2011-05-05T00:07:11 @later tell berick This is the problem I get when trying to use the makefile target to install PG9 for EG 2.1 using the deb backports source specified in the readme: http://paste.lisp.org/+2LYQ 2011-05-05T00:07:11 bshum: The operation succeeded. 2011-05-05T00:11:50 @later tell berick As you suggested, that's using the new deb entry as updated in the README. My attempts are on a fresh Squeeze install too, so not much is on that VM. 2011-05-05T00:11:50 bshum: The operation succeeded. 2011-05-05T00:18:05 bshum: do you have postgresql-8.4 already installed? 2011-05-05T00:21:46 you could try adding libpq5 libpg-dev to the list of packages to be installed, to see what happens there 2011-05-05T00:23:27 dbs: That VM doesn't have PG-8.4 installed, but I can give that a whirl. 2011-05-05T00:23:38 no, you don't want 8.4 on there 2011-05-05T00:24:22 Eek, I meant I'll give your second suggestion a try :) 2011-05-05T00:25:27 hrm. 2011-05-05T00:25:43 hrm? 2011-05-05T00:25:45 can't break sip in the way that it was reported to me. 2011-05-05T00:26:10 checking to see how much it might just be 1.6 vs trunk 2011-05-05T00:26:17 but it seems to be a basic thing 2011-05-05T00:26:31 poking and prodding and "hrm"ing 2011-05-05T00:30:20 dbs: Installing those other two packages first didn't seem to help, it still claims those dependency problems I listed earlier. 2011-05-05T00:30:51 bshum: hmm. can you try the command without the -yq flags and step through the various recommendations? 2011-05-05T00:31:00 Okay 2011-05-05T00:31:16 there might be something in there, past the first one, that leads to nirvana 2011-05-05T00:32:40 postgresql-server-dev-9.0: Depends: libpq-dev (>= 9.0~) but 8.4.7-0squeeze2 is installed. 2011-05-05T00:32:43 Same for the libpq5 2011-05-05T00:33:11 It asks me to keep the following 5 packages at their current version, the PG9 packages as not installed) 2011-05-05T00:33:26 The prompt is Y/n/q/? for the "Accept this solution?" 2011-05-05T00:35:14 ah - libpq-dev 8.4.7 is installed? 2011-05-05T00:35:24 Hmm, sounds like it 2011-05-05T00:35:28 removal time 2011-05-05T00:38:23 Okay, removed 2011-05-05T00:44:29 and if you try that long installation line now? 2011-05-05T00:46:02 Still doesn't like it. 2011-05-05T00:46:09 Complaining about those two packages 2011-05-05T00:46:33 paste? 2011-05-05T00:46:36 okay, SIP2 skills leveled up. i see packet captures in my day tomorrow. 2011-05-05T00:47:08 unfortunately i don't see sleep in my night tonight :P 2011-05-05T00:49:03 I slept from 6:30 - 9:30. bad sleep patterns 2011-05-05T00:49:22 *** eeeevil has quit IRC 2011-05-05T00:49:28 dbs: It's the same thing as before, but I'll annotate the paste - http://paste.lisp.org/display/121778#2 2011-05-05T00:49:49 Oh, should I remove that -yq again? 2011-05-05T00:51:03 bshum: if you run "aptitude" and search for libpq5 ("/libpq5"), what do you see? 2011-05-05T00:51:34 I see "libpq5 9.0.4-1~bp" (meaning 9.0.4-1 from backports) 2011-05-05T00:54:13 Hmm 2011-05-05T00:54:25 The following packages depend on libpq and are broken: 2011-05-05T00:54:42 libpq-dev depends on libpq5 (= 8.4.7-0squeeze2) 2011-05-05T00:55:06 Guess that's where I'm broken 2011-05-05T00:56:33 huh. I don't have libpq-dev installed 2011-05-05T00:56:57 Maybe something in the Squeeze installer is broken on my VM 2011-05-05T00:57:08 And included something that it didn't really need 2011-05-05T00:57:24 I think it included the laptop + base system stuff 2011-05-05T00:57:38 I don't have postgresql-server-dev-9.0 installed either 2011-05-05T00:57:49 drop that from the install line 2011-05-05T00:59:40 Okay, that takes away the libpq-dev dependency complaint, but this still remains: postgresql-client-9.0: Depends: libpq5 (>= 9.0~) but 8.4.7-0squeeze2 is to be installed. 2011-05-05T01:01:37 I wonder if there are any other packages that we're installing that are creating that dependency 2011-05-05T01:04:21 It seems like my libpq5 is stuck with the original 8.4.7 squeeze version, and not the 9.0.4 one from backports. 2011-05-05T01:04:23 I thought you had removed libpq-dev? 2011-05-05T01:04:34 remove libpq5 too, for good measure 2011-05-05T01:05:01 Both were removed 2011-05-05T01:05:11 and try "aptitude -t squeeze-backports install libpq5" 2011-05-05T01:05:31 That did it. 2011-05-05T01:05:44 And now all is happy 2011-05-05T01:05:58 hmm. not looking forward to that Makefile.install change 2011-05-05T01:08:41 *** pmplett has quit IRC 2011-05-05T01:09:21 *** pmplett has joined #evergreen 2011-05-05T01:09:35 dbs++ Thanks for helping me poke at Squeeze stuff 2011-05-05T01:09:47 bshum: thanks for running the gauntlet 2011-05-05T01:09:55 It seems that Debian has been meaner to me than Ubuntu ;) 2011-05-05T01:10:03 But Ubuntu has been mean to me in other ways 2011-05-05T01:10:50 I'll try the rest of the install steps tomorrow for Squeeze and report back. 2011-05-05T01:10:53 Or later today I guess 2011-05-05T01:11:03 Goodnight! 2011-05-05T01:12:02 gnite 2011-05-05T02:42:51 *** Jbergy has quit IRC 2011-05-05T03:11:14 *** gmcharlt has quit IRC 2011-05-05T03:11:54 *** gmcharlt has joined #evergreen 2011-05-05T03:57:47 *** collum has quit IRC 2011-05-05T06:04:14 *** RBecker has quit IRC 2011-05-05T06:06:02 *** RBecker has joined #evergreen 2011-05-05T07:18:29 *** rsinger has joined #evergreen 2011-05-05T07:47:12 *** JMCraig has joined #evergreen 2011-05-05T07:48:11 *** JohnC has quit IRC 2011-05-05T07:48:12 *** JMCraig is now known as JohnC 2011-05-05T08:17:49 replacement_fans++ 2011-05-05T08:35:44 *** AaronZ-PLS has joined #evergreen 2011-05-05T08:43:28 *** Dyrcona has joined #evergreen 2011-05-05T08:56:07 *** collstech has joined #evergreen 2011-05-05T08:58:22 Good morning all. I've been crawling through documentation and IRC logs, but can't find a way to import holdings from Follett to Evergreen. Can anybody help me with this? 2011-05-05T09:00:06 *** JohnC has left #evergreen 2011-05-05T09:18:11 collstech: I am willing to share everything I know about that process. Unfortunetly, what I know can be summed up as follows: What is Follett? 2011-05-05T09:19:41 tsbere: lol...thanks. Follett is our current library system and one of the largest around. 2011-05-05T09:22:54 Never heard of it personally. 2011-05-05T09:23:01 Heh, I've heard of it. 2011-05-05T09:23:05 *** jenny has joined #evergreen 2011-05-05T09:23:24 But a quick search tells me it is aimed at schools. I don't work with school libraries right now. 2011-05-05T09:23:35 Which probably explains why I haven't heard of it. 2011-05-05T09:24:28 collstech: you might want to focus on how to get the data out of Follett; after that, getting it into Evergreen is more or less documented (at least on mailing lists and such) 2011-05-05T09:24:57 collstech: for example, the Koha mailing list / wiki / docs might have some examples of getting data out 2011-05-05T09:25:11 Well, I have the data from Follett and I've been able to add it to Evergreen. I'm just having problems with the actual holding data transfering over. 2011-05-05T09:25:51 collstech: okay, then the best thing you could do would be to provide an annotated sample of what Follett's holdings data looks like 2011-05-05T09:26:23 preferably on the mailing list - and then someone should be able to give you a hint about how that fits into asset.copy / asset.call_number etc 2011-05-05T09:26:41 * dbs rolls 2011-05-05T09:28:07 *** yboston has joined #evergreen 2011-05-05T09:28:32 OK, well thanks. I appreciate the information. 2011-05-05T09:30:59 *** dbs has quit IRC 2011-05-05T09:31:50 *** collstech has quit IRC 2011-05-05T09:36:44 *** bgoble has joined #evergreen 2011-05-05T09:53:57 *** Sally has joined #evergreen 2011-05-05T09:59:42 *** dbs has joined #evergreen 2011-05-05T10:11:12 happy 2.0.6/2.1-beta1 release day, folks! anyone have any last-minute thing we need to squeeze in? 2011-05-05T10:17:03 would a framework for redirection of merged bibs make sense, rather than "this record has been deleted" style messages? for patrons, auto-redirect, for staff give an optional redirect? 