2011-01-19T00:01:43 aha, methinks authority.by_heading_and_thesaurus index is missing from upgrades 2011-01-19T00:01:46 * dbs goes to check 2011-01-19T00:04:25 nope, failed during the upgrade due to dupes - as the comments explain. right-o 2011-01-19T01:22:43 *** jamesrf has joined #evergreen 2011-01-19T01:25:40 *** AaronZ-PLS has quit IRC 2011-01-19T01:25:40 *** Ghidorah has quit IRC 2011-01-19T01:25:41 *** mrpeters-isl has quit IRC 2011-01-19T01:25:42 *** berick has quit IRC 2011-01-19T01:25:43 *** mtate has quit IRC 2011-01-19T01:25:43 *** brendan2 has quit IRC 2011-01-19T01:25:44 *** phasefx_ has quit IRC 2011-01-19T01:25:45 *** senator has quit IRC 2011-01-19T01:25:45 *** dbwells has quit IRC 2011-01-19T01:27:21 *** senator has joined #evergreen 2011-01-19T01:36:43 *** dbs has quit IRC 2011-01-19T01:40:05 *** bshum has joined #evergreen 2011-01-19T01:40:51 *** bshum has left #evergreen 2011-01-19T01:47:26 dbs pasted "public.naco_normalize() in 2.0 still needs some love?" at http://paste.lisp.org/display/118783 2011-01-19T02:14:28 *** jamesrf has quit IRC 2011-01-19T03:28:45 *** phasefx_ has joined #evergreen 2011-01-19T03:28:46 *** brendan2 has joined #evergreen 2011-01-19T03:28:46 *** mtate has joined #evergreen 2011-01-19T03:28:46 *** Ghidorah has joined #evergreen 2011-01-19T03:28:46 *** AaronZ-PLS has joined #evergreen 2011-01-19T03:28:46 *** mrpeters-isl has joined #evergreen 2011-01-19T03:28:46 *** dbwells has joined #evergreen 2011-01-19T03:28:46 *** berick has joined #evergreen 2011-01-19T07:27:16 *** alxp has joined #evergreen 2011-01-19T07:30:47 *** rickd_ has joined #evergreen 2011-01-19T07:32:02 *** sfortin has joined #evergreen 2011-01-19T08:00:02 *** collum has joined #evergreen 2011-01-19T08:37:05 @later tell dbs interesting; I'll see how this does in trunk; probably should backport the trunk naco_normalize() work to rel_2_0 2011-01-19T08:37:05 gmcharlt: The operation succeeded. 2011-01-19T08:39:46 *** dbs has joined #evergreen 2011-01-19T08:42:06 *** kmlussier has joined #evergreen 2011-01-19T08:55:14 *** Dyrcona has joined #evergreen 2011-01-19T09:08:51 *** Lowery has joined #evergreen 2011-01-19T09:09:38 *** Meliss has joined #evergreen 2011-01-19T09:11:30 *** lowery_ has joined #evergreen 2011-01-19T09:13:14 *** Lowery has quit IRC 2011-01-19T09:16:08 *** lowery_ has quit IRC 2011-01-19T09:16:46 *** Dyrcona has quit IRC 2011-01-19T09:20:22 *** Dyrcona has joined #evergreen 2011-01-19T09:26:51 *** sfortin has quit IRC 2011-01-19T09:27:07 eeevil: I think I've got one more fix for RC3 - a more robustified authority.normalize_heading() (it broke on some ugly authority MARCXML corner cases; switched to spi_prepare/spi_exec_query rather than going down the foolish path of trying to escape input myself) 2011-01-19T09:28:05 gmcharlt: I think I'm going to blame authority.normalize_heading() first, but I wouldn't be averse to backporting the robustified/authoritative public.naco_normalize() from trunk 2011-01-19T09:28:15 *** sfortin has joined #evergreen 2011-01-19T09:29:03 Just running our ~2 million authority record test set through the robustified authority.normalize_heading() to ensure that it survives our many flawed records, first :) 2011-01-19T09:31:09 dbs++ 2011-01-19T09:34:28 the one that broke its back was the 956,943 record with Foo, Bar\ - the trailing slash of death! could have just stripped that out, but... invoking the robustified authority.normalize_heading() on that record worked nicely, so confidence is high 2011-01-19T09:34:39 I'll commit it to trunk for now 2011-01-19T09:35:31 *** jenny has joined #evergreen 2011-01-19T09:35:43 ahh, yes ... was it eascaping/interpolation? (there were issues with trailing \s re call numbers before ... pain, indeed) 2011-01-19T09:35:52 eeevil: yessir 2011-01-19T09:35:52 before as in back in 1.2, IIRC 2011-01-19T09:42:14 calling 0476 2011-01-19T09:42:21 ring, ring 2011-01-19T09:49:39 is the hard_due_date_name_check constraint on chdd necessary? or in particular, its dislike of spaces in the name? 2011-01-19T09:50:15 I just copied the code from the duration/fine rules. Is it needed there? 2011-01-19T09:50:55 I wasn't sure if there was a scripts reason for no spaces, and I believe the hard due dates are available via scripts as well as DB 2011-01-19T09:52:43 *** eby has quit IRC 2011-01-19T09:54:09 * tsbere would offer gmcharlt more information, but his entire logic set on that piece has been dealt with at this point 2011-01-19T09:54:41 Oh, I think someone else did question that constraint when I submitted it, but kept it when I pointed out I had just copied from the other circ rules. Forget who 2011-01-19T09:55:02 *** eby has joined #evergreen 2011-01-19T09:57:04 gmcharlt: you and your questioning of existing practices! It's so.... healthy 2011-01-19T09:57:20 tsbere: looking at the script code, I don't see any particular reason why allowing spaces in the rule names would break anything 2011-01-19T09:58:57 gmcharlt: I was just going for "things named in circ rules/scripts/whatever are consistent in their naming rules" when I copied the constraint from the duration rules. I figured it would be more likely to be accepted that way, question naming rules later. 2011-01-19T09:59:01 I could see somebody possibly coming to grief if they used diacritics in the rule names and using leagcy scripts 2011-01-19T09:59:25 tsbere: sure - I understanding wanting to keep it consistent 2011-01-19T10:00:15 One benefit of no spaces in general is "you don't get leading/trailing spaces", though. 2011-01-19T10:00:43 Two rules that differ in number of trailing spaces might be a problem 2011-01-19T10:01:23 and I don't care about the details of the constraint, per se - more that (a) AutoGrid currently doesn't pass check constraint exceptions back in any way visible to the user and (b) even if it did, it would be nicer if the validatingtextbox widget knew the appropriate regex to apply if a constraint of that form was in force 2011-01-19T10:01:58 tsbere: agreed re leading/trailing whitespace 2011-01-19T10:02:25 For the most part? I thought the constraint was silly, and mentally questioned "why is this here?" on the duration/fine rule names to begin with. I just thought of the leading/trailing whitespace now, and that is fixable with trimming. 