2012-01-04T00:38:38 *** dbs has quit IRC 2012-01-04T01:41:49 *** enhancin has joined #evergreen 2012-01-04T01:42:06 Is it possible to downgrade from 2.1.1 to 2.1.0 or would that cause more issues than it's worth? 2012-01-04T03:30:05 *** edoceo has quit IRC 2012-01-04T07:18:17 *** tfaile has quit IRC 2012-01-04T07:44:51 *** collum has joined #evergreen 2012-01-04T07:58:23 enhancin: there's not really a way to downgrade - probably best to reinstall the older version IMHO 2012-01-04T08:00:08 enhancin: having said that, you wouldn't have to start from scratch - I would tar up /openils and just do the Makefile.install/configure/make/make install dance 2012-01-04T08:08:31 *** wolf29 has joined #evergreen 2012-01-04T08:26:16 *** Dyrcona has joined #evergreen 2012-01-04T08:27:25 *** _bott_ has quit IRC 2012-01-04T08:27:58 *** _bott_ has joined #evergreen 2012-01-04T08:29:38 *** akilsdonk has joined #evergreen 2012-01-04T09:10:06 *** Meliss has joined #evergreen 2012-01-04T09:25:17 *** jenny has joined #evergreen 2012-01-04T09:26:05 *** tfaile has joined #evergreen 2012-01-04T09:30:01 *** dbs has joined #evergreen 2012-01-04T09:36:12 Anyone know if the conference hackfest is the 25th or 24th of April? 2012-01-04T09:41:10 Dyrcona: 25th I believe; hackfest for one day, conf sessions on 26, 27, 28 2012-01-04T09:42:04 * dbs isn't sure if anyone wants to hang out for an in-person dev day after the conference like we did last year (which primarily turned into a git indoctrination session) :) 2012-01-04T09:43:07 dbs: thanks. I plan to get registration, hotel and travel arrangements out of the way in one swoop. 2012-01-04T09:44:03 dbs: sounds lovely 2012-01-04T09:44:07 * dbs has registered and has hotel but has not yet arranged flight due to possibility of post-conference hijinx... but probably should do so soon 2012-01-04T09:44:59 * berick would be up for a post-conf shindig 2012-01-04T09:45:33 augh, http://evergreen2012.org/schedule/ has broken links for sponsor images (extra "/schedule/" in URL)... will alert shauna 2012-01-04T09:49:15 *** yboston has joined #evergreen 2012-01-04T09:50:26 *** tspindler has joined #evergreen 2012-01-04T10:00:58 berick: Post-conf shindig as in Sunday or Saturday afternoon? 2012-01-04T10:01:21 * moodaepo should have booked the hotel earlier 24th is gone 2012-01-04T10:01:39 * Dyrcona is worried. 2012-01-04T10:02:01 moodaepo: like last time, saturday afternoon, maybe part of sunday morning 2012-01-04T10:02:06 * Dyrcona wonders what moodaepo means by "24th is gone." 2012-01-04T10:02:48 Dyrcona: Just tried to book 24th - 28th and computer said no. (24th is fully booked at conf prices) 2012-01-04T10:03:14 * Dyrcona grumbles....I knew I should have taken care of this sooner.... 2012-01-04T10:07:36 Dyrcona: same here - we're in full scramble mode for flights and hotel to avoid a 10 hour van ride ;-) 2012-01-04T10:07:56 hmm 2012-01-04T10:08:06 this while coordinating next week's upgrade with the post-holiday blues 2012-01-04T10:08:09 * csharp weeps 2012-01-04T10:10:48 *** agJohn has joined #evergreen 2012-01-04T10:10:53 csharp: 10 hour van ride? Ouch 2012-01-04T10:11:13 mmm, pay a 30% premium to deal with only one connection & one airline, or add one connection and multiple airlines to the mix... 2012-01-04T10:11:28 * berick would drive if it was just a wee bit shorter 2012-01-04T10:12:48 13 hour drive for me; not going to do that again 2012-01-04T10:13:13 * tsbere is considering driving anyway as an excuse to visit family about halfway between here and there.....but that would be 2 days of driving for him 2012-01-04T10:13:19 2 days each way 2012-01-04T10:13:21 <_< 2012-01-04T10:13:35 tsbere++ # Road tripping 2012-01-04T10:15:51 train....choo choo! 2012-01-04T10:15:52 *** dzeiger has joined #evergreen 2012-01-04T10:20:14 train would be awesome 2012-01-04T10:22:45 if amtrak had wifi, we could have the conf on the train ;) 2012-01-04T10:24:00 berick++ 2012-01-04T10:24:41 Or at the least hack fests! 2012-01-04T10:25:20 looks like I'm flying out at 4:00 PM on the 29th 2012-01-04T10:25:33 on your left, you'll see some new code, on your right, the Rockies. 2012-01-04T10:33:12 grabbing 0662 2012-01-04T10:34:21 * dbs wonders what fer 2012-01-04T10:39:48 Oooh, post conf :P 2012-01-04T10:40:14 * dbs heads out into the cold 2012-01-04T10:42:31 post-conf++ 2012-01-04T10:44:53 *** dbs has quit IRC 2012-01-04T10:53:06 So, how close to "no 2.2 targeted pullrequests" are we going for before we push out the next alpha? 2012-01-04T11:13:41 tsbere: this close? 