Open Source Integrated Library System

Evergreen on IRC

#evergreen Logs for Tuesday, July 12th, 2011

< Monday, July 11th, 2011Raw Log FileWednesday, July 13th, 2011 >
#TimeNickMessage
#01:32:10dbs has quit IRC
#03:55:09foocraft_ has joined #evergreen
#04:20:41dbwells_ has joined #evergreen
#04:22:46tater has joined #evergreen
#04:29:36dbwells has quit IRC
#04:29:37mtate has quit IRC
#05:05:05artunit_ has joined #evergreen
#05:06:36artunit has quit IRC
#05:06:42artunit_ is now known as artunit
#05:33:20bjwebb has joined #evergreen
#05:33:20bjwebb has joined #evergreen
#05:37:05artunit_ has joined #evergreen
#05:38:18artunit has quit IRC
#05:38:27artunit_ is now known as artunit
#05:59:42miz^Gecko has joined #evergreen
#06:09:40miz^Gecko has left #evergreen
#06:20:15artunit_ has joined #evergreen
#06:22:26artunit has quit IRC
#06:22:37artunit_ is now known as artunit
#07:51:58foocraft_ has quit IRC
#07:56:03mrpeters-isl has joined #evergreen
#08:29:18mrpeters-islmorning all -- I'm having some trouble debugging survey editor interface. I can create/edit/delete/admin survey with all users in Administrator group, however, I cannot add questions/answers. The interface just seems to die when it tries to load anything after "Questions & Answers". Doesn't appear to be permissions related, since a user with "EVERYTHING" permission also cannot add Questions
#08:29:18mrpeters-isl& Answers. Any thoughts?
#08:30:07tsbereI recall having issues like that
#08:31:11tsbereHmmm. My dev box is borked.
#08:31:24tsbere initiates a full reinstall of evergreen with a pre-rigged alias
#08:31:32AaronZ-PLS has joined #evergreen
#08:32:43tsberemrpeters-isl: Gimme a min or two, I apparently can't use my dev box due to borked master. :(
#08:33:14mrpeters-islAt some point while we were on 1.6 users were able to add questions, because there are a few in there from 2010. Looking back through commits on "survey" but not seeing much that'd apply to > 1.6.
#08:33:18mrpeters-islyeah, no worries tsbere
#08:35:16tsberemrpeters-isl: I can only add questions/answers if the survey starts *in the future*.
#08:35:27mrpeters-islaha!
#08:35:28tsbereIf the survey starts, say, today I can't add questions/answers
#08:36:27mrpeters-islthat did the trick
#08:37:03dbwells_ is now known as dbwells
#08:46:00tsbereeeevil: You around?
#08:46:34csharpokay - back to the reporter setup (reporter process running on DB server, but needing to connect to a server with opensrf running)
#08:47:18csharpI've scoured opensrf.xml and opensrf_core.xml on the reporter/DB server and do not see where the server would be told to use another server's opensrf
#08:47:40csharpI can't find this on my production example either
#08:47:42tsberecsharp: You just need it to use the same jabber server.
#08:48:21csharp tries to process that suggestion...
#08:48:32tsbereThat may require (depending on your setup) a change to your hosts file to point public.localhost and private.localhost to the network IP of the server you want it to talk to
#08:49:09csharpah...
#08:50:17tsbere@later tell eeevil http://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=6343d40803699e3d287ba8adec979dd99c4687ef seems to say authority merge broke master's DB scripts. Specifically the second added line in 011.schema.authority.sql has an extra comma on the end.
#08:50:17pinesol_greentsbere: The operation succeeded.
#08:51:22kmlussier has joined #evergreen
#08:52:40Dyrcona has joined #evergreen
#08:53:29csharptsbere++ # that did it!
#08:54:00mrpeters-isl'cookieManager' is undefined with IE on rresult.js is pretty indicative of the user being behind some kind of filter, or pretty restrictive cookies settings yeah?