2011-01-19T10:03:54 *** tspindler has joined #evergreen 2011-01-19T10:05:53 *** sfortin has quit IRC 2011-01-19T10:06:19 *** sfortin has joined #evergreen 2011-01-19T10:06:47 gmcharlt: Could there be scripts out there that turn the various duration/fine rules into actual perl objects of some kind? The constraint rules seem very similar to naming rules in that regard. 2011-01-19T10:07:10 CREATE INDEX with the new authority.normalize_heading() succeeded on our 1.2 M authority records; if you guys want to take a look at r19201 before I backport it, I'll be happy to give you some time 2011-01-19T10:07:22 tsbere: none that I've seen, but I suppose anything is possible 2011-01-19T10:07:34 dbs: was just looking 2011-01-19T10:08:48 *** lowery has joined #evergreen 2011-01-19T10:09:04 dbs: looks great to me! 2011-01-19T10:09:24 dbs: agreed 2011-01-19T10:09:25 *** jenny has left #evergreen 2011-01-19T10:10:16 thanks, I appreciate the extra/skilled extra-skilled eyes at this stage of the 2.0 game 2011-01-19T10:10:29 hello all, i am trying to install opensrf on ubuntu and i keep getting /bin/bash: ./build: no such file or directory when i do the ./configure part. can anyone give me a hand please? 2011-01-19T10:10:54 lowery: any chance you forgot to run the prerequisites makefile? 2011-01-19T10:10:58 dbs: heh - and you get to pass some of blame on to us if it breaks things horribly ;) 2011-01-19T10:11:14 make -f Open-ILS/src/extras/Makefile.install 2011-01-19T10:11:22 remember to run the make file. had no problems with that step 2011-01-19T10:11:36 so, either ubuntu-lucid or something like that 2011-01-19T10:11:46 hmm, kind of seems like build-essentials may be missing? 2011-01-19T10:11:54 ubuntu-lucid 2011-01-19T10:11:55 gmcharlt: That is another (retroactive) reason to not mess with naming constraints in my patch. Not breaking things horribly ;) 2011-01-19T10:12:35 lowery: what directory are you in when you're running ./configure ? 2011-01-19T10:13:24 gmcharlt: we all hang together, right? 2011-01-19T10:13:27 i an in the opensrf directory 2011-01-19T10:13:49 lowery: can you type the results of "pwd" ? 2011-01-19T10:14:12 dbs: beats hanging separately 2011-01-19T10:14:28 pwd = /evergreen/opensrf 2011-01-19T10:14:32 re the no-spaces constraint, the original constraint was there because .-notation (foo.bar.baz) is easier to teach than associative array notation (foo["bar"]["baz"]) in the scripts (which were all that existed) and so it was used as a guide rail 2011-01-19T10:14:45 eeevil++ 2011-01-19T10:14:54 lowery: you have to cd into the OpenSRF source directory 2011-01-19T10:15:01 just to give some context 2011-01-19T10:15:09 "cd opensrf-1.6.2" or whatever 2011-01-19T10:15:40 so i would be in /evergreen/opensrf/src? 2011-01-19T10:15:42 eeevil: oh right - you were teaching librarians how to write javascript! 2011-01-19T10:16:15 dbs: silly me, yes :) 2011-01-19T10:17:03 lowery: your directory structure confuses me, most people would have downloaded the OpenSRF source tarball to /home//opensrf-1.6.2.tar.gz 2011-01-19T10:17:11 eeevil: heh. At least now we know. Can we give up on that silly idea in 2.1? ;) 2011-01-19T10:17:29 lowery: but you've created a new root /evergreen directory and unpacked the tarballs into that? 2011-01-19T10:17:30 since the library I'm working with wanted to be able to enter things like 'Branch Foo End of Winter Term' as the hdd name, unless there's any serious objection, I'm inclined to go ahead and drop the constraint in trunk 2011-01-19T10:17:41 lowery: wherever you find the NEWS, README, etc. for OpenSRF 2011-01-19T10:17:55 i created a directory top level called /evergreen downloaded opensrf-1.6.2 and renamed it to opensrf 2011-01-19T10:18:03 gmcharlt: I would say drop the constraint from all the rules, personally. 2011-01-19T10:18:05 tsbere: how about when we remove script support? FWIW, autogrid (well, autofieldwidget) can apply regex's if they are defined in the IDL 2011-01-19T10:18:10 so then /evergreen/opensrf 2011-01-19T10:18:19 yes 2011-01-19T10:18:24 eeevil: When will that be removed? 3.0? 2011-01-19T10:18:27 but usually, as dbs says, keeping the tarballs with their original names is good practice 2011-01-19T10:18:51 tsbere: it hasn't been decided, but certainly after 2.1 2011-01-19T10:18:55 I'd highly reccomend using the /home/opensrf/OpenSRF-1-x-x-x structure 2011-01-19T10:19:08 lowery: okay. maybe mrpeters-isl is right then: sudo aptitude install build-essentials 2011-01-19T10:19:32 * gmcharlt is going to punt for now and just write up a lp bug 2011-01-19T10:19:33 just do a wget of the tarball, and tar xzf OpenSRF-1.x.x.tar.gz 2011-01-19T10:20:04 dbs: build-essentials is in the Makefile IIRC, right? 2011-01-19T10:20:45 kinda makes me worried for what other deps may be missing 2011-01-19T10:20:48 lowery: the location and names shouldn't matter, really, as long as we're sure he's in the right directory 2011-01-19T10:20:55 gmcharlt: can autoblah things be shown how to convert "_" to " " and back? 2011-01-19T10:21:37 mrpeters-isl: build-essential is in the Makefile.install "DEBS" list 2011-01-19T10:21:51 yep, thought so...so we may see more problems 2011-01-19T10:21:58 **IF** that's what's missing 2011-01-19T10:22:09 lowery: can you link to the documentation you're following? 2011-01-19T10:22:25 it will help us help you along 2011-01-19T10:22:34 http://evergreen-ils.org/dokuwiki/doku.php?id=opensrf:1.6:install 2011-01-19T10:23:02 so you're crashing out at step 4? 