2012-01-04T11:14:14 ---> <--- 2012-01-04T11:15:09 tsbere: you happy with the authen-handler proxy? if so, is there an LP ticket I can push it from? 2012-01-04T11:15:52 eeevil: Made the LP ticket this morning. I am not happy with it, mind you, but the parts I am not happy with are more related to Apache being stupid and not supporting what I want to do ;) 2012-01-04T11:16:19 * tsbere dislikes his handler finishing running only to have apache segfault because it isn't happy 2012-01-04T11:16:42 Basically, you can't do everything from perl yet. So we are stuck with "valid-user" for now. >_> 2012-01-04T11:20:32 eeevil: Fair note, I only did really basic testing of replacing the original with mine. More complete testing should probably be done. 2012-01-04T11:21:39 *** kivilaht2o has joined #evergreen 2012-01-04T11:21:51 tsbere: does valid-user require an actual password file? unsure what valid-user does in this context, actually 2012-01-04T11:22:28 eeevil: "valid-user" basically says "if the backend says we are good, we are good". Adding in "user" or "group" entries and apache starts looking for more info *after* you finish authenticating them. 2012-01-04T11:22:31 tsbere: both exist now, so if it becomes troublesome in some contexts we can use the response handler 2012-01-04T11:22:47 k 2012-01-04T11:26:56 * Dyrcona updates his development branch with latest commit to master while the database restores. 2012-01-04T11:27:23 Hi I would like to know can I somehow configure the Evergreen search engine? 2012-01-04T11:27:39 simplify it a bit to make the searches go faster 2012-01-04T11:27:49 No. 2012-01-04T11:27:58 at the moment it takes ~25 seconds to get one keyword search returned 2012-01-04T11:28:26 I have used some really fast eg servers who have search times in seconds 2012-01-04T11:28:35 me, too. 2012-01-04T11:28:44 *** dbs has joined #evergreen 2012-01-04T11:28:59 kivilaht2o: there are lots of moving parts. first thing: is postgres tuned? (that's a big, loaded question, btw) 2012-01-04T11:29:16 i did the tunings documented in 1.6 documentation 2012-01-04T11:30:47 kivilaht2o: did you use pg_tune? 2012-01-04T11:31:15 nope 2012-01-04T11:31:26 i just edited the core postgresql.conf 2012-01-04T11:31:39 * bshum wonders what tunings were noted in 1.6 docs 2012-01-04T11:31:43 kivilaht2o: Is this for production or for testing/development? 2012-01-04T11:31:47 testing 2012-01-04T11:31:53 i am comparing evergreen and koha 2012-01-04T11:32:01 we are planning to migrate to open source 2012-01-04T11:32:06 http://docs.evergreen-ils.org/1.6/draft/html/configuringPostgreSQL.html ? 2012-01-04T11:32:10 * dbs really should have put forth a "PostgreSQL administration for Evergreen admins" round table proposal. oh well 2012-01-04T11:32:11 kivilaht2o: I suggest just running pg_tune and using its defaults. 2012-01-04T11:32:16 bshum: yes 2012-01-04T11:32:37 dbs: I'll make sure to suggest to moodaepo that we include that amongst the sys admin interest group discussion topics :) 2012-01-04T11:33:07 * tsbere looks at the 14 2.2-targeted pullrequests from one of dbs's links....and has 7 of the 14 branches last he checked 2012-01-04T11:34:00 Dyrcona: thanks, I'll try that 2012-01-04T11:34:06 for production, you'll want to do more tuning than pg_tune does by default. 2012-01-04T11:35:17 there was an old bug about being unable to save Library Adresses 2012-01-04T11:35:33 from 2.1 2012-01-04T11:35:57 do you know has it been resolved already? 2012-01-04T11:35:59 kivilaht1o: you might also want to ensure that the filesystem has cached the postgresql files that hold the indexes and such. first search on a cold database can crawl. 2012-01-04T11:38:09 how can we make a translation out of evergreen= 2012-01-04T11:38:11 Also fun: First *anything* when the host has dropped the entire *VM* to disk. <_< 2012-01-04T11:39:58 via launchpad it seems 2012-01-04T11:40:58 *** youdonotexist has joined #evergreen 2012-01-04T11:41:16 kivilaht1o: yep, translations.launchpad.net/evergreen is the translation interface, and then the translations get folded into releases. Usually. 2012-01-04T11:41:41 eeevil: did you notice my comment on https://bugs.launchpad.net/evergreen/+bug/894052 ? 2012-01-04T11:41:43 Launchpad bug 894052 in Evergreen "Point-to-point version upgrade scripts should be kept in master and backported through applicable releases" (affected: 1, heat: 6) [Undecided,Fix committed] - Assigned to Mike Rylander (mrylander) 2012-01-04T11:43:00 dbs: I didn't. pushing that now, thanks 2012-01-04T11:43:29 *** jbondharris has joined #evergreen 2012-01-04T11:44:42 Goodmorning (again). Actually got logged in as my admin account (Evergreen 2.1.1) Now experiencing an apparent Apache problem getting a client logged in. 2012-01-04T11:45:27 *** wolf29 has quit IRC 2012-01-04T11:45:48 Error message: The requested URL /xul/rel_2_1_1/server/main/ws_info.xul was not found on this server. The file/directory is there on the server, so ther emust be some subtle mistake in my setup. 2012-01-04T11:47:20 I went through the troubleshootin gguide successfully up to step 11, "Try to login from the staff client" 2012-01-04T11:48:26 general question... aside from quieting log errors, is there any advantage to having these (whatever)_custom.properties and custom.js files in place (or disadvantages to not having them)? 