2011-05-05T10:17:18 thinking out loud, and not as a response to your question, eeevil 2011-05-05T10:17:39 eeevil: I would not say no to my lazy circ or my new branch: http://git.mvlcstaff.org/?p=tsbere/ILS.git;a=shortlog;h=refs/heads/patron_info_copy 2011-05-05T10:19:44 eeevil: Also wouldn't say no to my sip_statcats, for that matter. 2011-05-05T10:21:41 *** atheos has quit IRC 2011-05-05T10:23:57 eeevil: there are also two patches i'd like to try to make happen for beta1: iisg's diacritics feature and atheos' phone number search patch (with index). what say you to noon cutoff? 2011-05-05T10:27:39 tsbere: has anyone had a chance to review those? I haven't had tuits (and won't today) 2011-05-05T10:27:45 senator: noon sounds fine 2011-05-05T10:28:18 eeevil: The latest one (patron_info_copy) is almost trivial. The others.....beyond me, I dunno. 2011-05-05T10:29:16 tsbere: did you share a use case for the phone number patch? i wondered where you found it most useful... 2011-05-05T10:29:43 for the phone number patch? 2011-05-05T10:30:02 As in the extra "click to copy" stuff? 2011-05-05T10:30:02 yes, a use case for the phone number patch. 2011-05-05T10:30:26 * tsbere isn't positive what jeff is asking about 2011-05-05T10:30:27 tsbere: sorry, no. the patron phone number search patch. 2011-05-05T10:30:40 jeff: That isn't even my patch. ;) 2011-05-05T10:30:55 doh. 2011-05-05T10:30:59 thanks for the correction. 2011-05-05T10:31:43 that's atheos' -- the use case as i understand it is looking up patron by phone number and not needing to know what punctuation was used to put the phone number in the system, when phone numbers are not consistent in the db and there's some kind of organizational issue blocking a good cleanup 2011-05-05T10:34:10 i wonder if it would be useful to still have it left-anchored (maybe it still is), so that you could find 231-555-1212 x.123 and 231.555.1212 x.123 and 231-555-1212 with the same search on 2315551212 2011-05-05T10:34:55 seems like it would be. if it isn't left anchored now i can make it so 2011-05-05T10:35:17 well, 2011-05-05T10:35:20 atheos might require an exact match for his purposes. i don't know. 2011-05-05T10:35:28 assuming a index can make that fast 2011-05-05T10:35:42 in the middle of something else so brain power diverted 2011-05-05T10:36:00 yeah. i'll mention it on the list and forget about it myself. 2011-05-05T10:36:17 senator: it becomes magically left-anchored 2011-05-05T10:36:57 eeevil: cool. i guess there is one other concern before i volunteer to make that change though: do some/all the field in the patron search already work in a left-anchored way? i guess introducing one inconsitent field would be bad 2011-05-05T10:37:01 haven't used it in a while 2011-05-05T10:37:11 and i see that atheos' patch isn't undoing that magic, so i'm done here. :) 2011-05-05T10:37:28 senator: all 2011-05-05T10:37:38 eeevil, jeff: ah ok cool 2011-05-05T10:37:49 senator: most of patron crazy search does magical left-anchoring. i've got a branch somewhere that gives an "exact name match" option. 2011-05-05T10:41:55 jeff: share share :-) maybe not in time for 2.1 :-( 2011-05-05T10:45:26 eeevil: there are a few tweaks to Makefile.install and the install instructions that will be necessary for a successful install of postgresql 9.0, based on bshum's attempts early this morning 2011-05-05T10:46:09 On Squeeze anyways ;) 2011-05-05T10:49:24 bshum: we pulled support for lenny from Makefile.install, so a working squeeze install is kind of important 2011-05-05T10:50:24 dbs: shall I wait, or is that fixable by documentation for b1? 2011-05-05T10:51:25 * berick wonders why http://paste.lisp.org/new/evergreen is keen on gobbling up my pastes 2011-05-05T10:51:46 berick: yeah, you have to go back to paste.lisp.org to pick up the link atm 2011-05-05T10:51:47 bshum: dbs: my squeeze installer does this and it seems to work ok: 2011-05-05T10:51:48 apt-get -yq -t squeeze-backports install postgresql-$PG_VERSION postgresql-client-$PG_VERSION postgresql-contrib-$PG_VERSION postgresql-plperl-$PG_VERSION libpq-dev 2011-05-05T10:52:04 dbs: ah, ok 2011-05-05T10:52:08 berick: nice, that's the direction I was heading 2011-05-05T10:52:13 but you rocked it 2011-05-05T10:54:48 eeevil: We can say it's fixable via documentation; it will be a hurdle, but being able to test the rejiggering of Makefile.install is going to take a reasonably significant amount of time 2011-05-05T10:56:36 berick++ # for keeping https://dev198.esilibrary.com/eg/opac up-to-date - and debranded! 2011-05-05T10:57:14 i have a squeeze clone i could test if the Makefile.install is ready to go 2011-05-05T10:57:24 with commented-out up-front pg installing... 2011-05-05T10:58:16 berick: it's not ready to go yet (not from me, anyway) 2011-05-05T10:59:18 Every time I look at that thing, more weird / ugly bits jump out at me 2011-05-05T10:59:25 dbs: do we need to redirect /eg/opac to /eg/opac/home (or myopac) ? /eg/opac was never intended as a landing spot and it seems to result in problems after login 2011-05-05T10:59:48 s/problems/404's/ 2011-05-05T11:00:35 or, optionally, just don't use /eg/opac directly 2011-05-05T11:00:36 berick: sure, a redirect makes sense; that way sites that want to encourage logging in can -> myopac, and sites that don't -> home 2011-05-05T11:00:50 we could leave that as an apache config 2011-05-05T11:00:56 exactly what I was thinking 2011-05-05T11:01:38 at some point we need to sort out multiple skins & locales 2011-05-05T11:01:47 * dbs will add to the todos 2011-05-05T11:02:39 indeed. some support exists for both, but it needs some love 2011-05-05T11:03:27 * dbs changes to "Expose support for ..." 2011-05-05T11:03:59 * dbs hasn't really looked for that support yet but is glad to hear something is there 2011-05-05T11:04:27 ugh, late for a meeting - part of the reason I'm not on top of Makefile.install / README 2011-05-05T11:04:31 biab 2011-05-05T11:04:54 phasefx: you around? 2011-05-05T11:05:11 i have xul questions for you momentarily if so 2011-05-05T11:06:10 senator: I can take a crack at them if you want. 2011-05-05T11:06:57 tsbere: ok thanks. just a sec here and i'll paste the code and describe the issue 2011-05-05T11:07:44 tsbere: code: http://nox.esilibrary.com/~lebbeous/joscha_0.patch.txt 2011-05-05T11:07:46 issue: 2011-05-05T11:08:44 see the change to marcedit.xul ? basically, that patch is supposed to popup a little menu full of buttons with diacritically marked characters when you press ctrl-s while editing a field in the marc editor. it's someone else's code so i'm not sure of all the parts, however 2011-05-05T11:09:27 in marcedit.xul i removed the onkeypress attribute from the xhtml:textarea element underlying the marc editor 2011-05-05T11:09:33 to get the ctrl-s thing to work 2011-05-05T11:09:42 that won't do though 2011-05-05T11:09:52 the onkeypress value is important 2011-05-05T11:10:43 not sure what's incompatible with what 2011-05-05T11:12:55 Looks more like a js issue 2011-05-05T11:13:31 well, a js issues specific to xulish events 2011-05-05T11:14:22 Lets see 2011-05-05T11:14:28 (side note about that patch ... it removes oils_lock_page() support in at least one place ... maybe more. need to let that come back) 2011-05-05T11:15:03 i think that's exactly what i was just talking about ? 2011-05-05T11:15:13 senator: ah! ok ... sorry 2011-05-05T11:16:03 senator: Well, one option is to move the code you removed into the event listener on text-editor-box in symbol_overlay.js, in an "else" on the if statement. 2011-05-05T11:16:09 basically i'm trying to figure out why the event that leads to oils_lock_page() is incompatible with diacritics menu and/or how to make the two play together 2011-05-05T11:16:41 tsbere: aha, will try that 2011-05-05T11:21:02 senator: or ... you could do the same thing that the popup code is doing via dojo, with addEventListener in symbol_overlay.js 2011-05-05T11:21:41 (or even use that event listener) 2011-05-05T11:23:45 hrm.. might be better to simply move /everything/ to the main one 2011-05-05T11:26:02 * senator tries it that way for neatness 2011-05-05T11:26:41 *** youdonotexist has joined #evergreen 2011-05-05T11:29:36 ok it seems to be working reasonably correctly. thanks guys 2011-05-05T11:36:50 Anyone have thoughts on my patron_info_copy branch? Was a bit of a request from one of our libraries. 