2011-01-19T10:23:05 lowery: okay - and what operating system are you running on? 2011-01-19T10:23:11 dbs: lucid 2011-01-19T10:23:42 yes, however just redid it in the home/user/opensrf directory and it worked. so my idea of going to a top level directory must have messed it up 2011-01-19T10:23:56 lowery: so you ran - cd /evergreen/opensrf && make -f src/extras/Makefile.install ubuntu-lucid ? 2011-01-19T10:24:50 lowery: hmm - that's pretty interesting & frighteningly fragile if that problem can be reproduced 2011-01-19T10:24:57 * mrpeters-isl tries 2011-01-19T10:25:47 10.4 = lucid right? 2011-01-19T10:25:54 dbs lowery mrpeters-isl etc: Could it be a permissions issue on his /evergreen folder? 2011-01-19T10:26:24 tsbere: that's an idea - i guess it would need to be chown'd as opensrf if he downloaded as a different user? 2011-01-19T10:26:25 tsbere: anything's possible 2011-01-19T10:27:07 Was thinking "root extracts tarball, what user(s) did it just give all the files to?" 2011-01-19T10:27:17 true 2011-01-19T10:28:08 lowery: if you still have the /evergreen/opensrf structure, could you do a ls -ltrh /evergreen/opensrf 2011-01-19T10:28:16 to check the ownership? 2011-01-19T10:28:30 did anyone else see my response to Anthony Ortega? 2011-01-19T10:28:45 (on the mailing list) 2011-01-19T10:28:46 gmcharlt: that constraint is the only thing holding back the use of spaces, and since we're looking to remove it at some point it would seem ok to me to drop it for that instance 2011-01-19T10:29:11 dbs: i did, i responded to him last week too 2011-01-19T10:29:13 eeevil: true - installed base for chdd users is tiny 2011-01-19T10:29:16 he never replied 2011-01-19T10:29:17 evergreen is root:root 2011-01-19T10:29:28 lowery: i kind of bet thats the problem...it should be owned by opensrf 2011-01-19T10:29:44 if you grab the tarball as opensrf, and extract as opensrf...you don't run into the permissions issue 2011-01-19T10:30:24 *** r123 has joined #evergreen 2011-01-19T10:30:40 aww, well ill tell you what the guy that made that folder and didnt changes the permission must feel like an ass for such a simple overlook. 2011-01-19T10:30:45 mrpeters-isl: I thought he did reply to you, in that he included the memcached / ejabberd info you requested, in his subsequent post 2011-01-19T10:30:56 dbs: crap, maybe i missed it then 2011-01-19T10:31:03 mrpeters-isl: thanks for letting me know I'm not crazy though :) 2011-01-19T10:31:08 i tuned out over the weekend 2011-01-19T10:31:31 * gmcharlt grabs 0477 2011-01-19T10:31:34 lowery: happens to everybody sometime 2011-01-19T10:33:13 dbs: FWIW it looks like he replied to your email from last night 2011-01-19T10:33:44 oh, actually maybe not...i thought i saw that earlier but i was dreaming 2011-01-19T10:33:54 he just reposted his initial message 2011-01-19T10:34:24 mrpeters-isl: yeah, that's why I was questioning my sanity 2011-01-19T10:34:39 ill shoot him an email directly and tell him to come join us here 2011-01-19T10:35:37 i know i've had this issue before...i'm just forgetting what caused it 2011-01-19T10:36:24 Suggestion: one day, we have a "break OpenSRF / Evergreen" party, where we find creative ways to break configs, document the broken config and the resulting log messages / evidence 2011-01-19T10:36:48 only if we keep score! i bet I'll have the most broken configs 2011-01-19T10:37:29 and then that documentation can help people find their way back to unbrokenness. For bonus points, revise the code to trap the known broken configs and throw more prescriptive error messages. 2011-01-19T10:37:30 i kind of bet OpenSRF C isn't started on Anthony's server 2011-01-19T10:37:41 mrpeters-isl: Don't forget, you have to document how they are broken, and how to fix them. 2011-01-19T10:37:48 only if the end result of that day is a series of patches and commits that bulk up the error handling! 2011-01-19T10:37:54 tsbere: no fair! i just want to break things! 2011-01-19T10:38:07 dbs: ah, you already said that. 2011-01-19T10:39:30 would a settings-tester.pl for OpenSRF be a possibility? 2011-01-19T10:39:46 to check things like jabber, memcache, etc? 2011-01-19T10:40:11 it's helped me catch a jabber config issue or a postgres permission issue many of times 2011-01-19T10:40:17 mrpeters-isl: yes, but better errors would be even better 2011-01-19T10:40:56 true 2011-01-19T10:41:07 Anthony just emailed back, he will be in here in a bit 2011-01-19T10:41:17 * mrpeters-isl thinks he gets a daily digest 2011-01-19T10:41:32 in one of his replies, it's a digest 2011-01-19T10:43:07 *** jamesrf has joined #evergreen 2011-01-19T10:48:55 mrpeters-isl: that makes sense 2011-01-19T10:50:23 * gmcharlt grabs 0478 2011-01-19T10:53:58 *** jamesrf has quit IRC 2011-01-19T10:54:24 looks like we've found a potential blocker :( ... we need to get a work-ou filtered, ordered tree into some dojo data stores, and there's code that trys to do it, but falls down when work-ous are ordered in particular ways 2011-01-19T10:55:21 so, I'm going to attack that. and berick is poking at https://bugs.launchpad.net/evergreen/+bug/699914 2011-01-19T10:58:30 *** lowery has quit IRC 2011-01-19T11:09:30 * gmcharlt has finished backporting the naco_normalize() changes to rel_2_0 2011-01-19T11:09:42 I'd appreciate extra eyes and testing 2011-01-19T11:13:28 * berick submitted a patch that could use some independent testing for https://bugs.launchpad.net/evergreen/+bug/699914 2011-01-19T11:22:49 messing with hard due dates requires permission at the consortium level? 2011-01-19T11:31:55 Dyrcona: according to fm_IDL.xml, requires CREATE/UPDATE/DELETE_CIRC_DURATION + work_ou; maybe the UI or middle layer code introduces some other complications? 