2012-01-04T11:49:02 BTW, I can successfully access the machine via its IP address 2012-01-04T11:49:50 csharp: Reduction of requests. A 404 is not usually cached by a lot of things, an empty file is. 2012-01-04T11:50:00 * tsbere still doesn't go to any effort to put them in place 2012-01-04T11:50:33 tsbere: okay - makes sense 2012-01-04T11:53:00 jbondharris: So you login as admin via the client, did you get a prompt to register the workstation? 2012-01-04T11:56:23 My login was from srfsh, so once I am logged in, I only have the srfsh# prompt. I was attempting to login from the client with the same admin login and password 2012-01-04T11:56:52 moodaepo: The error he is getting seems to be "the workstation info/registration box isn't available on the server" 2012-01-04T11:57:06 * tsbere isn't sure why that would be other than botched install 2012-01-04T11:57:25 tsbere: That's what I thought just wanted to confirm he didn't get a prompt 2012-01-04T11:57:46 What might I have missed? Thought I followed the install successfully 2012-01-04T11:58:28 jbondharris: I presume you've run autogen? 2012-01-04T11:58:43 jbondharris: can you pastebin the output of 'ls -l /openils/var/web/xul'? 2012-01-04T11:59:03 * moodaepo probably has since it's in the instructions! 2012-01-04T12:01:52 Done. 2012-01-04T12:03:24 jbondharris: Done as in ran autogen again or copied output to pastebin? 2012-01-04T12:09:40 Hmm.... 2012-01-04T12:10:11 csharp: Downside to empty file compared to 404 is "404 tells some thing to look for the next file, empty file with normal response tells things to stop looking" for the record 2012-01-04T12:10:44 OK, so I'm a noob with pastebin - thought I could just redirect output to pastebinit - seems to work, but don't see it appear. I hate to feel so stupid, but what am I missing?? 2012-01-04T12:13:53 I suppose the other question is, "How do I configure a client for the server?" Where is the documentation, if that's the better question? 2012-01-04T12:15:48 tsbere: good - thanks 2012-01-04T12:16:34 jbondharris: just do 'ls -l /openils/var/web/xul | pastebinit' and see if that works 2012-01-04T12:16:55 you want to pipe it to pastebinit, not redirect it with ">" 2012-01-04T12:18:26 That's what I did, but I'll do it again. 2012-01-04T12:19:31 csharp: some of our people have fun when they use our phone call list module and get a 200 NO CONTENT from the server. They wonder why the page won't load. 2012-01-04T12:20:56 jbondharris: before you go to too much trouble over pastebinit, what I'm looking for is that the symlink "server" points to rel_2_1_1/server - I thought that might be an obvious (to me) thing to rule out ;-) 2012-01-04T12:21:12 Dyrcona: heh 2012-01-04T12:23:12 lrwxrwxrwx 1 opensrf opensrf 16 2012-01-03 09:54 server -> rel_2_1_1/server 2012-01-04T12:23:35 okay - then that settles that ;-) 2012-01-04T12:23:56 drwxr-xr-x 3 opensrf opensrf 4096 2012-01-03 09:52 rel_2_1_1 2012-01-04T12:25:12 and 'ls -l /openils/var/web/xul/server/main/ws_info.xul' returns okay? 2012-01-04T12:26:49 I'm trying to remove the example sub libraries and I can't. the staff client flashes red when trying to delete it and the database says that two entries refer to each other so they can't be deleted 2012-01-04T12:26:51 Yes 2012-01-04T12:27:30 jbondharris: then my next step would be to go to the logs and look for clues there 2012-01-04T12:28:46 jbondharris: You'll need to use SQL in a transaction. 2012-01-04T12:29:00 jbondharris: I would start with the logs in /var/log/apache2/ 2012-01-04T12:29:24 I'll patsebin the code that I used to setup our org tree for migration. It should give you some ideas. 2012-01-04T12:30:07 csharp: I'll take a look. 2012-01-04T12:30:54 * dbs wonders if the SSL cert was accepted? 2012-01-04T12:32:22 If the problem is setting up a new hierarchy, then I'd suggest doing it with SQL. I almost never mess with that from the client. 2012-01-04T12:33:03 Dyrcona: If it involves config, in general, you avoid the client. 2012-01-04T12:33:16 Yeah, I think the client tools for the org unit hierarhcy are misleading, they've almost never worked 2012-01-04T12:33:18 true, I do. 2012-01-04T12:35:01 For setting up the hierarchy during migration or during setup, I'd recommend using something like this: http://pastebin.com/m4C07r2b 2012-01-04T12:38:37 * dbs notes with amusement that the gitstats thing seems to be stuck in November 2011 :) 2012-01-04T12:40:42 maybe something broke when we had to restart it a while back 2012-01-04T12:40:44 jbondharris: can you browse to the URL https://HOSTNAME/xul/rel_2_1_1/server/main/ in a regular web browser? 2012-01-04T12:41:07 eeevil: any ETA on the return of testing.esilibrary.com? 2012-01-04T12:42:23 dbs: re gitstats, I blame letting branches sit for months before merging. :P 2012-01-04T12:42:43 tsbere: I think you should work to rule! 2012-01-04T12:42:59 *** jamesrf has quit IRC 2012-01-04T12:43:07 dbs: sadly, no ETA ATM. but I will get out my pokin' stick 2012-01-04T12:43:42 eeevil: thanks! 