2011-05-05T11:37:55 I will add it to my queue of things to pull. if eeevil comes down with the guillotine at noon even though ... :-) 2011-05-05T11:38:17 there may be some lag .. gotta eat 2011-05-05T11:45:53 *** youdonotexist has quit IRC 2011-05-05T11:46:27 *** youdonotexist has joined #evergreen 2011-05-05T11:47:07 *** youdonotexist_ has joined #evergreen 2011-05-05T11:48:12 senator: I don't think Joscha included a DCO or an explicit license for his work? 2011-05-05T11:49:11 (asking for that has been on my to-do list, had a draft message together but wanted to include sane comments on the code at the same time) 2011-05-05T11:50:33 gah no. i can seek it now. if he did that work under contract for someone else (iish) i don't suppose one from iish would do? 2011-05-05T11:50:41 *** youdonotexist has quit IRC 2011-05-05T11:50:41 *** youdonotexist_ is now known as youdonotexist 2011-05-05T11:51:23 regarding sane comments and bugfixes, i hate to shove something in like this, but i hoped it would get that in the beta phase 2011-05-05T11:51:42 since any later might be too late to just add the feature outright 2011-05-05T11:53:48 Understood, with the conference etc. we've all been plenty busy :) 2011-05-05T11:53:54 and it's a very nice feature! 2011-05-05T11:54:08 reducing_user_pain__ 2011-05-05T11:54:11 err, ++ 2011-05-05T11:54:33 reducing_user_pain++ # for the pinesols 2011-05-05T11:55:05 I think the pinesols are much dumber than zoia and don't understand karma_bumps++ that aren't at the end of a line 2011-05-05T11:55:09 @karma karma_bump 2011-05-05T11:55:09 dbs: An error has occurred and has been logged. Please contact this bot's administrator for more information. 2011-05-05T11:55:13 heh 2011-05-05T11:55:25 @karma reducing_user_pain 2011-05-05T11:55:25 dbs: An error has occurred and has been logged. Please contact this bot's administrator for more information. 2011-05-05T11:55:31 @karma berick 2011-05-05T11:55:31 dbs: An error has occurred and has been logged. Please contact this bot's administrator for more information. 2011-05-05T11:55:37 okay, okay 2011-05-05T11:55:39 heh bummer 2011-05-05T11:56:29 pinesol_green-- 2011-05-05T11:56:45 * moodaepo goes to check on pinesol_green 2011-05-05T11:58:07 *** Petaris has joined #evergreen 2011-05-05T11:59:35 Hello 2011-05-05T12:00:38 I am trying to use the Evergreen 2.0.3 Squeeze virtual machine with VMware but am having troubles with the OVF file 2011-05-05T12:01:17 VMware's OVFtool keeps complaining about incompatible or not supported bits in it 2011-05-05T12:02:02 I Googled and found this: http://www.vmware.com/appliances/getting-started/learn/ovf.html but it didn't work for me 2011-05-05T12:02:38 Is there a known good procedure for making the OVF file work with VMware ESXi or VMware Fusion? 2011-05-05T12:03:21 Petaris: could you give the exact error? 2011-05-05T12:03:44 sure, sec 2011-05-05T12:03:55 starting over with a fresh copy of the OVF file 2011-05-05T12:04:25 mrpeters-isl: Error: OVF Package is not supported by target: 2011-05-05T12:04:25 - Line 36: Unsupported hardware family 'virtualbox-2.2'. 2011-05-05T12:04:29 i'd imagine OVF Tool would work 2011-05-05T12:05:05 after I fix that to say vmx-07 I get more errors 2011-05-05T12:05:56 eeevil: re: beta1, there is some fleshing out of functionality I'd like to do for unified vol/copy editor, etc. that might could be considered bug fixes, and thus safe to go in after beta1 (so as to not hold you up) 2011-05-05T12:06:01 Error: OVF Package is not supported by target: 2011-05-05T12:06:01 - Line 77: OVF hardware element 'ResourceType' with instance ID '5': No support for the virtual hardware device type '20'. 2011-05-05T12:06:08 eeevil: these are the indexes you felt niles' phone number search thing needed ? http://pastebin.com/4F4DvUbU 2011-05-05T12:06:21 mrpeters-isl: Those errors are from ovftool 2011-05-05T12:06:34 yeah 2011-05-05T12:06:51 senator: indeed they are 2011-05-05T12:06:52 mrpeters-isl: I'm running ./ovftool -tt=vmx /source /dest 2011-05-05T12:06:56 eeevil: kewl thanks 2011-05-05T12:07:27 http://www.virtualbox.org/ticket/7982 2011-05-05T12:07:40 maybe try that --lax commandline option to get some more error info 2011-05-05T12:07:43 Petaris: I've had issues converting from vbox to vmware - just gave up. 2011-05-05T12:08:01 yeah, proabbly easier to build from source in a vmdk haha 2011-05-05T12:08:23 phasefx: is one of those bug fixes providing more than just Dewey call numbers, hopefully? :) 2011-05-05T12:08:59 mrpeters-isl: I'll try that option and see what happens 2011-05-05T12:09:38 dbs: err, maybe? I remember you mentioning something along those lines, but it's not in the list I have handy 2011-05-05T12:09:58 phasefx: we can always sort it out post-beta. might be a config issue, who knows 2011-05-05T12:10:23 mrpeters-isl: Well the --lax seems to have worked 2011-05-05T12:10:40 I'll try the VM and see if its valid or not 2011-05-05T12:13:47 Petaris: awesome - if it does, please share the exact command that worked for you and I'll add that to future READMEs 2011-05-05T12:14:00 tsbere: I'm not particularly comfortable pushing the lazy-circ patch in this moment ... IMO, 1 setting vs 3 settings should be discussed (if it was, I don't recall it), and I'd like to have more eyes on it that are familiar with the code in that area :( 2011-05-05T12:14:09 hrm, Fusion is saying there is no hard disk 2011-05-05T12:14:11 :( 2011-05-05T12:14:43 senator: the encodeURI/decodeURI stuff to global_util.js, does that leave most values unchanged? I'm wondering about incumbent values in a given staff client prior to encountering that code change 2011-05-05T12:15:42 eeevil: the lazy_circ is going to get some real testing next week. As in, its going on our demo server and we'll likely go live with it whether it makes it into 2.1 or not. 2011-05-05T12:16:17 So will the sip_statcat stuff, in part at least. 2011-05-05T12:16:33 dbs: It didn't work 2011-05-05T12:16:42 tsbere: as for the stat cats via sip, I know there was more discussion on that, but I didn't follow it, so I'll have to defer to (IIRC) berick and senator? 2011-05-05T12:16:54 Looks like it lost half of the contents of the OVF file 2011-05-05T12:17:01 yep. i've not bothered seriously publishing the sip fee payment code, yet. it needs a lot of testing. 2011-05-05T12:17:02 Petaris: hmm, alas. 2011-05-05T12:17:37 standards_that_don't_work-- 2011-05-05T12:17:39 speaking as tsbere's boss, we're OK with those things waiting until later. we're going to be testing them soon. 2011-05-05T12:17:43 Dyrcona: understood ... I still have concerns about the ou setting ... if we need to add more components then things could get complicated 2011-05-05T12:17:52 phasefx: well, i think so. i'm not at this moment aware of things that don't make the roundtrip through encodeURI and decodeURI correctly 2011-05-05T12:18:04 it does sound like the kind of thing that would have exceptions though 2011-05-05T12:18:24 senator: when things get encodeURI'ed, do they get stuffed into the database in that format? 