2011-01-19T11:34:57 *** finnapz has quit IRC 2011-01-19T11:36:27 dbs: That looks like "we don't have a table field specified for where to get a permission value from, need it everywhere" to me. Even though there is an "owner" id (that would need linking to aou, I think) 2011-01-19T11:37:31 * tsbere wonders how the permissions would have to work for chddv though 2011-01-19T11:38:07 *** youdonotexist has joined #evergreen 2011-01-19T11:38:42 tsbere: you can link through to another related table to use the aou field "over there" 2011-01-19T11:39:03 eeevil: Good to know. No clue how yet, may look at that later. 2011-01-19T11:40:45 * tsbere found an example in the IDL 2011-01-19T11:41:04 :) 2011-01-19T11:41:35 ccmcmt uses ccmm for permissions. 2011-01-19T11:42:35 *** finnapz has joined #evergreen 2011-01-19T11:44:45 *** gdunbar has joined #evergreen 2011-01-19T12:11:20 *** jenny has joined #evergreen 2011-01-19T12:30:54 *** tspindler has quit IRC 2011-01-19T12:46:24 *** sfortin has quit IRC 2011-01-19T12:46:39 hey all, looking to submit a new triggered event to Evergreen Contrib, but seeking some help with the proper way to seed the action_trigger.event_params and action_trigger.environment rows - currently, I just have a file at.event_params that contains - http://pastie.org/1477933 - with the comment to update with the proper id matching the new event_def 2011-01-19T12:46:44 is there a better way to do this? 2011-01-19T12:48:35 if i remember correctly, i can't do both a SELECT and VALUES within the INSERT INTO, right? 2011-01-19T12:51:03 there's a way to use the normal sequence (which would put it out of the reserved range of id's), and then reference that same id value elsewhere. I can dig up an example once there's no food on my fingers 2011-01-19T12:51:14 haha 2011-01-19T12:51:16 mrpeters-isl: you can use functions, like currval to get the last value pulled from a specific sequence (the ev_def seq, in this case) 2011-01-19T12:51:18 thanks phasefx 2011-01-19T12:51:21 mrpeters-isl: "You can do SELECT 'SOMEID', 'sender_email', email FROM table" as your source 2011-01-19T12:51:40 er, extra_quotes-- :( 2011-01-19T12:51:48 But you should get the idea 2011-01-19T12:52:06 mrpeters-isl: http://www.postgresql.org/docs/current/static/functions-sequence.html 2011-01-19T12:52:23 (what phasefx and I are talking about) 2011-01-19T12:52:33 eeevil: so that'd be accepted? i guess it makese sense since one would hope this is the only event being added at the particular time 2011-01-19T12:53:08 mrpeters-isl: it's per-session ... it's safe 2011-01-19T12:53:14 ok 2011-01-19T12:53:40 only thing i guess i don't follow is how i can do both an INSERT VALUES and SELECT together...it's not possible right? 2011-01-19T12:54:00 maybe someone has a simple foo/bar type example? 2011-01-19T12:54:31 why would you need to? 2011-01-19T12:54:48 mrpeters-isl: Use literals in your select (see above) 2011-01-19T12:55:06 INSERT INTO blah (a,b,c) VALUES ('literal',CURRVAL('sequence_name'),42); 2011-01-19T12:55:15 Or that 2011-01-19T12:55:43 ah ok i was thinking i need to do a direct SELECT on the id_seq 2011-01-19T13:01:57 awesome. works like a charm 2011-01-19T13:01:58 thank you guys 2011-01-19T13:12:43 mrpeters-isl: your separate .sql files, that may cause problems with currval if you're using it one file but depending on a sequence that got incremented in another file. \i file1.sql \i file2.sql \i etc. in a single psql shell may fix that, but I'm not sure 2011-01-19T13:14:16 phasefx: you know, you kind of make a good point...it might be safer to have this all in one transaction too 2011-01-19T13:14:37 yeah 2011-01-19T13:14:56 like an upgrade/ script in trunk 2011-01-19T13:16:14 see 0355.data.missing_pieces_format.sql, though it's not using currval 2011-01-19T13:17:19 cool. i'll submit this as a single SQL within a transaction 2011-01-19T13:17:46 tsbere: going way back - duh, on my part, for failing to read alllll the way over to the "global_required" attribute 2011-01-19T13:20:37 eeevil: as it's been a full 24 hours (hah), any progress on the security mailing list? 2011-01-19T13:22:33 tsbere had a suggested follow-on to my early morning commits and it would be good to be able to start talking about those a little more broadly 2011-01-19T13:25:29 *** lowery has joined #evergreen 2011-01-19T13:26:46 hello again, so having another problem with the install. for shits and giggles jus to make sure i had not done anything wrong i have done a reformat and gotten to this step and get the same proble. when doing the install of opensrf i get STEP 5 2011-01-19T13:26:46 epmd -kill 2011-01-19T13:26:46 killall beam; killall beam.smp 2011-01-19T13:26:46 rm /var/lib/ejabberd/* 2011-01-19T13:26:46 echo 'ERLANG_NODE=ejabberd@localhost' >> /etc/default/ejabberd 2011-01-19T13:26:46 user@Evergreen:/$ sudo etc/init.d/ejabberd start 2011-01-19T13:26:46 Starting Jabber Server: ejabberd 2011-01-19T13:26:47 Crash dump was written to: /var/log/ejabberd/erl_crash.dump 2011-01-19T13:26:47 kernel pid terminated (application_controller) ({application_start_failure,kerne;,{kernel,start,[normal.[]]}}}) 2011-01-19T13:26:48 crash dump was written to: /var/log/ejabberd/erl_crash.dump 2011-01-19T13:26:48 kernel pid terminated (application_controller) ({applcation_start_failure,kernel,{shutdown,kernel,start,[normal,[]]}}}) 2011-01-19T13:26:49 . 2011-01-19T13:27:06 sorry for spam size of post 2011-01-19T13:27:59 * csharp notes that the pastebin link is not visible in the topic on his irssi screen... 2011-01-19T13:28:46 lowery: http://paste.lisp.org/new/evergreen 2011-01-19T13:29:19 lowery: what OS are you building on? 2011-01-19T13:29:29 lucid 2011-01-19T13:30:02 lowery: had you modified anything ejabberd-related before receiving this? 2011-01-19T13:30:04 echo 'ERLANG_NODE=ejabberd@localhost' >> /etc/default/ejabberd is that right? 2011-01-19T13:30:14 ubuntu 2011-01-19T13:30:23 i think there is a good bit more to modify in ejabberd.cfg 2011-01-19T13:30:27 nm - I see now 2011-01-19T13:30:28 did standard install 2011-01-19T13:30:31 postgresql install 2011-01-19T13:30:39 and then followed the walkthrough 2011-01-19T13:30:53 lowery: just to be clear, you didn't install ejabberd manually...you got it through the makefile? 