2012-01-04T12:44:10 *** djfiande` has joined #evergreen 2012-01-04T12:44:15 dbs: I can browse to the hostname, but NOT farther. It complained about my unsigned certificate, but can't find the file 2012-01-04T12:44:29 *** djfiande` is now known as djfiander 2012-01-04T12:44:38 * tsbere feels sad that at least some of his stuff was apparently pushed to working back in september........and is still sitting there. 2012-01-04T12:44:49 *** jamesrf has joined #evergreen 2012-01-04T12:45:04 tsbere: yeah, it would suck for that stuff not to make it into 2.2 2012-01-04T12:48:09 dbs: What sucks is that some of it I am waiting on before I build on the affected areas further <_< 2012-01-04T12:49:11 tsbere: also the drift against master can cause pain. If you have good test scenarios that exercise the new functionality, I wouldn't mind trying to work through one or two of those tickets 2012-01-04T12:49:52 * tsbere may need to go and start playing with rebase as a double-check 2012-01-04T12:52:15 *** jeffdavis has joined #evergreen 2012-01-04T12:53:01 jbondharris: How about with http://HOSTNAME/xul/rel_2_1_1/server/main/ ? 2012-01-04T12:54:15 *** djfiander has quit IRC 2012-01-04T12:54:45 *** djfiander has joined #evergreen 2012-01-04T12:54:52 dbs: Let you know in a second or two. 2012-01-04T12:55:40 dbs: https://.../main/ gave me a directory listing now. Don't know what changed... 2012-01-04T12:56:13 jbondharris: hmm. can you try the staff client again? 2012-01-04T12:56:24 jbondharris: also, have you hit the "Add SSL Exception" button on the staff client? 2012-01-04T12:57:23 dbs: Yes, I got that. Here's a wierd message, tho, when I added ws_info.xul: This page uses an unsupported technology that is no longer available by default in Firefox. 2012-01-04T12:57:49 jbondharris: that's remote XUL, not a surprise 2012-01-04T12:58:16 Is this a browser issue? should I be using only certain browsers? (BTW, the page didn't display) 2012-01-04T12:58:59 *** enhancin_ has joined #evergreen 2012-01-04T12:59:05 jbondharris: no, I just wanted to ensure that Apache was at least serving up the files, I didn't expect it to render XUL files 2012-01-04T12:59:28 So the staff client still isn't finding the registration page? 2012-01-04T12:59:28 Is there an easy way to remove all the books and users from the database but keep the org unit structure? 2012-01-04T13:00:29 *** adbowling-isl has quit IRC 2012-01-04T13:01:15 *** tspindler has quit IRC 2012-01-04T13:01:15 *** enhancin has quit IRC 2012-01-04T13:02:47 *** jbondharris has quit IRC 2012-01-04T13:08:34 *** kmlussier has joined #evergreen 2012-01-04T13:11:05 * dbs attempts to remember if grace period calculation changed in 2.1 or in master, will go to the source :) 2012-01-04T13:11:49 2.1 2012-01-04T13:12:05 that would explain why we're suddenly giving people grace periods :) 2012-01-04T13:14:45 * tsbere has rebased those branches of his that needed rebasing in the 2.2 list there 2012-01-04T13:15:49 dbs: Heh, got caught by the upgrade script populating the fine rule's field? 2012-01-04T13:16:54 tsbere: Yup! If we get in the release notes habit, we should probably have an upfront section mentioning new defaults / things you might want to consider configuring _right now_ :) 2012-01-04T13:17:36 heh 2012-01-04T13:20:37 is there a way to change the owning lib on a workstation? I can't do it directly in the database but I can't find anything in the staff client either 2012-01-04T13:21:19 enhancin_: I usually delete the workstation entry in the database, then you have to register again. 2012-01-04T13:21:19 enhancin_: You can delete the database side entry, delete the...I think it is ws_info file from the workstation.... 2012-01-04T13:22:00 yeah, deleting ws_info on the client also works if you don't have database access, but have access to the workstation. 2012-01-04T13:22:05 Or on later versions of evergreen (I forget if it made it into 2.1 or not) add a /id to the end of the hostname, where the id part is anything you want. 2012-01-04T13:23:29 * dbs suspects that we might want to add ADMIN_RECURRING_FINE_RULE to the Circulation Administrator stock perms list 2012-01-04T13:24:39 * tsbere throws out a standard "change the stock ones all you want, but be careful in how you do ugprade scripts for those....if you create any to begin with" opinion statement 2012-01-04T13:26:21 dbs: On the subject of pullrequests, 798255 should, even without an explicit set of "do this" instructions, be easily tested. Turn the flags on, see if data populates when you check an item out. Has the catch of "need to create/populate some stat cats" though. 2012-01-04T13:26:56 tsbere: well, given that the Circulation Administrator group was new with 2.1, I'd say its fair to treat it as an oversight and change the stock perms for that group 2012-01-04T13:27:58 dbs: I argue that changing permissions at any point *other* than when a permission was *added* to the system is a bad thing, and creating new groups shouldn't happen with upgrade scripts anyway. <_< 2012-01-04T13:28:35 AKA, we customized our groups and permissions. Thus we would prefer that upgrade scripts not try and second-guess our decisions. 