2011-05-05T12:18:27 eeevil: I don't recall any 1 vs 3 settings thing on the lazy_circ bit. The only pain I noted was the fact that the settings interface doesn't actually know how to do anything of use ;) 2011-05-05T12:18:44 senator: if nothing obvious blew up in your testing, it's probably fine. May lose a sticky setting or two, but historically such things didn't survive through upgrades very well anyway 2011-05-05T12:18:46 And launchpad still has the "just use a standard string for config" version 2011-05-05T12:19:10 dbs: if i know the flow of the code in question correctly (and i may not) the db isn't involved, this is strictly in-staff client permutations 2011-05-05T12:19:14 or so i believe. verifying that 2011-05-05T12:19:31 tsbere: right ... forcing users to type in JSON correctly is suboptimal ... why not use 3? (there's a GSOC project to wrangle the UI, so I'm not particularly worried about adding more) 2011-05-05T12:19:32 senator: okay, cool 2011-05-05T12:19:50 phasefx: regular punct and unicode as well seem to do fine, no blowing up so far 2011-05-05T12:19:54 senator: dbs: these values get stored as mozilla preferences 2011-05-05T12:20:09 Does anyone know if Evergreen can be installed (without a large amount of additional work) on RHEL or CentOS? 2011-05-05T12:20:20 eeevil: Because the way it is coded *each* prefix can have a length and padding character, independantly. 2011-05-05T12:20:53 Petaris: it has been installed on RHEL 5, but it's still rather ugly 2011-05-05T12:20:59 Petaris: Yes it can be installed on RHEL, I'm not entirely sure about CentOS 2011-05-05T12:21:01 tsbere: I don't follow why that matters 2011-05-05T12:21:31 tsbere: surely you only have one prefix per branch? 2011-05-05T12:21:34 dbs, ok I just figured I would ask as I have a CentOS 5.5 box here I could install it on 2011-05-05T12:21:55 eeevil: No, we have branches with multiple. 2011-05-05T12:21:59 moodaepo: RHEL and CentOS are binary compatable :) 2011-05-05T12:22:01 Petaris: if you go that route, use the rhel target in Makefile.install in OpenSRF and Evergreen 2011-05-05T12:22:22 eeevil: Take two libraries with their own prefix codes, merge them into one........ 2011-05-05T12:22:35 * moodaepo has sadly lost track of distros 2011-05-05T12:22:40 tsbere: well... it just got more complicated then ;) ... doesn't seem like a candidate for YAOUS at all 2011-05-05T12:22:49 dbs: I think I'll just throw up a debian VM on one of my ESXi boxes as thats what the install instructions are for 2011-05-05T12:22:58 sounds like a new table 2011-05-05T12:23:18 Petaris: that's the sanest route 2011-05-05T12:23:28 sitting on atheos patch by the way pending his dco. left him a vmail too. personally willing to commit it aynway, wait, or not commit and move on with beta, in that order, but deferring to eeevil and dbs 2011-05-05T12:24:12 eeevil: An entire table for an optional org unit limited feature on a single interface? I felt that was overkill. 2011-05-05T12:24:51 dbs: 32 or 64 bit? Or does it not matter? 2011-05-05T12:25:11 Petaris: both are fine for evaluation purposes 2011-05-05T12:25:29 dbs: Ok, thanks 2011-05-05T12:25:52 tsbere: but an OUS is the wrong tool for the job of a complex (and possibly changable) data structure 2011-05-05T12:27:32 atheos may be out today 2011-05-05T12:27:39 havent heard from him today, and usually do by now 2011-05-05T12:27:40 my suggestion would be to proceed with what you have, because you guys need it, and generalize to a table-based solution later .. should be relatively straight-forward to provide an upgrade path from your OUS version to a table 2011-05-05T12:28:13 I know 100% that he will provide the DCO, so feel free to commmit and then apply the DCO retroactively 2011-05-05T12:28:36 he wouldnt want to hold up the beta release - ill see if i can get him on Facebook or something 2011-05-05T12:28:50 but I wouldn't encourage building a specialized ui for one OUS ... and we'd need something like that 2011-05-05T12:29:39 thanks for checking via facebook mrpeters-isl. licensy things make me uncomfortable 2011-05-05T12:29:56 yep - just talked to him on phone 2011-05-05T12:30:01 mrpeters-isl++ 2011-05-05T12:30:04 he is out buying a new hard drive for his apple 2011-05-05T12:30:22 so he can't submit it anyways at the moment -- he wants me to submit it on his behalf, and on behalf of Evergreen Indiana 2011-05-05T12:30:59 mrpeters-isl: that should be fine; you're asserting that you have the right to contribute the code, in that case 2011-05-05T12:31:40 tsbere: your patron_info_copy branch seems straightforward and utterly unobjectionable. making sure it actually does what it says it does (30 seconds).... you have a standing dco at this point i imagine? :-) 2011-05-05T12:31:57 senator: where shall i email 2011-05-05T12:31:58 senator: Yes, I have a standing DCO. On the dokuwiki, even. 2011-05-05T12:32:02 or i guess i just copy your "Signed-off-by" in the commit message 2011-05-05T12:32:03 thanks 2011-05-05T12:32:03 eeevil: fwiw, we did just that with a print related OUS. Knew it was wrong at the time, but expediency :) 2011-05-05T12:32:28 eeevil: I have until the end of the month to have something working that you are happy-ish with before we go live, even if you are only happy to put it as far as trunk. 2011-05-05T12:33:18 * tsbere had other features he was considering but ditched for reasons of not complicating the OUS syntax anyway 2011-05-05T12:33:28 phasefx: and now it's causing all sorts of pain :( 2011-05-05T12:33:38 is it? 2011-05-05T12:34:03 phasefx: the one-big-blob stuff? I'd say so 2011-05-05T12:34:33 you're thinking item attribute templates as user settings? yeah, I wish that was done differently 2011-05-05T12:34:34 phasefx: also harmed by similar implementation: copy templates (IIRC), no? 2011-05-05T12:34:46 DCO sent to list for atheos 2011-05-05T12:34:49 this print one is just an array of strings, IIRC 2011-05-05T12:35:18 tsbere: not that the standing DCO is more than a signal of intention (which is worth something at least), until we revise the "Contributing" document to adopt "Signed-off-by" 2011-05-05T12:36:32 phasefx: if it's completely (complete) managed by a specialized UI, it's not the end of the world, but data structure changes will eventually crop up, causing pain, for anything that's not an opaque list-o-strings that have no positional meaning ... that may be the case for the printing settings, is it? 2011-05-05T12:36:42 ("Contributing" document patches welcome!) 2011-05-05T12:37:11 with the copy templates, the 1.4 -> 1.6 circ modifier transition comes to mind as one previous example. 2011-05-05T12:37:12 I guess a rule would be: if you want to use an array, don't expect index to matter 2011-05-05T12:37:23 *** natschil has joined #evergreen 2011-05-05T12:38:03 *** kmlussier has joined #evergreen 2011-05-05T12:38:26 eeevil: it's described like so: Disable automatic print attempts from staff client interfaces for the receipt types in this list. Possible values: "Checkout", "Bill Pay", "Hold Slip", "Transit Slip", and "Hold/Transit Slip". Has a datatype of array 2011-05-05T12:38:43 code treats it as a set 2011-05-05T12:38:50 eeevil: If I am going to the effort of jamming it into a table anyway, what do you think of adding the option of possible suffix as well? Just for kicks. 2011-05-05T12:38:58 phasefx: so, indexOf() > -1 is all that matters 2011-05-05T12:39:03 yeah 2011-05-05T12:39:31 tsbere: I see no reason not to, sir ... some other ILS' tack letters onto the end of barcodes 2011-05-05T12:39:41 to differentiate patron from item 2011-05-05T12:40:23 tsbere: any reason not to add an item version of this? many more integration points, of course... 2011-05-05T12:41:24 eeevil: For one, where do I put the disambiguation buttons when more than one matches? On the patron by barcode interface that is easy. 2011-05-05T12:41:45 I suppose I could figure out making a popup, but that is a tad more difficult. 2011-05-05T12:42:27 * tsbere notes that he deals with "patron starts with 2, item starts with 3, then the library's 4 digit real prefix code" locally, but was trying to come up with all kinds of other uses for the feature when he thought of suffixes the first time 2011-05-05T12:44:24 click-to-copy++ 2011-05-05T12:44:27 tsbere++ 2011-05-05T12:44:46 atheos++ 2011-05-05T12:44:49 tsbere++ 2011-05-05T12:44:51 mrpeters-isl++ 2011-05-05T12:44:56 ok done with my last minute patching 2011-05-05T12:45:04 now we're all incremented and feeling good 2011-05-05T12:45:07 grabbing ab ite 2011-05-05T12:45:10 *bite 2011-05-05T12:45:29 trying out pg_restore with -j 12 and hoping for goodness 2011-05-05T12:48:51 *** Sally has quit IRC 2011-05-05T13:03:11 *** Sally has joined #evergreen 2011-05-05T13:04:50 *** natschil has quit IRC 2011-05-05T13:08:10 *** dbs has quit IRC 2011-05-05T13:22:59 *** dbs has joined #evergreen 2011-05-05T13:23:26 *** christoph_ has joined #evergreen 2011-05-05T13:32:11 *** dbs has quit IRC 2011-05-05T13:40:06 *** finnx has left #evergreen 2011-05-05T13:45:57 *** dbs has joined #evergreen 2011-05-05T13:56:27 ok ... we ready to cut stuff? 