2011-01-19T13:30:58 yes 2011-01-19T13:31:11 so, ps aux | grep ejabberd 2011-01-19T13:31:14 is it stopped still? 2011-01-19T13:31:27 lowery: can you pastebin your /etc/ejabberd/ejabberd.cfg file (srubbed of passwords, please)? 2011-01-19T13:32:01 s/srubbed/scrubbed/ 2011-01-19T13:32:06 ignore my comment about the "echo..." i see what you did now 2011-01-19T13:32:11 no passwords in ejabberd.cfg :) 2011-01-19T13:32:27 ejabberd 31582 0.0 0.0 10760 348 ? S 11:38 0.00 /usr/lib/erlang/erts-5.7.4/bin/epmd -daemon 2011-01-19T13:32:37 hmm so it seems still to be running 2011-01-19T13:32:45 just plain /etc/init.d/ejabberd stop doesn't work? 2011-01-19T13:33:03 correct 2011-01-19T13:33:18 hmm 2011-01-19T13:33:37 and you tried epmd -kill, etc. 2011-01-19T13:34:35 yes 2011-01-19T13:34:51 and you get the crash when you try to kill jabber? 2011-01-19T13:35:26 just crash when i try to stop or start. no messages when i do a kill 2011-01-19T13:35:34 and you're starting/stoping/killing as root, not opensrF? 2011-01-19T13:35:58 lowery: were you able to start ejabberd without that ERLANG_NODE=ejabberd@localhost line? 2011-01-19T13:36:06 i do a sudo to do the stop / start 2011-01-19T13:36:12 ok 2011-01-19T13:36:26 no i was not able to start ejabberd either way 2011-01-19T13:36:45 you can ping private.localhost / public.localhost ok? 2011-01-19T13:37:01 I haven't installed on ubuntu in a long time, and haven't seen that particular ejabberd issue for a long time either. Coincidence? 2011-01-19T13:37:28 i have a lucid server here...let me give a crack at this 2011-01-19T13:37:41 which OpenSRF version again? 2011-01-19T13:37:47 can ping both just fine 2011-01-19T13:37:56 1.6.2 2011-01-19T13:38:05 k let me catch up to you here 2011-01-19T13:38:06 mrpeters-isl: the problem is ejabberd at the moment, let's solve that first 2011-01-19T13:38:28 dbs: right, i have Lucid here so i was going to see if i ran into the same crash with jabber 2011-01-19T13:39:40 lowery: to be clear, did you make any modifications to ejabberd's configuration, other than the ERLANG_NODE bit, before trying to start it or stop it? 2011-01-19T13:41:00 just the addition fo the public and private localhost bit from the walkthrough 2011-01-19T13:41:23 to /etc/ejabberd/ejabberd.cfg ? 2011-01-19T13:41:26 {hosts, ["localhost"]}. to {hosts, ["localhost", "private.localhost", "public.localhost"]}. 2011-01-19T13:41:52 need to stop jabber before changing the config perhaps? 2011-01-19T13:41:56 ejabberd will crash like crazy if that file has bad syntax 2011-01-19T13:42:23 lowery: you might need to do a "dpkg -purge ejabberd" to clean up the ejabberd database and remove the package, then "aptitude install ejabberd" for a clean start. I think in the past ubuntu had problems with some versions of ejabberd that were cleared up by subsequent updates; if you haven't run "sudo aptitude update; sudo aptitude safe-upgrade" now might be a good time 2011-01-19T13:43:13 hmmm starting ejabberd................failed 2011-01-19T13:43:30 during the run of the makefile 2011-01-19T13:43:43 can i get a cookie for making your's fail too? 2011-01-19T13:43:52 did yours do the same? 2011-01-19T13:43:54 during the install process? 2011-01-19T13:44:10 just sits on "Starting jabber server" for a good minute, and then fails 2011-01-19T13:44:17 not problems with ejabberd till this point 2011-01-19T13:44:27 alternately, perhaps the current version of the ejabberd package for Ubuntu is busted 2011-01-19T13:44:28 * csharp suspects two separate issues 2011-01-19T13:44:35 perhaps 2011-01-19T13:44:44 quite possible 2011-01-19T13:45:19 for lowery, please purge the ejabberd package, update aptitude, and reinstall it to get to a known point 2011-01-19T13:45:37 ok trying that now 2011-01-19T13:46:10 * csharp 's ejabberd on lucid is running fine 2011-01-19T13:46:18 and no updates to it 2011-01-19T13:46:42 looks like i have 2.1.2-2 2011-01-19T13:47:10 same failure after aptitude update/upgrade 2011-01-19T13:47:24 mrpeters-isl: that's the version I'm on 2011-01-19T13:47:26 must be the other jabber servers that mrpeters-isl is running that's blocking his port 2011-01-19T13:47:36 after reinstall i have 2.1.5-2 2011-01-19T13:47:47 mrpeters-isl: time to check the /var/log/ejabberd.log (or whatever) 2011-01-19T13:47:57 yeah don't worry about me 2011-01-19T13:48:10 lowery: can you start ejabberd right now? 2011-01-19T13:49:02 works now. 2011-01-19T13:49:08 yay 2011-01-19T13:49:39 so stop it, and then make the config changes to /etc/ejabberd/ejabberd.cfg - very carefully :) 2011-01-19T13:49:59 btw - debian squeeze release is scheduled for February 5/6 - huzzah 2011-01-19T14:03:26 *** rsoulliere has joined #evergreen 2011-01-19T14:04:08 *** finnapz has quit IRC 2011-01-19T14:06:25 *** finnapz has joined #evergreen 2011-01-19T14:09:15 able to start and stop without problems after the uninstall reinstall 2011-01-19T14:12:00 *** jenny has quit IRC 2011-01-19T14:13:41 *** alxp has quit IRC 2011-01-19T14:13:41 *** mtate has quit IRC 2011-01-19T14:13:41 *** brendan2 has quit IRC 2011-01-19T14:13:42 *** phasefx_ has quit IRC 2011-01-19T14:13:42 *** Ghidorah has quit IRC 2011-01-19T14:13:43 *** AaronZ-PLS has quit IRC 2011-01-19T14:13:44 *** youdonotexist has quit IRC 2011-01-19T14:13:45 *** kmlussier has quit IRC 2011-01-19T14:13:46 *** mrpeters-isl has quit IRC 2011-01-19T14:13:47 *** berick has quit IRC 2011-01-19T14:13:48 *** dbwells has quit IRC 2011-01-19T14:13:49 *** leed has quit IRC 2011-01-19T14:13:49 *** Callender has quit IRC 2011-01-19T14:13:50 *** kbeswick has quit IRC 2011-01-19T14:13:51 *** finnapz has quit IRC 2011-01-19T14:13:51 *** Dmagick has quit IRC 2011-01-19T14:13:52 *** eeevil has quit IRC 2011-01-19T14:13:52 *** _bott_ has quit IRC 2011-01-19T14:13:52 *** denials has quit IRC 2011-01-19T14:13:53 *** csharp has quit IRC 2011-01-19T14:16:45 *** pmplett has joined #evergreen 2011-01-19T14:16:45 Hi, I see it was said at yesterday's DEV meeting that I "requested a SIP backport" 2011-01-19T14:16:48 Actually, that was more of a question/enquiry. 