2012-01-04T13:28:53 tsbere: I agree; having the stock permissions be *templates* that could be copied, and having the *templates* be updated automatically during upgrade would be better IMO 2012-01-04T13:29:20 tsbere: And I counter (as usual: hey, we should do this as an onstage skit at the conference) with "How the hell else are most sites going to get all this great new stuff working?" 2012-01-04T13:29:24 dbs: Also, isn't that permission a global one? I wasn't aware the circulation administrator group was supposed to be for global administration of circ stuff? 2012-01-04T13:29:32 and then gmcharlt has to go and be all reasonable 2012-01-04T13:29:43 tsbere: I'm just following along with what was there 2012-01-04T13:30:10 dbs: I know, isn't it terrible? 2012-01-04T13:30:14 So the max fine and duration rule permissions are granted to it too? 2012-01-04T13:30:14 tsbere: all of the existing perms in the Circ Admin group appear to be depth 0, grantable 2012-01-04T13:30:35 tsbere: aye 2012-01-04T13:30:41 tsbere: could also modify that file I guess. 2012-01-04T13:30:53 tsbere: sorry that was from a while ago. hah 2012-01-04T13:30:55 seriously, another part to my wild idea is giving active permission groups a flag to indicate if they want to inherit or copy from template permission groups 2012-01-04T13:31:10 dbs: See, we don't use that group. I am thus not all that familiar with it. Still, don't touch existing permission groups, IMO, unless it is part of creating the new permissions to begin with (and tied to existing permissions you can search for and such) :P 2012-01-04T13:31:37 I AM however having an issue with finding where to delete users... 2012-01-04T13:32:13 enhancin_: Load user, delete is an option in one of the "actions" type menus. I think. Bulk delete is a different matter. 2012-01-04T13:32:33 * tsbere isn't sure that there is a way to bulk delete.....user buckets, maybe? 2012-01-04T13:32:36 gmcharlt: of course, we need to be capable of creating usable config UIs... you think we can handle slush buckets or something like that? 2012-01-04T13:33:14 * dbs wonders if joelewis can be seduced/coerced back this summer, if he's even eligible 2012-01-04T13:33:14 dbs / gmcharlt: I keep toying with "re-build the entire permission system" as a project..... 2012-01-04T13:33:29 *** edoceo has joined #evergreen 2012-01-04T13:33:38 * Dyrcona has yet to find a more usable config ui than plain text. 2012-01-04T13:34:58 * dbs is mostly with Dyrcona 2012-01-04T13:36:06 Hmm. Thinking about turning our local admins loose on the circ rules to set their own grace intervals, but having tried in the past to rely on a stock set of recurring fine rules I'm sure we're going to end up getting tug-of-war conflicts 2012-01-04T13:36:11 *** edoceo has left #evergreen 2012-01-04T13:36:17 *** edoceo has joined #evergreen 2012-01-04T13:36:52 Should the Event Envronment settings for Action Triggers be allow to contain a space or other special characters? 2012-01-04T13:37:11 which will likely result in "BR1_1_week_50_cents_per_day" with a grace period of 4 hours vs. "BR2_1_week_50_cents_per_day" with a grace period of 1 day and the like 2012-01-04T13:37:56 (and we're still using script-based circ, so I'll still end up having to manage that part until we can cut over to in-db... *sigh*) 2012-01-04T13:38:00 edoceo: A/T event environments basically contain nothing but table and columns references 2012-01-04T13:38:57 so /^[a-z_.]+$/ should suffice for validating the path 2012-01-04T13:44:22 *** jbondharris has joined #evergreen 2012-01-04T13:44:32 gmcharlt: when I am able to edit them in the UI they are entirely free-form; except that when I read that data in my Validator ($env->{target}->usr->stat_cat_entries) the names of the stat cat allow all kinds of chars but the value seems to only be \w+ 2012-01-04T13:46:39 right now when i majke a new workstation I can't select BR1, it only shows BR2, 3, 4 BM, etc 2012-01-04T13:46:58 but we need to use BR1...and it won't allow people to i guess 2012-01-04T13:48:29 *** dzeiger has quit IRC 2012-01-04T13:58:21 phasefx: merge conflicts are indeed not fun. You are at least only a couple months out, I went and fixed some that are 3+ months out today. :( 2012-01-04T14:00:47 * phasefx will go branch hunting tomorrow, see what he can do 2012-01-04T14:01:13 still need to fix my .gitignore/alternative for rel_2_1/etc :) 2012-01-04T14:06:58 yeah, but it doesn't seem to show up in the diff. 2012-01-04T14:07:27 * tsbere suspects that went to the wrong window 2012-01-04T14:08:22 so it did... :) 2012-01-04T14:14:55 -= THIS MESSAGE NOT LOGGED =- 2012-01-04T14:16:40 I'm unable to register workstations with the main organization 2012-01-04T14:16:56 and I can't delete the others that we don't even need, so we can't add books to the right area 2012-01-04T14:17:58 enhancin_: that branch "can_have_users" for its matching actor.org_unit_type? And the user you're logging in with has been granted workstation perms at the branch in question? 