2011-05-05T13:59:50 * Dyrcona fires up the chainsaw..... *bzzzzzzz* 2011-05-05T14:03:46 * tsbere is working on learning how to use fancy_prompt 2011-05-05T14:03:56 Thus, I have no objections to cutting of stuff 2011-05-05T14:10:29 @later tell dbwells BillDueber over in #code4lib wants to hear your issues with Library::CallNumber::LC - you game? 2011-05-05T14:10:29 dbs: The operation succeeded. 2011-05-05T14:11:06 tsbere: what should I be updating in the staff client area at release time? 2011-05-05T14:11:23 or phasefx, whoever is handy 2011-05-05T14:13:22 windowssetup.nsi 2011-05-05T14:13:51 tsbere: hrm.. that's just 2.1+, right? 2011-05-05T14:14:05 Yea. 2011-05-05T14:14:23 whew... thanks, didn't see it in my 2.0.6 checkout :) 2011-05-05T14:14:32 *** natschil has joined #evergreen 2011-05-05T14:15:55 Hi, I try to restore a backup of a 1.6.0.0 Evergreen to our new library computer on which I installed the version 2.0.5. I am doing it along the dokuwiki but get lots of errors, e.g. type "tsquery" is only a shell. What should I do otherwise? 2011-05-05T14:16:13 christoph_: have you gone through the upgrade steps on the database itself? 2011-05-05T14:16:23 christoph_: What mrpeters-isl said : ) 2011-05-05T14:16:25 the 1.6 > 2.0 upgrade is a hefty one so a 1.6 database wont work with a new version 2011-05-05T14:16:56 mrpeters-isl: moodaepo: I am here with christoph_ trying to get this to work. 2011-05-05T14:17:07 See in Open-ILS\src\sql\Pg\upgrade 2011-05-05T14:17:09 moodaepo: mrpeters-isl: so we are trying to load an old backup simply into a new evergreen database 2011-05-05T14:17:20 err sorry - no upgrade subdir 2011-05-05T14:17:29 you need to get 1.6.0.0 up to 1.6.1 first 2011-05-05T14:17:40 then you can run the 1.6.1-2.0-upgrade-db.sql 2011-05-05T14:17:40 moodaepo: mrpeters-isl: and seeing that the backup is simply a sql file, shouldn't one be able to simply load that sql file into a database? 2011-05-05T14:17:51 depends on how you backed it up 2011-05-05T14:17:56 I think this is detailed somewhat (could use a little love?) in the upgrade dokuwiki pages. 2011-05-05T14:18:21 natschil: pg_dump -Fc -U evergreen -f /path/to/evergreen_db.backup.gz evergreen is your friend :) 2011-05-05T14:18:24 natschil: And there may be remnant oddities between your actual production DB and the upgrade scripts. 2011-05-05T14:18:41 That have to be resolved in order to successfully complete an upgrade to the next level. 2011-05-05T14:19:00 http://pastie.org/1868862 2011-05-05T14:19:02 mrpeters-isl: bshum: could you also include christoph_ in whatever you respond, as he is the one mainly doing the upgrade 2011-05-05T14:19:15 best way, in my experience, to backup and restore a snapshot 2011-05-05T14:19:18 natschil: christoph_: Sure ;) 2011-05-05T14:19:42 yep 2011-05-05T14:19:58 eeevil: On a different note, on adding partial barcode support to items for checkout and such. Pre-cats could be an issue if the pre-cat barcode matches, after prefix/suffix/whatever, a non-precat barcode. Any thoughts? 2011-05-05T14:19:59 christoph_: did you run pg_dump from the newer version of PostgreSQL? 2011-05-05T14:20:09 natschil: christoph_ I don't see why you've have issues just loading the 1.6 data as mrpeters-isl mentioned using pg_dump but importing a huge SQL file 'should' work 2011-05-05T14:20:15 mrpeters-isl: yes I used pg_dump 2011-05-05T14:20:15 mrpeters-isl: I totally go with j=48, I love super crunching DB hardware :) 2011-05-05T14:20:26 You might see messages if the PG versions are different 2011-05-05T14:21:04 christoph_: ^^ PostgreSQL versions are the same? 2011-05-05T14:22:56 mrpeters-isl: I am running lucid with the database of hardy. Is there a difference between the two? 2011-05-05T14:23:00 PostgreSQL docs are very clear that if source database server version < target database server version, pg_dump must be run from target database server version 2011-05-05T14:23:20 the OS distro shouldnt matter, it's more of the PostgreSQL versions 2011-05-05T14:23:21 christoph_: probably 8.3 on hardy, 8.4 on lucid 2011-05-05T14:23:36 i think dbs is right 2011-05-05T14:23:36 Yes I can confirm it's 8.3 on hardy by default 2011-05-05T14:23:41 but there might be a backport of 8.4 2011-05-05T14:23:50 http://packages.ubuntu.com/hardy/postgresql says 8.3 on hardy :) 2011-05-05T14:24:24 christoph_: dbs: you mean to run pg_dump from a machine that is running 8.4? 2011-05-05T14:24:25 and mrpeters-isl's pastie is exactly what I'm doing now (going from 8.3 -> 9.0) 2011-05-05T14:24:30 natschil: correct 2011-05-05T14:24:56 then once it's loaded up from the snapshot you get the fun of doing all of the little upgrades between 1.6.0.0 and 2.0.x 2011-05-05T14:25:00 building i18n for 2.0.6 2011-05-05T14:25:04 dbs: allright. Let me install 8.3. on some machine, load the backup into that, and then get it out from the 8.4 machine, and then we'll see. 2011-05-05T14:25:08 dbs: Would love to see your 8.3>9.0 notes to compare with mine 2011-05-05T14:25:19 you'll probably need a 1.6.1.x tarball 2011-05-05T14:25:24 for those incremental upgrades 2011-05-05T14:25:25 I have it posted on the IRC a few times 2011-05-05T14:26:01 a note on the upgrade scripts --- don't just psql < somescript.sql them 2011-05-05T14:26:14 go through them bit by bit because if you've done any backporting you may run into rollbacks 2011-05-05T14:26:18 natschil: well, hmm - rereading the postgresql docs, I might have incorrectly remembered the pg_dump bit 2011-05-05T14:26:32 "Because pg_dump is used to transfer data to newer versions of PostgreSQL, the output of pg_dump can be expected to load into PostgreSQL server versions newer than pg_dump's version" 2011-05-05T14:27:00 maybe you need to upgrade the source db first? 2011-05-05T14:27:09 moodaepo: my 8.3 -> 8.4 notes are on paste.lisp.org from somewhere around January, first time I ran through this process 2011-05-05T14:27:27 mrpeters-isl: naw 2011-05-05T14:28:35 moodaepo: what do you mean with "just loading the 1.6 data"? 2011-05-05T14:28:37 dbs: Will do..I personally used 8.3 pg_dump to import into 9.0 the main complains were that plperl etc were already installed in 9.0 2011-05-05T14:29:13 christoph_: Basically a pg_dump and pg_restore should work as mrpeters-isl and dbs mentioned 2011-05-05T14:29:38 or not ... dbs: 2011-05-05T14:29:50 http://paste.lisp.org/display/121795 2011-05-05T14:30:11 dbs: thoughts? 2011-05-05T14:30:54 eeevil: sounds like non-JSON in Searcher.js 2011-05-05T14:31:04 and yet ... it looks ok 2011-05-05T14:31:13 * dbs goes and looks 2011-05-05T14:31:19 eeevil: quoted keys? 2011-05-05T14:31:25 no, not qutoed 2011-05-05T14:31:30 need to be? 2011-05-05T14:31:53 aha 2011-05-05T14:32:01 eh? 2011-05-05T14:32:02 yes, JSON requires that keys be quoted 2011-05-05T14:32:11 and in trunk Searcher.js is fine 2011-05-05T14:32:15 k 2011-05-05T14:32:21 looks like that didn't get backported to rel_2_1 2011-05-05T14:32:49 fixing in the export ... anyone feel like getting that back into the repo through 2.0.6 tag? 2011-05-05T14:33:06 dbs: dbs: christoph: moodaepo: There seems to be a lot of confusion now... so *should* I simple be able to take the pg_dump of the old database and load it into a new (8.3) database? 