#08:54:47csharpthe hosts file was not correct anyway, but adding the entry for the brick I was connecting to (along with the opensrf_core.xml and opensrf.xml that I copied from the brick) worked to get the reporter process going
#08:56:19csharp was thinking that the configuration was at the EG level, not at the OS level
#08:57:10tsbereWell, some of us decided to use actual DNS names for our public/private and thus don't have to play with hosts at all ;)
#08:57:42csharpone more question... our prod reporter server runs clark_kent.pl as root - any reasons why it would be root vs. opensrf?
#08:58:07tsberePermissions issues with network shared locations is about all I can think of
#08:58:18csharptsbere: I bet that's it
#09:04:09csharpis there a nicer way to stop clark than killing it and removing the lock?
#09:05:34tsbereDid you try kill without -9?
#09:05:48Dyrconacsharp: writing a script to kill it and remove the lock for you?
#09:06:14csharpyeah - I found the process, killed it (no -9) and removed the lockfile
#09:06:20csharpDyrcona: understood ;-)
#09:06:20Dyrconacsharp: may i suggest naming it kryptonite? or jimmy_olsen?
#09:06:33csharpheh
#09:06:52csharpdidn't Batman kill Superman at one point? (may have that turned around)
#09:07:00Dyrcona fires up his Spin Doctors mp3s. The first track is Jimmy Olsen's Blues.
#09:07:20mrpeters-islupdating "source" on a bunch of biblio.record_entry rows, dies out with "ERROR: error from Perl function "maintain_control_numbers": Wide character in subroutine entry at /usr/local/share/perl/5.10.1/MARC/Charset/Table.pm line 96". Any thoughts on what this is dying on?
#09:07:42gmcharlt has quit IRC
#09:07:53Dyrconamrpeters-isl: old version of MARC::File::XML/MARC::Record?
#09:08:17Dyrconamrpeters-isl: record are UTF-8 but the LDR doesn't reflect that?
#09:08:41kmlussier has quit IRC
#09:08:47mrpeters-isltotally possible
#09:09:22Meliss has joined #evergreen
#09:09:22kmlussier has joined #evergreen
#09:10:15csharp seeks clark_kent.pl invocation examples
#09:10:20mrpeters-islMARC-XML-0.93 latest?
#09:10:55csharpah - on the install page!
#09:11:05csharphelps to read the friendly manual
#09:16:46Dyrconastarting and stopping clark_kent is something that could be added to oils_ctl-improvements branch.
#09:17:11Dyrcona doesn't think that has gone into master, yet.
#09:17:24dbs has joined #evergreen
#09:17:24dbs has joined #evergreen
#09:17:54csharpDyrcona: I would think at least adding the invocation documentation to the script header would be an improvement ;-)
#09:18:04csharp likes inline documentation
#09:19:21Dyrcona believes all comments lie.
#09:20:00csharpha!
#09:23:42DyrconaThe code is its own (and only) documentation. All other documentation is subject to error, omission, and prevarication.
#09:24:08dbsThat's true only for projects that don't really want users.
#09:24:25csharpI understand from the dev perspective, but I would not want to have to decode perl just to know how to run a simple daemon script
#09:24:48dbsLots of other options out there for people to run systems with better QA, documentation, etc.
#09:27:37Dyrconadbs: In my experience that is true in all cases for both open and closed source projects.
#09:28:02mrpeters-islDyrcona: so, FWIW same error after updating to 0.93
#09:28:29DyrconaHaving done this professionally for over 14 years, I've seen a lot of comments, documents, etc. that were just plain wrong.
#09:28:57Dyrconamrpeters-isl: check that the leader on the record that is choking has 'a' in position 9.
#09:29:08mrpeters-islwell, i'm not sure how i can find out what the record is
#09:29:13dbsDyrcona: Sure. But to not even try to provide documentation is pathetic
#09:29:28mrpeters-isli just have a list of id's i'm updating "source" on
#09:29:30dbsOr to argue that it's not worthwhile providing documentation
#09:29:31jenny1 has joined #evergreen
#09:29:39Dyrconadbs: I didn't say not to try. I just said it lies.