2012-01-04T14:20:26 anyone want to signoff & push the last two commits in user/dbs/makefile_and_egcat_tests to master? they're trivial, not like those complicated "add the release notes" branches that I habitually foist on folks 2012-01-04T14:21:05 it has the same ou_type as the others that I can add it to, only the parent is different. i'm looking for the perms at the branch. it should, because i'm trying it with the admin 2012-01-04T14:21:21 * tsbere will consider looking at a few branches....after he finishes writing this "we can't give you reports if you don't tell us what you want" message 2012-01-04T14:24:12 so in the work per missions they don't have permissions for 1, 2, or 3 but for 4-9. why are the 1,2,3 missing? 2012-01-04T14:24:50 gmcharlt: oops, I way mis-spoke - I mean about the Event Parameters section, that is where it appears my value is getting mangled 2012-01-04T14:30:44 edoceo: ah - the event parameter value can be (essentially) any Perl expression 2012-01-04T14:30:56 and is in fact evaluated as such 2012-01-04T14:31:50 Yea, Param is coming in properly, with spaces and all but my Value is mangled 2012-01-04T14:32:20 *** djfiander has quit IRC 2012-01-04T14:32:30 dbs: I can't get it to display the BR1 when creating the workstation name. it can have users and i'm using the admin 2012-01-04T14:34:46 gmcharlt: I'm going to write to the list on this, paragraphs of info.... 2012-01-04T14:35:20 *** djfiander has joined #evergreen 2012-01-04T14:42:00 Is there an easy way to remove all the books and users from the database but keep the org unit structure? 2012-01-04T14:42:34 enhancin_: Easy is relative. 2012-01-04T14:43:24 Dyrcona: well, i can handle most I just want to make it quick and painless. I'm not super familiar with postgresql so I assume I can't just pull specific tables out, restore to pristine, and then restore those tables? 2012-01-04T14:43:29 that would cause issues, right? 2012-01-04T14:43:38 enhancin_: you could pg_dump the actor schema. rebuild the database and then pg_restore the actor schema. 2012-01-04T14:43:47 ah so I CAN do that 2012-01-04T14:43:57 that will, however, not get rid of users. 2012-01-04T14:44:27 Could I modify the pg_dump file and remove the users and their addresses? 2012-01-04T14:44:30 you would need to dump the actor schema tables except for those that concern users. 2012-01-04T14:44:43 or that 2012-01-04T14:45:06 that sounds good, I just wanted to make sure I can do it. postgresql makes me feel like I'm in inception because I've been using MySQL my whole life and now there are schemas and i get confused. 2012-01-04T14:46:17 enhancin_: Yes, it is an adjustment. I've used a couple of other DBMS products, and while schemas don't throw me, little things like isnull(), ifnull(), coalesce(), etc. sometimes throw me. 2012-01-04T14:46:24 You could create your own customized version of 950.seed_data.sql 2012-01-04T14:46:43 dbs++ 2012-01-04T14:46:44 Back again -- still can't get Evergreen client to access EG website. From log files, I can log in with my admin account, but the ws_info.xul file cannot be found 2012-01-04T14:46:59 Then when you rebuild the database schema, you get exactly what you want for any of the settings 2012-01-04T14:47:05 That's a good suggestion, or make a script like one I pastebinned earlier. 2012-01-04T14:47:42 http://pastebin.com/m4C07r2b 2012-01-04T14:48:02 Yeah, we had developed a SQL script to update / insert actor.org_unit* back when we were migrating too 2012-01-04T14:48:02 has anyone encountered an error like this before?: http://paste.lisp.org/display/126866 2012-01-04T14:48:14 we're on 2.1/PG9.1 2012-01-04T14:48:35 this is what the logs say when I try to add a hold policy 2012-01-04T14:49:49 csharp: never seen that one. 2012-01-04T14:50:29 omg...950.seed_data.sql. while true ; do dbs=$(dbs+1) ; done ; 2012-01-04T14:50:31 I'd try Google with just the Apache2....abort part. 2012-01-04T14:50:49 enhancin_: heh, see http://bzr.coffeecode.net/unicorn2evergreen/drop_patrons.sql for my ancient "drop patrons from the database" script 2012-01-04T14:51:26 I'm sure it would no longer function, as the schema has evolved significantly since 2009-02, but the idea is there if you really want it 2012-01-04T14:51:53 looks useful, however we have a lot of book test records we have to remove as well : ) 2012-01-04T14:52:08 enhancin_: if you're using git, I'd suggest making a branch in your working repo for you customized 950.seed_data.sql. 2012-01-04T14:52:45 we do a lot of different libraries so I just have separate source packages on each machine that i've changes 2012-01-04T14:54:42 git is your friend. 2012-01-04T14:55:36 well it's really specific to just one server so we just make the modifications there 2012-01-04T14:56:06 git is a very good friend 2012-01-04T14:56:20 heh, just never got into it I guess 2012-01-04T14:58:49 * dbs thinks its time to create a .gitignore for OpenSRF, too 2012-01-04T14:59:42 So, I can log in from a client, but not see anything. I will put my config files in pastebin - could someone take a look and see if there is an obvious problem? 