2011-05-05T14:33:18 natschil: Yes. 2011-05-05T14:33:20 natschil: 2.0 requires 8.4 2011-05-05T14:33:47 * moodaepo will shut up 2011-05-05T14:33:58 You *should* be able to take the pg_dump of the old database and load it into a new 8.4 or 9.0 database 2011-05-05T14:34:13 christoph_: dbs: yep, I have 8.4. In fact, it's on a completely different machine to the old server, as the old server was also running a samba PDC and dns server, and had quite weak hardware. The only thing I have here apart from the new server is the backup (dump) of the old server. 2011-05-05T14:34:33 natschil: so, maybe you could paste the errors that you were worried about in the first place 2011-05-05T14:34:48 moodaepo: I didnt use pg_restore but like in the wiki psql -U evergreen -h localhost -f evergreen_db.backup.2008-06-15 new_evergreen 2011-05-05T14:34:49 eeevil: I'll get it going (backport, sigh) 2011-05-05T14:34:55 dbs++ 2011-05-05T14:35:27 christoph_: Could you paste the dokuwiki link? 2011-05-05T14:35:28 christoph_: That should work actually, so long as the DB was created on the new server place correctly. 2011-05-05T14:35:36 dbs: christoph_ is doing the actual install so he's on the server machine, I'll ask him to send the errors across. 2011-05-05T14:35:58 We used to do that style before we learned how to do a pg_restore 2011-05-05T14:36:27 eeevil: or... hrm, did I not commit that at all? 2011-05-05T14:36:45 * dbs smacks forehead, hard 2011-05-05T14:37:09 moodaepo: http://www.open-ils.org/dokuwiki/doku.php?id=evergreen-admin:maintenance:backups 2011-05-05T14:37:15 dbs: Hmm can't seem to find your 8.3>9.0 upgrade logs, if you have any from this time a pastie would be great > http://paste.lisp.org/list/evergreen?3 2011-05-05T14:37:26 moodaepo: I think christoph_ is referring to the backups page, which doesn't include a line to create the db beforehand with the necessary contribs 2011-05-05T14:37:47 christoph_: Oh yea that's ancient 2011-05-05T14:38:06 Might be why they're getting tsquery errors on reload, since they wouldn't have all the tsearch.sql contrib installed first. 2011-05-05T14:38:15 ^^ 2011-05-05T14:38:22 bshum: That doesn't require one to create an db though 2011-05-05T14:38:28 *a db 2011-05-05T14:38:28 doesn't specify locale, etc. either 2011-05-05T14:38:36 moodaepo: could we depricate the page? 2011-05-05T14:38:46 dbs: that didn't seem to do it 2011-05-05T14:38:58 hrm.. well 2011-05-05T14:38:59 eeevil: committing now 2011-05-05T14:39:39 moodaepo: is there a newer page with instructions? 2011-05-05T14:40:18 christoph_: mrpeters-isl's steps in the pastie before should work much nicer than what's on that page. 2011-05-05T14:40:51 dbs: thanks much, sir 2011-05-05T14:41:01 moodaepo's pastie: http://pastie.org/1868862 2011-05-05T14:41:13 christoph_: http://pastie.org/1868862 at lines 7-13 are the pieces that you need to setup a DB first. 2011-05-05T14:41:43 christoph_: natschil: No they are valid I was just noting that it was ancient. Let me read through the page, one sec. 2011-05-05T14:41:47 christoph_: At that point, you should be able to use either pg_restore as indicated, or psql -f like the old page. Either way ought to work if past experiences are any indication. Though ideally pg_restore is better. 2011-05-05T14:42:59 btw, on an unrelated note, have there been any major changes in opensrf since January or so this year? 2011-05-05T14:43:07 moodaepo: Heh, that page was last edited 2008/06/28 by dbs, *whistle* 2011-05-05T14:43:26 idea being that you want to get the newer definitions for tsearch, tablefunc, etc in place before you pg_restore, so that the older equivalents will error out, but the rest of the schema and data will load happily 2011-05-05T14:43:38 bshum: CLICK THE EDIT BUTTON 2011-05-05T14:43:55 A wiki is a better place for working documents that paste bins 2011-05-05T14:43:57 bshum: Yea that's why I mentioned ancient : ) 2011-05-05T14:44:18 I vaguely recall these old admin pages. 2011-05-05T14:44:23 Back in 1.4 days 2011-05-05T14:44:35 dbs: I just did : ) 2011-05-05T14:44:53 moodaepo++ 2011-05-05T14:45:20 natschil: some unicode-handling bug fixes that were important 2011-05-05T14:46:09 dbs: nice. API-wise it's all still backwards compatible though, right? 2011-05-05T14:46:10 eeevil: okay, backported through rel_2_0_6 2011-05-05T14:46:17 natschil: yep 2011-05-05T14:47:26 Whoa, that's a long changelog for 2.0.6 :) 2011-05-05T14:47:57 yeah, 2.0.6 has been building up a lot of steam 2011-05-05T14:49:13 * tsbere forgot some of his stuff was backported to 2.0 2011-05-05T14:50:18 * dbs notes that "von Hans-Christian Müơller" is already corrupted, probably by Unicorn in the first place, but at least Evergreen doesn't throw it away now. heh. 2011-05-05T14:52:54 bshum: Should I even suggest -j6 for pg_restore on the page? 2011-05-05T14:53:06 moodaepo: Eh..... depends 2011-05-05T14:53:14 moodaepo: That feature was only added in 8.4+ 2011-05-05T14:53:31 Right, leaving it out for now. 2011-05-05T14:53:32 moodaepo: Which is why I switched to using pg_restore instead of the flat psql dump file loading 2011-05-05T14:53:43 moodaepo: I would suggest adding a comment that "-j" is really useful on 8.4 and up 2011-05-05T14:53:57 I don't think I've ever used psql for restore : ) 2011-05-05T14:53:59 I would add it as an additional line like, If using 8.4+, recommend assigning -j=X for parallel loads 2011-05-05T14:54:05 dbs: Will do 2011-05-05T14:54:13 * dbs feels insecure with his puny -j 12 on his 8-core test db server 2011-05-05T14:54:28 not like bshum's manly -j 48 2011-05-05T14:55:38 dbs: :D 2011-05-05T14:58:15 eeevil: i18n going okay? 2011-05-05T14:59:45 dbs: curious about i18n. So we can pull a PO from LP and stick it into where the locales are stored; then run the steps written in the svn-install page (which reads for 1.4) but that ought to get us more up-to-date translations, right? 2011-05-05T15:00:30 Or at least, more up-to-date with LP's translations. 2011-05-05T15:01:09 (actually it's 1.6) Referring to this page: http://open-ils.org/dokuwiki/doku.php?id=server:svn_install 2011-05-05T15:01:33 bshum: cd build/i18n; # drop PO file in place; make install_all_locales 2011-05-05T15:02:22 yeah, that page is largely still accurate. although dojo should be in place first now 2011-05-05T15:02:40 now I finished to follow the lines 7-13 in moodaepo's pastie and with the last line of the old wiki and the old data is now in the new database. But I lost the overview, how should I do now the database upgrades from 1.6.0.0 to 2.0.5 2011-05-05T15:03:03 make LOCALE=es-ES install # much faster than make install-all-locales, for obvious reasons 2011-05-05T15:03:31 Well it was a repast by bshum of a paste by mrpeters-isl I had nothing to do with it except copy it over to the dokuwiki page you were looking at earlier : ) 2011-05-05T15:03:45 christoph_: > http://evergreen-ils.org/dokuwiki/doku.php?id=upgrading:evergreen 2011-05-05T15:03:52 moodaepo++ 2011-05-05T15:04:07 christoph_: run every point release database upgrade from 1.6.0.0 through 1.6.1.8, then 1.6->2.0 script, then 2.0.1 through 2.0.5 2011-05-05T15:04:10 dbs: Neat, I'll have to give that a try later. Thanks! 2011-05-05T15:04:32 christoph_: What 1.6 version are you at currently? 2011-05-05T15:05:03 dbs: it is 2011-05-05T15:05:03 (i was keying off of "how should I do now the database upgrades from 1.6.0.0 to 2.0.5") 2011-05-05T15:05:18 dbs: By the time christoph_ get's to 2.0.5 there might be a 2.0.6 aye : ) 2011-05-05T15:05:19 huzzah for uncommitted fixes lying around in SVN checkouts 2011-05-05T15:05:37 moodaepo: depends on how many more hurdles we can throw in eeevil's way 2011-05-05T15:05:42 hah 2011-05-05T15:06:48 moodaepo: I am at 1.6.0.0 but I am a bit slow it might even be 2.1.something 2011-05-05T15:07:11 christoph_++ 2011-05-05T15:07:52 *** jenny1 has joined #evergreen 2011-05-05T15:09:17 *** jenny has quit IRC 2011-05-05T15:09:23 * senator looks for sawhorses in the back to literally throw into eeevil's office 2011-05-05T15:10:12 senator: I think eeevil would be impressed if you merely placed them gently in his office 2011-05-05T15:10:54 please don't bonk the release cutter on the head ;) 2011-05-05T15:10:55 bah, these walls could use some character :-) 2011-05-05T15:16:39 Hmm so we removed OPAC_LOGIN permissions from a group and then added it back in but users still cannot login to OPAC any suggestions? > http://paste.lisp.org/display/121796 2011-05-05T15:16:54 rad ... I can't scp to lupin 2011-05-05T15:18:37 heh 2011-05-05T15:18:41 sftp? 2011-05-05T15:18:46 rsync? 