now... autogen.sh -u ran cleanly |
| # | 21:52:05 | dbs | restart apache, and... http://opac2.library.ac.ge comes up with the locale selector working |
| # | 21:55:23 | dbs tries to wrap his head around adding SRU support for NAF/SAF requests |
| # | 22:13:23 | jennam has joined #evergreen |
| # | 22:13:49 | jennam is now known as Guest47999 |
| # | 22:14:00 | Guest47999 is now known as jennam |
| # | 22:14:02 | jennam has joined #evergreen |
| # | 23:07:51 | elene has left #evergreen |
April 10-13