#09:29:44Dyrcona:)
#09:30:03Dyrconadbs: Me, personally, I'll believe the code and not the documentation.
#09:30:51DyrconaIt also helps if people write clear, concise code. In 14 years, I've found that is also a futile battle.
#09:30:53dbsDyrcona: Fair enough. I believe that many places don't want to employ people to spend time unravelling mysteries of undocumented code and unexpected interactions.
#09:31:25Dyrcona"undocumented code and unexpected interactions" is 95% of debugging. :)
#09:31:57dbsDyrcona: No shit. As I said, most places just want to run a system, they don't want to invest in intensive debugging to make the thing work.
#09:32:39dbs needs to put together a full-day training agenda for tomorrow a new set of libraries coming on board Conifer. Top to bottom in one day, that should be fun. In French to boot
#09:32:54DyrconaBonne chance! :)
#09:33:42dbs is perhaps a bit weary of spending time debugging and handling problem reports and has an overly negative perspective this morning
#09:39:00gmcharlt has joined #evergreen
#09:41:30eeeviltsbere: I'll correct the baseline schema, thanks for spotting that
#09:43:38tsbereeeevil: Also, the hold targeter is broken.
#09:44:17eeeviltsbere: due to authority-control-sets, or because of something you were chasing the other day?
#09:44:31tsbereeeevil: I am blaming commit 20735bf13454c3d9bdcc958805228c85ec7af031
#09:44:42tsbere has been fighting with this for an hour or so
#09:46:09eeeviltsbere: error? because that hasn't shown a problem otherwise...
#09:46:54tsbereeeevil: No error, but holds haven't targeted, *at all*, since we installed that commit. I think part_maps isn't actually defined at that point in the file UNLESS it is a type P.
#09:47:34tsbere doesn't believe we have many, if any, part holds in the system right now
#09:48:04dbs is going to sign off before losing any lingering respect he might have by going off the deepend on a testing / QA / time efficiency rant
#09:48:06dbs has quit IRC
#09:48:41tsbereeeevil: I reverted that commit and, hey, holds target again.
#09:49:16eeevil looks
#09:50:36eeevil$_ is a CDBI object, which gets that method from OpenILS/Application/Storage/CDBI.pm via asset::copy->has_many( part_maps => 'asset::copy_part_map' );
#09:51:27eeeviltsbere: any chance you have a stale OpenILS/Application/Storage/CDBI.pm? (or related)
#09:51:38tsbereDoubtful.
#09:51:43Dyrcona wonders if fm_IDL.xml is up to date, or what eeevil just said.
#09:54:57eeeviltsbere: or that the schema changes for part holds isn't up to date?
#09:57:24tsbereeeevil: Well, reverting that commit fixes targeting individual holds and non-parallel holds targeting for us for now. Beyond that, I am not sure.
#09:57:33eeevilok ... off to a meeting
#10:05:42yboston has joined #evergreen
#10:18:44leed has quit IRC
#10:23:25leed has joined #evergreen
#10:32:03gmcharlt has quit IRC
#10:33:47gmcharlt has joined #evergreen
#10:44:20rri has joined #evergreen
#10:59:43tater is now known as mtate
#11:00:56Tstallis has joined #evergreen
#11:06:20joseph_ has joined #evergreen
#11:13:44youdonotexist has joined #evergreen
#11:22:14kuys has quit IRC
#11:23:41Tstallis has quit IRC
#11:51:37tsbereAnyone have a block of code that would allow me to use CDBI via a standalone script?
#11:52:11artunit_ has joined #evergreen
#11:53:23artunit has quit IRC
#11:53:35artunit_ is now known as artunit
#12:04:58foocraft has joined #evergreen
#12:13:10tsbereeeevil: AHA! Found an error! Processing of hold failed: Not an ARRAY reference at /usr/local/share/perl/5.10.1/OpenILS/Application/Storage/Publisher/action.pm line 1274.