2011-05-05T15:19:20 eeevil: I pmed you 2011-05-05T15:19:21 ok ... I can't connect to port 22 2011-05-05T15:19:36 Yea the port is diff I passed it over 2011-05-05T15:19:56 moodaepo: thanks, sir 2011-05-05T15:27:50 moodaepo: ok ... well, I'm not in sudoers, so I can't move the new tarball into place ... thoughts? 2011-05-05T15:28:09 wtf 2011-05-05T15:28:54 eeevil: Let's try that again. 2011-05-05T15:29:56 WHEEE! 2011-05-05T15:30:04 moodaepo: thanks, man 2011-05-05T15:30:33 moodaepo-- # overlooking DA MAN 2011-05-05T15:30:45 *** bgoble has quit IRC 2011-05-05T15:31:00 ok, it's in place 2011-05-05T15:31:15 eeevil++ 2011-05-05T15:31:19 the 2.1 beta may have to wait until tonight or tomorrow 2011-05-05T15:31:27 but at least we can get 2.0.6 out :) 2011-05-05T15:32:33 *** jenny has joined #evergreen 2011-05-05T15:32:39 eeevil: Since you got 2.0.6 uploaded, any thoughts on auto-complete on item barcodes in relation to "what if you wanted to pre-cat the exact match and the prefix matches as well"? 2011-05-05T15:32:41 *** jenny has left #evergreen 2011-05-05T15:34:16 eeevil: Do we need to run reingest if moving from 2.0.5 to 2.0.6? 2011-05-05T15:35:26 *** jenny1 has quit IRC 2011-05-05T15:35:42 moodaepo: it's not required, but it wouldn't hurt. also, you'll want to say UPDATE authority.record_entry SET active = active; 2011-05-05T15:36:13 moodaepo: if you were ever on 1.6 2011-05-05T15:39:49 *** _bott_ has quit IRC 2011-05-05T15:47:23 hmm, tons of "INFO: task postgres:20016 blocked for more than 120 seconds" messages in dmesg during pg_restore 2011-05-05T15:48:11 hopefully won't interfere with the restore 2011-05-05T15:49:07 I'm trying to test my OpenSRF install but the test command of request opensrf.math add 2 2 is coming back with "Received no data from server" 2011-05-05T15:49:23 Petaris: did you run osrf_ctl -l -a start_all ? 2011-05-05T15:49:35 I left out the -l 2011-05-05T15:49:43 Petaris: what did it say? 2011-05-05T15:49:44 The first upgrade script went through without any problems now from 1.6.0.1 to 1.6.0.2 I get the following error: psql:Open-ILS/src/sql/Pg/1.6.0.1-1.6.0.2-upgrade-db.sql:19: ERROR: duplicate key value violates unique constraint "upgrade_log_pkey" 2011-05-05T15:49:52 as it said you should only do it if you need to force it 2011-05-05T15:50:20 Petaris: when you ran it, what did it output to the screen? 2011-05-05T15:50:36 Petaris:( that being osrf_ctl without the -l) 2011-05-05T15:50:43 yes it did 2011-05-05T15:50:57 I just stopped the service and am starting it with the -l 2011-05-05T15:51:02 see if that fixes it 2011-05-05T15:51:37 ok, that fixed it 2011-05-05T15:51:58 sorry to bother 2011-05-05T15:53:53 Petaris: np 2011-05-05T15:55:26 *** jenny1 has joined #evergreen 2011-05-05T16:03:35 moodaepo: I saw that you pasted more than a year ago the same update error ( ERROR: duplicate key value violates unique constraint "upgrade_log_pkey"). Did you find an easy way to solve that? That'd be great. 2011-05-05T16:04:43 *** jenny has joined #evergreen 2011-05-05T16:06:24 christoph_: looks like http://svn.open-ils.org/trac/ILS/browser/branches/rel_1_6_0/Open-ILS/src/sql/Pg/1.6.0.1-1.6.0.2-upgrade-db.sql says "1.6.0.1" in the first INSERT stmt, which would be the problem 2011-05-05T16:06:28 *** jenny1 has quit IRC 2011-05-05T16:07:47 christoph_: you might want to compare the upgrade script that is in the 1.6.0.2 tarball for that purpose with the one from SVN 2011-05-05T16:08:49 christoph_: nm, same problem in the tarball 2011-05-05T16:09:20 adjust to say "1.6.0.2" and rerun the script 2011-05-05T16:10:14 interesting, 1.6.0.3 tarball has a good version of that script 2011-05-05T16:11:32 same script is busted in the 1.6.1.8 tarball though 2011-05-05T16:11:45 * dbs goes home 2011-05-05T16:11:53 *** dbs has quit IRC 2011-05-05T16:19:45 christoph_: What dbs said ^^ 2011-05-05T16:21:19 eeevil: Reminder: When you cut beta2 make sure you drop it into downloads/previews/ 2011-05-05T16:21:29 bshum: Same applies to you sir for client 2011-05-05T16:22:04 moodaepo: Yep, starting 2.0.6 SC building, bumping xulrunner version to 1.9.2.17 (latest one) 2011-05-05T16:23:00 But you mean downloads/ right? 2011-05-05T16:23:03 Not previews 2011-05-05T16:23:11 Oh, you mean when beta1 is finished 2011-05-05T16:23:15 Got it 2011-05-05T16:24:14 *** gdunbar has quit IRC 2011-05-05T16:27:34 moodaepo: csharp: I don't think lupin is handling email (feedback@evergreen-ils.org and docs@evergreen-ils.org) 2011-05-05T16:28:47 phasefx: Oy I didn't know that needed to be setup. 2011-05-05T16:29:41 in the case of feedback, there is a .procmailrc file in the feedback user's home directory, that copies incoming mail to a backup file and forwards to openils-feedback@esilibrary.com 2011-05-05T16:30:29 look in /etc/aliases for some of others, like docs, events, newsletter, eg-sites 2011-05-05T16:31:04 webmaster 2011-05-05T16:31:14 phasefx: Thanks 2011-05-05T16:36:40 *** natschil has quit IRC 2011-05-05T16:37:08 *** natschil has joined #evergreen 2011-05-05T16:38:00 *** natschil has joined #evergreen 2011-05-05T16:39:00 *** natschil has joined #evergreen 2011-05-05T16:40:21 *** natschil has joined #evergreen 2011-05-05T16:40:42 *** natschil has joined #evergreen 2011-05-05T16:41:28 bshum: Thoughts? > http://paste.lisp.org/display/121796 2011-05-05T16:41:42 *** natschil has quit IRC 2011-05-05T16:42:00 *** natschil has joined #evergreen 2011-05-05T16:42:43 *** Petaris has quit IRC 2011-05-05T16:43:10 *** natschil has quit IRC 2011-05-05T16:43:34 moodaepo: What level do they have the permission defined at? And is it on their primary permission group (or ancestor thereof)? 2011-05-05T16:43:49 *** natschil has joined #evergreen 2011-05-05T16:45:23 *** natschil has quit IRC 2011-05-05T16:45:34 tsbere: It's on the level "Users" and it's the ancestor of group "Adult" which is the users group I was trying to login as..it was working fine till we removed the permission to make sure they can't login and added it back. The users still cannot login : ) 2011-05-05T16:45:54 *** natschil has joined #evergreen 2011-05-05T16:46:06 moodaepo: When I said level I should have said "depth", I think. 2011-05-05T16:46:08 By level, I think he means 0, 1, 2, or 3 (or more) 2011-05-05T16:46:17 *** natschil has joined #evergreen 2011-05-05T16:47:24 *** natschil has quit IRC 2011-05-05T16:48:13 *** natschil has joined #evergreen 2011-05-05T16:48:49 *** natschil has joined #evergreen 2011-05-05T16:49:40 *** _bott_ has joined #evergreen 2011-05-05T16:49:50 *** natschil has quit IRC 2011-05-05T16:50:14 *** natschil has joined #evergreen 2011-05-05T16:51:10 *** natschil has quit IRC 2011-05-05T16:51:28 *** natschil has joined #evergreen 2011-05-05T16:51:57 *** natschil has joined #evergreen 2011-05-05T16:52:21 *** natschil has joined #evergreen 2011-05-05T16:52:43 *** natschil has joined #evergreen 2011-05-05T16:54:33 *** natschil has joined #evergreen 2011-05-05T16:55:28 *** natschil has quit IRC 2011-05-05T16:56:06 *** natschil has joined #evergreen 2011-05-05T16:56:31 tsbere: bshum: The permission is set on level 1 and is consortium wide depth 2011-05-05T16:56:48 Pretty sure level 1 is system depth by default 2011-05-05T16:56:51 0 = consortium 2011-05-05T16:56:55 Yea 2011-05-05T16:59:04 *** natschil has quit IRC 2011-05-05T16:59:15 *** Sally has quit IRC 2011-05-05T16:59:24 *** natschil has joined #evergreen 2011-05-05T17:01:57 2.0.6 SC uploaded 2011-05-05T17:02:05 Going to start tweaking wiki pages now 2011-05-05T17:02:55 Oh, nvm, moodaepo++ :) 2011-05-05T17:04:33 *** kmlussier has quit IRC 2011-05-05T17:07:47 *** AaronZ-PLS has quit IRC 2011-05-05T17:09:10 phasefx: Who should feedback be going to...you and afterl? 