2011-01-19T14:17:33 *** collum has left #evergreen 2011-01-19T14:22:28 I used launchpad to report just in case someone using 1.6 and SIP checkin might be able to see that it can be done. 2011-01-19T14:22:37 *** finnapz has joined #evergreen 2011-01-19T14:22:37 *** youdonotexist has joined #evergreen 2011-01-19T14:22:37 *** kmlussier has joined #evergreen 2011-01-19T14:22:37 *** alxp has joined #evergreen 2011-01-19T14:22:37 *** berick has joined #evergreen 2011-01-19T14:22:37 *** dbwells has joined #evergreen 2011-01-19T14:22:37 *** mrpeters-isl has joined #evergreen 2011-01-19T14:22:37 *** AaronZ-PLS has joined #evergreen 2011-01-19T14:22:37 *** Ghidorah has joined #evergreen 2011-01-19T14:22:37 *** tater has joined #evergreen 2011-01-19T14:22:37 *** brendan2 has joined #evergreen 2011-01-19T14:22:37 *** phasefx_ has joined #evergreen 2011-01-19T14:22:37 *** Dmagick has joined #evergreen 2011-01-19T14:22:37 *** leed has joined #evergreen 2011-01-19T14:22:37 *** Callender has joined #evergreen 2011-01-19T14:22:37 *** eeevil has joined #evergreen 2011-01-19T14:22:37 *** kbeswick has joined #evergreen 2011-01-19T14:22:37 *** _bott_ has joined #evergreen 2011-01-19T14:22:37 *** csharp has joined #evergreen 2011-01-19T14:22:37 *** denials has joined #evergreen 2011-01-19T14:22:49 Also, if I can be of any assistance in testing or sharing my logs let me know. 2011-01-19T14:33:23 *** rsoulliere has quit IRC 2011-01-19T14:34:27 *** rsoulliere has joined #evergreen 2011-01-19T14:34:39 *** pmplett has quit IRC 2011-01-19T14:39:24 *** lowery has quit IRC 2011-01-19T14:41:40 *** pmplett has joined #evergreen 2011-01-19T14:42:31 *** bshum has joined #evergreen 2011-01-19T14:50:56 *** kmlussier has quit IRC 2011-01-19T14:57:39 *** sfortin has joined #evergreen 2011-01-19T15:00:10 *** ChanServ changes topic to "Welcome to the #Evergreen library system channel! | We are publicly logged. | Large pastes at http://paste.lisp.org/new/evergreen" 2011-01-19T15:12:05 berick: Hmm, about the patch to add back permission checking for patron registration in 2.0... I've been trying to apply it and have found that I can't seem to find the positioning for the new code pieces. Was it built against something other than RC1/2? 2011-01-19T15:13:24 Some of the surrounding code doesn't appear in those versions, so I guess I was wondering if maybe I'm doing it wrong :) 2011-01-19T15:15:38 *** rsoulliere has quit IRC 2011-01-19T15:16:51 bshum: built against trunk, lemme make sure it will apply cleanly. in the meantime, make sure you apply with patch -p1, since git adds an extra bit to the path 2011-01-19T15:17:25 berick: Ah, alright. Thanks :) 2011-01-19T15:17:33 I didn't know about the -p1 part 2011-01-19T15:17:42 I started out by trying to copy/paste pieces by hand 2011-01-19T15:17:48 And realized that things didn't quite look the same 2011-01-19T15:17:52 my bad for not posting a patch that will cleanly apply.. 2011-01-19T15:19:43 hmm, trunk register.js is a little different.. 2011-01-19T15:19:55 ah, the save-on-exit stuff 2011-01-19T15:23:04 I'm still trying to make the biblios and loc z39.50 servers not appear in the MARC import interface... 2011-01-19T15:23:29 I have removed biblios and loc as sources from the GUI (verified this in the database) 2011-01-19T15:23:52 and I have commented out their config in /openils/conf/opensrf.xml 2011-01-19T15:24:40 I've also cleared my local staff client cache, but they are still showing up (along with an entry "#text" - perhaps that's a typo?) 2011-01-19T15:25:01 what other steps are there to this? 2011-01-19T15:26:15 I can see that the z3950.js file loops over a list of active_servers... where does that list get its values? 2011-01-19T15:26:21 csharp: You restarted opensrf right? I think dbs mentioned a specific service 2011-01-19T15:26:49 bshum: pretty sure I did, but I'll do it again 2011-01-19T15:27:13 csharp: autogen (?) 2011-01-19T15:27:31 moodaepo: hmm - does that require autogen? 2011-01-19T15:27:37 I can try that too 2011-01-19T15:27:54 Just throwing it out there 2011-01-19T15:28:04 *** jenny has joined #evergreen 2011-01-19T15:28:11 * csharp starts by restarting opensrf 2011-01-19T15:31:01 bshum: this should apply cleanly to rel_2_0 http://dev198.esilibrary.com/~berick/patches/reg_perm_2_0.patch (no -p1 needed) 2011-01-19T15:31:38 opensrf restarted, cache cleared, still visible 2011-01-19T15:31:43 * csharp tries autogen 2011-01-19T15:32:08 csharp: you removed them from the DB and opensrf.xml? 2011-01-19T15:32:29 berick: yes - removed from DB and commented them out in opensrf.xml 2011-01-19T15:32:46 and restart opensrf.setting and open-ils.search (at minimum) 2011-01-19T15:32:56 or just all perl services 2011-01-19T15:34:22 berick: yes and still there 2011-01-19T15:34:34 * csharp sticks fist through monitor 2011-01-19T15:36:27 csharp: For fun thoughts, what if you deleted them entirely from opensrf.xml. Not just commenting it out. 2011-01-19T15:36:39 bshum: good thought 2011-01-19T15:38:47 * tsbere celebrates security list welcome message 2011-01-19T15:39:18 yay! 2011-01-19T15:39:29 apparently commenting out is not enough ;-) 2011-01-19T15:39:36 Evil. 2011-01-19T15:39:43 It should have worked, oh well 2011-01-19T15:39:44 * csharp seeks tools to repair his monitor 2011-01-19T15:39:44 dbs: Which of us should start that discussion? 2011-01-19T15:39:53 csharp: how did you comment out? 2011-01-19T15:40:15 tsbere: single #'s at the beginning of each line 2011-01-19T15:40:40 csharp: I think you need to use those silly <-- --> arrows? 2011-01-19T15:40:41 csharp: is XML commmenting 2011-01-19T15:40:42 csharp: Aren't the config files XML? And thus needing XML style comments? Like 2011-01-19T15:40:46 Erm, ! 