2012-01-04T15:00:03 * eeevil thinks .gitignore should ignore vim swap files 2012-01-04T15:01:47 whee. the new WWW::Proxy::Authen module works with my WWW::PhoneList module/ 2012-01-04T15:02:05 * Dyrcona smirks at smiley icons. :) 2012-01-04T15:02:20 tsbere++ 2012-01-04T15:02:55 jbondharris: you can log in from a client, but not see anything? Not sure I follow 2012-01-04T15:03:24 * Dyrcona was thinking of asking for a post of a screen shot somewhere. 2012-01-04T15:04:35 dbs: By looking at the log files, the actual login is successful from the client. BUT the web interface reports that the server/main/ws_info.xul file cannot be found 2012-01-04T15:05:16 TypeError: g.my_libs_tree is null 2012-01-04T15:05:16 So, it looks like nothing worked, but it seems felated to the servers web files 2012-01-04T15:05:51 enhancin_: autogen.sh usually fixes that. 2012-01-04T15:05:54 jbondharris: I would concur. Anything in the apache error log? 2012-01-04T15:06:01 duh..i didn't even think about that 2012-01-04T15:06:11 i was completely changing the whole library structure and didn't run that -.- 2012-01-04T15:06:57 I've run autogen several times, without any errors. I can do it again... 2012-01-04T15:08:51 Dyrcona: Still nothing. Is it because I don't have a SYS2 anymore? it seems like you can only put workstations on SYS2... 2012-01-04T15:09:55 does it jsut ignore the main one or something? 2012-01-04T15:10:34 enhancin_: sounds like your tree is incorrect in the database or something. 2012-01-04T15:10:55 well I tried to make my own custom one because we don't need 30 different branches like the example puts in there 2012-01-04T15:10:58 we just need the main one 2012-01-04T15:11:05 enhancin_: triple check the depths and parents of your org_units. 2012-01-04T15:11:17 dbs: autogen output is in http://pastebin.com/8tFNvDpm 2012-01-04T15:11:47 ls 2012-01-04T15:13:29 yeah I have one Consortium, one System, and one Branch. the Branch's parent is System, and System's parent is Consortium 2012-01-04T15:13:41 jbondharris: that output looks fine to me. 2012-01-04T15:13:55 enhancin_: the branch's parent is? 2012-01-04T15:14:05 System 2012-01-04T15:14:13 Sorry [ls] - I have also run the settings-tester.pl script without finding any problems. There's gotta be something about how Apache is set up that is just slightly different from a typical system. 2012-01-04T15:14:45 Dyrcona: org_unit_type table is the same as default, i just changed it to only have those three in the org units 2012-01-04T15:15:07 enhancin_: what about can_have_users and can_have_volumes or whatever their exact names are? 2012-01-04T15:15:30 hmm, do i have to set the Consortium to that? that's the only one that is not 2012-01-04T15:15:40 enhancin_: no. 2012-01-04T15:15:55 enhancin_: did you change shortnames? 2012-01-04T15:15:59 yeah 2012-01-04T15:16:15 enhancin_: and you are sure you're using the new shortnames? 2012-01-04T15:16:39 enhancin_: you quit the client and restarted? 2012-01-04T15:17:02 do i need to completely restart osrf as well? 2012-01-04T15:17:19 ah-ha! 2012-01-04T15:17:26 enhancin_: might not be a bad idea.... 2012-01-04T15:17:37 enhancin_: autogen.sh -u? 2012-01-04T15:17:55 and restart of the staff client 2012-01-04T15:25:55 berick: you wanna give user/dbs/README_202 in the OpenSRF working repo a once-over? Would like to prep for the coming 2.0.2 release 2012-01-04T15:26:13 dbs: on it 2012-01-04T15:35:59 *** jbondharris has quit IRC 2012-01-04T15:36:12 dbs: looks good to me. 2012-01-04T15:36:25 *** jbondharris has joined #evergreen 2012-01-04T15:47:16 Any Apache experts on-line? My EG clients can't access the server files 2012-01-04T15:48:04 *** jhaig has joined #evergreen 2012-01-04T15:50:22 jbondharris: you should post your apache conf and logs from /var/log/apache2/* 2012-01-04T15:50:30 post to pastebin, not here! 2012-01-04T15:51:13 Hello. I'm having problems with connecting via the staff client. I'm getting a 'Network or system' failure but I cannot understand the error. I should point out that I have recently made changes to the network and, although the evergreen server and database server have not moved, the staff client is now the other side of a router. For the moment, though, I have moved it back onto the same network. 2012-01-04T15:51:53 I'm putting the error in pastebin.ca but it is taking forever to submit :-/ 2012-01-04T15:52:43 _bott_: is there a version of your jspac copy Actions code for EG 2.1? 2012-01-04T15:54:37 Couldn't get either of the pastebins in the topic here to respond so I borrowed one from another channel: http://pastebin.centos.org/38235 2012-01-04T15:55:36 jhaig: wow, that's some error 2012-01-04T15:56:12 <_bott_> phasefx: Actions code? ...off to a meeting, but all of our jspac code is out here: https://github.com/grpl-eg/rel_2_1/tree/master/web/opac/skin/grpl 2012-01-04T15:56:22 cd .. 