#12:20:38tsbere has no clue how to fix it, but at least he found the error
#12:20:53phasefx thinks he encountered that or something like it before, and it was fixed
#12:23:35Dyrconaphasefx: this has to do with checking part holds, so it is relatively new code.
#12:23:44phasefxright
#12:24:02artunit_ has joined #evergreen
#12:24:12mmorgan has joined #evergreen
#12:24:35tsbereDyrcona: phasefx signed off on the offending commit, according to git. So he has likely seen it ;)
#12:24:49Dyrcona:)
#12:25:11artunit has quit IRC
#12:25:24artunit_ is now known as artunit
#12:25:24phasefxProcessing of hold failed: Not an ARRAY reference at /
#12:25:28phasefxusr/local/share/perl/5.10.1/OpenILS/Application/Storage/Publisher/action.pm line 1274.. yeah, same one
#12:25:57phasefxtsbere: you have http://git.evergreen-ils.org/?p=evergreen/equinox.git;a=commit;h=e11ca376239610ced428750584f1f0196dc8bcd0 ?
#12:27:56phasefxhrmm, master has that
#12:28:02phasefxso something else maybe
#12:28:37tsberephasefx: Yes, I have that. For the record, line 1274 is treating a Class::DBI::Iterator or such as an array.
#12:35:01kmlussier has quit IRC
#12:36:42kmlussier has joined #evergreen
#13:00:51jenny1 has quit IRC
#13:03:35Melissa has joined #evergreen
#13:10:51jeffis there an authoritative git repo for SIPServer?
#13:11:05tsberegit.evergreen-ils.org's, I think.
#13:11:07tsbereAt this point
#13:13:37jefffresh clone and "diff -ur -x .git SIPServer SIPServer-mvlc" is clean. success.
#13:14:21jeff(of course, mvlc has remote branches that differ)
#13:16:13tsbereBeing lazy, I rigged git.evergreen-ils.org to update several of MVLC's repos whenever someone commits ;)
#13:17:44jeffso our public pc management software doesn't do guest passes the way we want, so we're just going to use a second sip server with a simple backend to provide the needed features.
#13:17:47tsberephasefx: On a different note, open-ils.storage.action.hold_request.targetable_holds.id_list doesn't want to return anything for me. No clue why, can't find errors for it.
#13:18:05tsberejeff: What is "the way you want"?
#13:19:45jeffguest passes that are only valid for specific groups of workstations, guest passes that can be printed on something like perforated business card sheets, that don't expire unless used...
#13:20:17jeffconfigurable as to if guest cards expire on use or can be used multiple times in the day they are first used, etc.
#13:21:46tsbereOk. Have fun with that. :P
#13:22:15tsbere hasn't done work to make most of that work, well, maybe the "specific groups of workstations" with the stat cat export, but not auto-expire on use
#13:22:15jeffwilldo.
#13:22:47foocraft has quit IRC
#13:23:03jeffspecific groups of workstations with the stat cats exposed via sip2 would work well if any of these users will have an account in the ILS
#13:23:13jeffbut that's not intended.
#13:24:57tsbereYou should be able to use the extra fields support in SIPServer for that, though
#13:26:58jeffhow are you thinking?
#13:27:09jeff"hi, i'm a patron and would like to use your computers."
#13:27:12jeffer,
#13:27:15jeffs/patron/tourist/
#13:27:31tsbereWithout changing SIPServer (just giving it a backend) you can have the backend provide extra fields to spit out in the response packet.
#13:27:32jeff"sure, you can use this guest pass to get on one of those express stations over there!"
#13:28:48jeffah, i thought you were saying something else entirely.
#13:28:59jeffyes, i don't think i'll need to make any changes to SIPServer to make this happen.
#13:29:39jeffjust a new backend -- somehow i thought you were suggesting that extra_fields would enable us to do guest passes with an Evergreen backend and no guest users in the ILS.