2011-01-19T15:40:47 Yeah 2011-01-19T15:41:05 * csharp forehead slap 2011-01-19T15:41:13 thanks all ;-) 2011-01-19T15:41:43 *** alxp has quit IRC 2011-01-19T15:41:52 * csharp needs more sleep, apparently 2011-01-19T15:41:54 C-a C-[Space] C-e C-w is how I comment things out. :) 2011-01-19T15:42:07 or C-k C-k is faster. 2011-01-19T15:43:11 Dyrcona: which editor? 2011-01-19T15:44:05 berick++ # the patch applied nicely and testing so far appears quite functional. Awesome! 2011-01-19T15:44:27 bshum: great, thanks for testing 2011-01-19T15:46:01 *Which* editor? Why the One, True editor, Emacs.... :) 2011-01-19T15:46:34 Or, if you're old-fashioned, you can hit [Esc] dd 2011-01-19T15:46:37 Dyrcona: I'm a vim guy ;-) 2011-01-19T15:46:52 * csharp prepares for fisticuffs 2011-01-19T15:48:36 csharp: syntax:on 2011-01-19T15:48:44 berick: ha! 2011-01-19T15:49:05 er, :syntax on 2011-01-19T15:49:29 Hmm, how are statistical categories ordered in 2.0? By ID rather than name? 2011-01-19T15:49:37 Erm, patron statistical categories 2011-01-19T15:49:42 From the dropdown selection 2011-01-19T15:50:04 bsum: could be. 2011-01-19T16:00:19 *** Meliss has quit IRC 2011-01-19T16:02:47 *** brendan2 has quit IRC 2011-01-19T16:04:11 *** sfortin has quit IRC 2011-01-19T16:06:24 apologies, folks, I was just out snowshoeing in the -18C weather 2011-01-19T16:07:30 tsbere: if you have time to open it up, please do! 2011-01-19T16:07:41 dbs: Will do. Was already writing an email. 2011-01-19T16:07:46 tsbere++ 2011-01-19T16:18:09 * tsbere has sent his email, thus starting the first security list discussion 2011-01-19T16:31:44 * tsbere grumbles about greylisting 2011-01-19T16:32:31 weird, I got phasefx's reply (on one of my accounts, only), haven't seen tsbere's post yet. greylisting, heh 2011-01-19T16:33:27 My own fault for implementing it, although phasefx's reply was apparently "to all" not "to list" so I am getting a copy via google. Which hasn't made it through the greylister yet. 2011-01-19T16:35:15 Huh. I haven't gotten anything from a google-based account sent to me in over a month. 2011-01-19T16:37:00 dbs: the Finding stories one? weird, I thought I _just_ sent that less than a minute ago, but IRC says 3 minutes. spooky time travel :) 2011-01-19T16:37:10 gmail says 1 minute ago 2011-01-19T16:37:42 phasefx: no, security 2011-01-19T16:37:49 ah, okay *8) 2011-01-19T16:53:34 dbs: I see that time travel phenomena quite a bit on some lists. I get replies sometimes before getting the original message. 2011-01-19T16:56:11 Do money.billing entries with a timestamp of '2011-01-19 23:59:59-05' (i.e. tonight before midnight) show up in the patron records? Or is this just a placeholder event till it actually passes? 2011-01-19T16:56:45 Also, when should the fine_generator.pl script run at night usually? 2011-01-19T16:57:29 bshum: they show, the event is not a "placeholder", just a quirk. i'd like for it not to be a quirk, but that will be... complex. 2011-01-19T16:57:51 bshum: some people run the fines generator more frequently, if they have things like hourly fines. 2011-01-19T16:57:58 jeff: Right, that makes sense. 2011-01-19T16:58:15 jeff: I guess I'm wondering if this is where the communication is breaking down between us and front-line staff. 2011-01-19T16:58:49 if you run it after close and it completes before midnight, i think it might stop generating those "future" billings. 2011-01-19T16:59:18 that might be the simplest fix for the "quirk" that annoys me so. :) 2011-01-19T16:59:34 i don't know the internals well enough at this point. 2011-01-19T16:59:54 i know it changed from 1.4 to 1.6 to accomodate hourly fines, etc. 2011-01-19T17:00:14 that's when the due time of 23:59 came into existance also. 2011-01-19T17:00:18 jeff: but wouldn't that effectively change the amount being fined? 2011-01-19T17:00:24 jeff: I'm trying to figure out exactly what's happening, because now I'm not sure whether we're seeing "extra" fines that shouldn't be there. Vs. fines that haven't happened yet. Vs. fines that aren't being reported right. 2011-01-19T17:01:14 dbwells: it could, and that may or may not be "fixed" at checkin... see my disclaimer about not knowing the internals well enough yet :) 2011-01-19T17:02:32 jeff: fair enough. Then I guess I will also chime in to say that fines for 11:59 of the current day are 'normal' for us, too. 2011-01-19T17:02:47 bshum: been there, it was Really Fun for circs that had fines that spanned 1.4 and 1.6 ;-) 2011-01-19T17:03:08 those I can guarantee got at least one day's worth of "extra" fines. 2011-01-19T17:03:14 (for our system, at least) 2011-01-19T17:03:21 jeff: Well I'm just trying to make sure that it's not our perception of fines or if there really are extra days. 2011-01-19T17:03:30 This all relates to the bug I filed about backdating and voiding of fines. 2011-01-19T17:03:42 Because if that extra day really exists, then maybe it's not a real bug 2011-01-19T17:03:47 But our mistaken perception of it. 2011-01-19T17:04:08 And if it does exist, then we need to try getting rid of it, because it's really wigging people out 2011-01-19T17:04:15 The extra day I mean 2011-01-19T17:04:27 That's an interesting though about running fines before midnight though 2011-01-19T17:04:28 this assumes that none of the circulations in question are "leap circs". due to gravitational forces related to large library collections, every 1048576th circulation is a "leap circ"... 2011-01-19T17:04:32 * jeff stops 2011-01-19T17:04:43 :D 2011-01-19T17:04:54 Gahh......... 2011-01-19T17:05:55 If the fine generator only runs once a day, and finishes its run before midnight, then these current day entries would probably not generate till it runs later in the evening. 2011-01-19T17:06:21 time and written language: two things that computers (and people) have a hard time with. 2011-01-19T17:06:28 Sigh 2011-01-19T17:06:34 Just can't win. 2011-01-19T17:07:04 bshum: add a few minutes to the dates and see what happens. You can play with this in a test server, right? 