2011-05-05T17:10:09 moodaepo: openils-feedback@esilibrary.com 2011-05-05T17:10:24 that has me, dbs, aterlaga, and I think bshum 2011-05-05T17:10:32 I'm not one of the lucky 2011-05-05T17:10:44 But I think I'm okay with that for now ;) 2011-05-05T17:14:22 moodaepo: When you say level 1, do you mean like 1 - Users ? 2011-05-05T17:14:23 *** jlamos has quit IRC 2011-05-05T17:14:59 bshum: Yes "Users" 2011-05-05T17:15:10 Oh, then I'm stumped :) 2011-05-05T17:15:19 Cause that's exactly what we have. 2011-05-05T17:15:35 It's not that the patron is expired or inactive? 2011-05-05T17:15:49 Heh it's running 2.04 I think it just wants some attention and an upgrade to shiny 2.0.6 2011-05-05T17:16:00 Oh well, that's also reasonable ;) 2011-05-05T17:16:08 moodaepo: Did you clear memcached and such? 2011-05-05T17:16:11 That's what I'm going to be doing all tomorrow... 2011-05-05T17:16:28 Nope that would have given a diff error message not PERM_FAILURE 2011-05-05T17:16:51 I was thinking "perhaps the *lack* of perm was cached" ;) 2011-05-05T17:17:03 tsbere: Hmm nope didn't seem like it was required, will give it a whirl and see 2011-05-05T17:22:34 tsbere: No luck. Shutdown and started everything in the proper sequence. 2011-05-05T17:26:22 moodaepo: Did you double-check with a SQL query that the right permission is granted? 2011-05-05T17:26:33 bshum: This looks correct doesn't it?It's freakin EG 101...heck 100 > http://i.imgur.com/gfYFt.png 2011-05-05T17:28:41 That does seem correct. 2011-05-05T17:28:52 There's no specific entry associated with the user right? 2011-05-05T17:29:41 Or wait, you've got all users unable to login? 2011-05-05T17:31:39 *** Dyrcona has quit IRC 2011-05-05T17:44:44 *** jenny has left #evergreen 2011-05-05T18:00:24 *** yboston has quit IRC 2011-05-05T18:17:57 *** christoph_ has quit IRC 2011-05-05T18:30:43 @karma help 2011-05-05T18:30:43 moodaepo: An error has occurred and has been logged. Please contact this bot's administrator for more information. 2011-05-05T18:55:15 @karma help 2011-05-05T18:55:15 moodaepo: An error has occurred and has been logged. Please contact this bot's administrator for more information. 2011-05-05T18:55:26 @karma bshum 2011-05-05T18:55:26 moodaepo: An error has occurred and has been logged. Please contact this bot's administrator for more information. 2011-05-05T18:55:38 *** AaronZ-PLS has joined #evergreen 2011-05-05T18:57:30 *** pmplett has quit IRC 2011-05-05T18:59:01 *** AaronZ-PLS has quit IRC 2011-05-05T19:06:53 *** rsinger has quit IRC 2011-05-05T19:07:41 *** rsinger has joined #evergreen 2011-05-05T19:10:07 @karma bshum 2011-05-05T19:10:07 moodaepo: Karma for "bshum" has been increased 59 times and decreased 0 times for a total karma of 59. 2011-05-05T19:11:15 Just for the record I had to dump the sqlite db from sqlite3 to sqlite for pinesol to read the file - sqlite3 data/#/Karma.sqlite.db .dump | sqlite ~/Karma.sqlite.db 2011-05-05T19:11:49 We have multiple versions running and need to figure things out sometime. 2011-05-05T19:13:13 Well on the new server there was just sqlite so I had to install sqlite3 to do the conversion, which is why there are multiple versions. 2011-05-05T19:13:22 @karma help 2011-05-05T19:13:22 moodaepo: help has neutral karma. 2011-05-05T19:16:44 *** youdonotexist has quit IRC 2011-05-05T19:52:01 moodaepo++ 2011-05-05T20:27:44 looks like beta cuttin' time 2011-05-05T21:03:44 *** mrpeters-isl has quit IRC 2011-05-05T21:20:30 *** _bott_ has quit IRC 2011-05-05T21:22:52 *** _bott_ has joined #evergreen 2011-05-05T21:23:19 *** dbs has joined #evergreen 2011-05-05T21:23:19 *** dbs has joined #evergreen 2011-05-05T21:39:23 woo hoo! Evergreen-ILS-2.1-beta1.tar.gz and Evergreen-ILS-2.1-beta1.tar.gz.md5 uploaded to download/previews/ 2011-05-05T21:45:46 eeevil++ 2011-05-05T21:45:56 I'm poking at the downloads page and wiki pages now 2011-05-05T21:46:22 Have to rewrite the steps for PG9 using backports 2011-05-05T22:07:53 eeevil++ 2.1++ 2011-05-05T22:08:03 @karma 2011-05-05T22:08:03 berick: Highest karma: "dbs" (385), "miker_" (140), and "berick" (125). Lowest karma: "----------------------------------" (-16), "my_laptop" (-15), and "ie" (-8). You (berick) are ranked 3 out of 1026. 2011-05-05T22:08:54 bshum / moodaepo: have you guys got the "upload files to special directory" instructions added to the "rolling a release" wiki pages? 2011-05-05T22:10:28 dbs: I doubt it. That was something that came up during poking at site stuff for alpha/beta/rc files to keep things from getting too confusing for the 2.0 release materials. We can add that though. 2011-05-05T22:10:37 bshum: please do! 2011-05-05T22:12:43 dbs: Thanks for mentioning that actually. Means all the links in the 2.1 install steps are broken cause it's missing the /preview/ part of the folder structure 2011-05-05T22:13:26 ugh 2011-05-05T22:14:01 I think the initial idea there was to have staff clients placed in previews, but moodaepo asked for all alpha/beta/rc to be put there I think. 2011-05-05T22:15:07 Bleh 2011-05-05T22:15:15 The same goes for opensrf-2.0.0-rc2 then 2011-05-05T22:17:40 Perhaps we should revisit the use of the previews folder if it's going to affect this many pages / links. Not to mention the READMEs and official documentation. 2011-05-05T22:18:05 I think the main purpose of using the previews was to stop the new staff-clients archive page from pulling in the various 2.0 alphas/beta, etc. 2011-05-05T22:18:15 We could put things back to not use the previews folder 2011-05-05T22:18:31 And just figure out how to fix the code on that staff client archive instead 2011-05-05T22:20:38 *** pmplett has joined #evergreen 2011-05-05T22:58:59 Nifty, the make staff client action for the 2.1 beta seems to fail because it's looking for the 1.9.2.16 xulrunner, but the newest one is now 1.9.2.17 2011-05-05T23:05:52 bshum: you tweaked the makefile and sallied forth? 2011-05-05T23:06:16 phasefx: Yes, though I did forget to install nsis first 2011-05-05T23:06:27 :D 2011-05-05T23:06:32 Two warnings: 2011-05-05T23:06:34 2 warnings: 2011-05-05T23:06:34 !include: could not find: "extras.nsi" (windowssetup.nsi:118) 2011-05-05T23:06:34 !include: could not find: "extras.nsi" (windowssetup.nsi:269) 2011-05-05T23:06:40 Not sure what that means 2011-05-05T23:06:46 I think those are harmless 2011-05-05T23:06:55 hooks for add-ons to the installer 2011-05-05T23:08:11 Oh okay 2011-05-05T23:08:44 I was following the instructions you wrote for 2.1 unofficial 2011-05-05T23:08:54 But not entirely sure what that needs to look like for an actual one 2011-05-05T23:09:12 Presumably, the make STAFF_CLIENT_STAMP_ID=rel_2_1_beta1 would be important 2011-05-05T23:09:16 What does devbuild do? 2011-05-05T23:09:43 copies some developer preferences into place and enables some debug buttons at the bottom of the login window 2011-05-05T23:09:56 may affect performance, but I never really measured it 2011-05-05T23:10:08 (disables or tries to disable some caching, etc) 2011-05-05T23:10:49 I'm not sure about the official process either.. there's something with xulrunner-stub.release.exe 2011-05-05T23:11:29 http://evergreen-ils.org/dokuwiki/doku.php?id=mozilla-devel:building_the_staff_client#windows_client 2011-05-05T23:11:49 Oh, interesting, there's an extras.example.nsi file in the directory 2011-05-05T23:11:57 Guess that's an example I should have looked at, heh 2011-05-05T23:13:16 may want to give tsbere a ping (oh, see what I did?) 2011-05-05T23:13:40 Heh 2011-05-05T23:14:35 so we might suppose to rename xulrunner-stub.release.exe to xulrunner-stub.exe 2011-05-05T23:14:40 then make 2011-05-05T23:34:41 I think I like this new builder. 2011-05-05T23:34:49 Just have to learn all the new tricks, but it seems fun 2011-05-05T23:58:57 Quirk, when trying to install the 2.1 beta1 client, all the references in the installer screens are doubled for the name. Like it'll read "Evergreen Staff Client 2.1-beta1 2.1-beta1" everywhere, instead of just "Evergreen Staff Client 2.1-beta1" alone. The extra "2.1-beta1" is a bit disconcerting.