#13:41:43eeeviltsbere: I can certainly fix that ... but @{...} should be setting wantarray for that call, which is disturbing
#13:42:50tsbere is trying to figure out why he can effectively run targetable_holds from action.pm manually in this perl file and get results, but isn't getting results when he calls it with srfsh
#13:43:31eeeviltsbere: if you're in a testing mood, mind chaning that line to: ($hold->hold_type ne 'P' ? $_->part_maps->count == 0 : 1)
#13:44:18tsbereeeevil: Don't have my dev system fully functional, although I might be able to swap to using our training system for some testing.
#13:45:41tsberespeaking of training....
#13:46:05tsbere kicks something, as the training system is fine with the open-ils.storage.action.hold_request.targetable_holds.id_list call, but production isn't
#13:47:04tsbereStupid thing is supposedly running the same code and a copy of the database, and it is working. <_<
#13:51:12tsbereeeevil: That change looks to work in our training system.
#13:52:34tsbereeeevil: Disclaimer: I attempted to target one hold, got [] back, made the change, restarted everything, attempted to target the same hold, got an object back. And that was the extent of my testing.
#13:54:52eeeviltsbere: thanks ... will poke here before committing
#14:01:30jenny1 has joined #evergreen
#15:01:25gmcharltmrpeters-isl: something for you to pass along to Bob - the '\' in many of the TCNs of the bib he's trying to load ... is being interpreting as a regex metacharacter by maintain_901
#15:01:31gmcharltleading to the fun you're seeing
#15:01:41gmcharltI'll be posting a patch shortly
#15:28:25mrpeters-isl10-4 passing along
#15:31:56joseph_ has quit IRC
#15:45:43kmlussier has quit IRC
#16:00:14Meliss has quit IRC
#16:11:45collum has quit IRC
#16:20:48jenny2 has joined #evergreen
#16:24:06jenny1 has quit IRC
#16:47:09joseph_ has joined #evergreen
#16:47:26rri has quit IRC
#17:04:24mmorgan has left #evergreen
#17:15:09artunit_ has joined #evergreen
#17:17:09artunit has quit IRC
#17:17:16artunit_ is now known as artunit
#17:21:45gdunbar has quit IRC
#17:31:33Dyrcona has quit IRC
#17:41:03phasefxgrabbing 0576
#17:41:59gmcharlt has quit IRC
#17:42:20jenny2 has left #evergreen
#18:20:01yboston has quit IRC
#18:45:00gmcharlt has joined #evergreen
#18:54:39youdonotexist has quit IRC
#19:10:02artunit_ has joined #evergreen
#19:12:15artunit has quit IRC
#19:12:24artunit_ is now known as artunit
#19:18:35rri has joined #evergreen
#19:23:03Melissa has left #evergreen
#19:30:18ernieSimuro has joined #evergreen
#19:31:30ernieSimuroI created an invalid workstation associated wityh an IP address, does anyone know how to remove this invalid entry
#19:32:53ernieSimuro has quit IRC
#19:53:08joseph_ has quit IRC
#20:14:10artunit_ has joined #evergreen
#20:15:15artunit has quit IRC
#20:15:17artunit_ is now known as artunit
#20:19:48Dyrcona has joined #evergreen
#20:44:17bjwebb has quit IRC
#20:45:04artunit_ has joined #evergreen
#20:47:33artunit has quit IRC
#20:47:41artunit_ is now known as artunit
#21:40:01artunit_ has joined #evergreen
#21:42:06artunit has quit IRC
#21:42:08artunit_ is now known as artunit
#22:07:41rri has quit IRC
#22:09:38rri has joined #evergreen
#22:15:01rri has quit IRC
#22:16:58rri has joined #evergreen
#22:29:36rri has quit IRC
#22:31:25rri has joined #evergreen
#22:48:41rri has quit IRC
< Monday, July 11th, 2011Raw Log FileWednesday, July 13th, 2011 >