2012-01-04T15:56:27 tsbere: By the way, thanks for the answer several weeks ago. I did find it via a log of the channel but I have only now logged back on and had it given to me by pinesol_green. :-) 2012-01-04T15:56:28 _bott_: muchos gracias. Action Buttons 2012-01-04T15:56:58 * tsbere doesn't even remember what jhaig is talking about, must have been a while ;) 2012-01-04T15:57:28 *** Meliss has quit IRC 2012-01-04T15:57:32 tsbere: About volumes and bibs and copies. 2012-01-04T15:57:39 jhaig: DBD::Pg::st execute failed: no connection to the server, that's talking about the database.. but it's a weird error if the client works for you in one circumstance and not another 2012-01-04T15:58:06 From the evergreen server I can connect to the database from the command line with psql 2012-01-04T15:58:17 Hmmm....no connection to server, works sometimes but not others....... 2012-01-04T15:58:28 jhaig: You run out of DB connections at the postgres level? 2012-01-04T15:58:58 Unlikely. It only has the one evergreen server as a client. 2012-01-04T15:59:36 Currently this is a 2 server system with one staff client, as someone is adding books before going live. 2012-01-04T16:00:08 I thought I may have messed something up with the change of the networking but now everything is back as it was. 2012-01-04T16:00:57 Well, storage can't connect, but cstore/pcrud likely can... so check the configs for storage? 2012-01-04T16:02:07 dbs: Feeling scrollback-lazy today. You had some specific branch you poked about (don't recall a LP bug number associated with it, so I doubt I will find it there) 2012-01-04T16:02:37 tsbere: Erm, are the configs for storage part of the evergreen configs? 2012-01-04T16:03:04 edoceo: I pasted a tar file to http://pastebin.com/jnf70a5N 2012-01-04T16:03:10 tsbere: user/dbs/makefile_and_egcat_tests perhaps? 2012-01-04T16:03:29 jhaig: opensrf.xml, I believe, contains them (or should) 2012-01-04T16:04:55 dbs: That looks like what I think you were poking about 2012-01-04T16:05:06 * tsbere goes to take a look 2012-01-04T16:05:10 very low-hanging fruit 2012-01-04T16:06:01 The bits inside in the section look correct. 2012-01-04T16:06:17 edoceo: And the log files are in http://pastebin.com/MkYpwj7T (another tar file) 2012-01-04T16:06:24 * dbs laughs at doc/Roadmap.txt in OpenSRF: "Roadmap for OpenSRF === The high points as of Febuary 2005" 2012-01-04T16:07:31 * dbs never noticed doc/OpenSRF-Messaging-Protocol.html before either 2012-01-04T16:09:46 *** collum has quit IRC 2012-01-04T16:13:10 has anybody done an elisp binding for OpenSRF? 2012-01-04T16:13:19 jbondharris: your apache error log has a ton of bad errors 2012-01-04T16:13:36 suggesting 1) that you haven't set the apache user to "opensrf" at a minimum 2012-01-04T16:15:00 dbs: The ones to pay attention to are the latest ones. I was a bad citizen by not clearing out the old log entries first. I apologize. 2012-01-04T16:16:20 dbs: envvars has APACHE_RUN_USER=opensrf -- wasn't that where it happens?? 2012-01-04T16:16:28 library 2012-01-04T16:16:33 *** jenny has quit IRC 2012-01-04T16:17:06 yep. but that was earlier, as you mentioned, so might not be an issue. Hard to sort out the valid from what's posted there 2012-01-04T16:17:31 jbondharris: any chance you can stop apache, clear the apache logs, start apache, and try a connection from the staff client to get the bare minimum? 2012-01-04T16:17:42 * dbs needs to scoot home but will check IRC logs later 2012-01-04T16:18:19 *** dbs has quit IRC 2012-01-04T16:19:18 Folks, I need to go, and will not be back until next Tuesday. I will restart the VM, clear the logs, and repost the problem and logs next week. 2012-01-04T16:19:50 I really appreciate your help, and I'm sorry if I made it more difficult than it needs to be. Have a good week. 2012-01-04T16:19:55 *** jbondharris has left #evergreen 2012-01-04T16:28:46 *** akilsdonk has quit IRC 2012-01-04T16:29:25 *** djfiander has quit IRC 2012-01-04T16:30:52 OK, so it looks like my problem was down to moving the staff client onto a different network (and a different IP address) - I had put it back on the same network when I got the error earlier but now I have restarted evergreen as well and it connected OK. If I move the client again do I need to change something in the database to let it connect? 2012-01-04T16:38:07 maybe in the firewall, or DNS, but not in the database 2012-01-04T16:40:33 Hmmm, looks like it will just involve a bit of experiment (which will involve me getting disconnected from here). Thanks and bye (for now). 2012-01-04T16:41:46 *** jhaig has quit IRC 2012-01-04T16:45:01 *** jenny has joined #evergreen 2012-01-04T16:53:52 *** kmlussier has quit IRC 2012-01-04T17:11:24 *** sal_ has joined #evergreen 2012-01-04T17:29:23 *** Dyrcona has quit IRC 2012-01-04T17:46:35 *** jenny has quit IRC 2012-01-04T17:52:09 *** yboston has quit IRC 2012-01-04T19:26:53 *** sal_ has quit IRC 2012-01-04T19:43:18 *** youdonotexist has quit IRC 2012-01-04T20:18:18 *** youdonotexist has joined #evergreen 2012-01-04T21:21:45 *** youdonotexist has quit IRC