2011-01-19T17:07:35 Dyrcona: Sort of, yeah 2011-01-19T17:07:40 *** AaronZ-PLS has quit IRC 2011-01-19T17:13:14 *** mjg_ has joined #evergreen 2011-01-19T17:17:55 * tsbere wonders if this email from denials caught in the greylister was supposed to go to the security list too 2011-01-19T17:20:58 bshum: not sure if this is still true concerning newer releases, but in 1.6, you can think of the the overdue timestamp as "this overdue charge covers until x" 2011-01-19T17:21:51 that is, if the bill is dated 1/19/2011 at midnight, the fine generator won't generate another fine until that time has passed. 2011-01-19T17:22:01 tsbere: I replied all to yours, and failed on Dyrcona's 2011-01-19T17:22:13 bshum: think it's safe for me to push that user registration perm patch? 2011-01-19T17:22:26 bshum: by overdue timestamp, I mean the timestamp on the overdue billing. 2011-01-19T17:22:50 berick: I think so. Seemed to work well on our test server with several different custom group permissions. 2011-01-19T17:23:17 dbwells: Right, I think I get that. That's what the money.billing_ts is for an "Overdue Materials" entry. 2011-01-19T17:23:23 bshum++ great, thanks again 2011-01-19T17:23:43 dbwells: But that seems to me that we have an extra entry when there really shouldn't be one. 2011-01-19T17:24:20 dbwells: Because that entry that's there, means we won't generate a new fine, sure. But that still counts towards the total billed/owed by the patron 2011-01-19T17:24:34 Even if they were to return the item today before closing time 2011-01-19T17:24:40 The entry seems to stick 2011-01-19T17:25:45 dbs: I don't see anything from you on the list yet? Not according to archives, anyway. 2011-01-19T17:25:58 hrmm, I'm on trunk, and I see this log line from open-ils.actor, editor[!|1] checking perms user=1, org=4, perm=VIEW_CIRCULATIONS, but then immediately after get an error from open-ils.storage.permission.user_has_perm, where it's trying to pass an empty string instead of an integer, SELECT permission.user_has_perm('1','VIEW_CIRCULATIONS',''); 2011-01-19T17:26:08 tsbere: "awaiting moderator approval" is what I got back 2011-01-19T17:26:52 * tsbere wonders why 2011-01-19T17:27:40 bshum: In my experience, if an item is due on 1/18 (at 11:59pm), when the generator runs anytime after that it will create a billing dated 1/19 at 11:59pm. That essentially means bill for today (1/19) and don't bill again until after this billing date (1/19 at 11:59pm) 2011-01-19T17:28:06 is this different that what you are seeing? 2011-01-19T17:28:42 So the bill is not for the future, but it 'covers' the future. 2011-01-19T17:28:51 dbwells: No, you're right. I just need to count from the initial date forwards properly. 2011-01-19T17:29:32 bshum: but I think you are then correct that the backdate is one day too generous. 2011-01-19T17:30:21 *** finnapz has quit IRC 2011-01-19T17:30:21 *** Dmagick has quit IRC 2011-01-19T17:30:22 *** eeevil has quit IRC 2011-01-19T17:30:22 *** _bott_ has quit IRC 2011-01-19T17:30:22 *** denials has quit IRC 2011-01-19T17:30:23 *** csharp has quit IRC 2011-01-19T17:30:53 dbwells: Tell me if this makes any sense 2011-01-19T17:30:54 dbwells: An item that was due on 1/12 isn't returned till today 2011-01-19T17:30:55 There's entries showing for 1/13, 1/14, 1/15, 1/18, and 1/19 2011-01-19T17:31:05 So that's 5 days of fines 2011-01-19T17:31:11 Even though the patron returned the item today 2011-01-19T17:31:15 *** Dyrcona has quit IRC 2011-01-19T17:31:17 They still get charged for the 1/19 entry 2011-01-19T17:32:12 Yes, that is what I would expect, unless you have a grace period. The simplest example is one day overdue. 2011-01-19T17:32:25 Hmm 2011-01-19T17:32:33 if its due on the 12th, they bring it back the 13th, they are charged for the 13th. 2011-01-19T17:32:45 Yeah, you're right. 2011-01-19T17:34:00 *** finnapz has joined #evergreen 2011-01-19T17:34:00 *** Dmagick has joined #evergreen 2011-01-19T17:34:00 *** eeevil has joined #evergreen 2011-01-19T17:34:00 *** _bott_ has joined #evergreen 2011-01-19T17:34:00 *** denials has joined #evergreen 2011-01-19T17:34:00 *** csharp has joined #evergreen 2011-01-19T17:34:04 That makes sense. 2011-01-19T17:34:04 Okay, I think I'm back on the ball now. Kind of. 2011-01-19T17:34:17 Yep, I think I finally get it. Thanks dbwells. 2011-01-19T17:34:38 That really helps to put things back into perspective. Been running myself in circles trying to think it through. 2011-01-19T17:35:24 The quirkiest part really is the use of future dates on the billings (for me at least), but it allows really simple code like 'next if ($last_fine > $now);' 2011-01-19T17:35:56 Indeed, I think that's where some folks have gotten so confused. 2011-01-19T17:36:09 Myself included, apparently 2011-01-19T17:36:15 tsbere: maybe it's filtering my denials@gmail.com mail 2011-01-19T17:36:17 not much more work to adjust by fine_interval, but I'm not sure how many places would need adjusting. 2011-01-19T17:40:53 sec mailing list needs some adjustment (private archive, etc) ... then members won't be moderated. sorry for the delay 2011-01-19T17:42:19 so, tonight or tomorrow morning for completion 2011-01-19T17:45:57 *** mjg_ has left #evergreen 2011-01-19T17:51:51 *** jenny has left #evergreen 2011-01-19T17:54:47 *** pmplett has quit IRC 2011-01-19T17:57:42 eeevil: The archive defaults to private. I checked that they were before starting stuff on it. And I wasn't moderated, so I dunno. 2011-01-19T18:39:24 *** dbs has quit IRC 2011-01-19T19:20:40 *** youdonotexist has quit IRC 2011-01-19T19:48:31 tsbere: I'm not personally addressing the mailing list config, thus "tonight or tomorrow", but it looks like "tonight" was it in any case :) 2011-01-19T19:49:23 *** dave-esi is now known as dave-esi-away 2011-01-19T21:08:26 * jeff grins 2011-01-19T22:21:37 *** bshum has quit IRC 2011-01-19T23:57:46 *** moodaepo has quit IRC