Evergreen ILS Website

IRC log for #evergreen, 2018-08-08

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat

All times shown according to the server's local time.

Time Nick Message
00:06 foxcookie3 joined #evergreen
04:45 orb joined #evergreen
05:27 channels joined #evergreen
06:31 pinesol News from qatests: Testing Success <http://testing.evergreen-ils.org/~live>
07:02 agoben joined #evergreen
07:03 bsanford joined #evergreen
07:04 JBoyer joined #evergreen
07:06 rjackson_isl joined #evergreen
07:11 Metacity22 joined #evergreen
07:22 FrozenFox29 joined #evergreen
07:29 dwgreen joined #evergreen
07:36 bdljohn joined #evergreen
08:01 dwgreen joined #evergreen
08:15 collum joined #evergreen
08:16 kmlussier joined #evergreen
08:19 * kmlussier suspects the email on the general list may be related to the patch for bug 1775719. jeffdavis mentioned that he had put it in production.
08:20 pinesol Launchpad bug 1775719 in Evergreen 3.0 "Multiple IndexedDB connections (via tabs) can result in data inconsistency" [High,Confirmed] https://launchpad.net/bugs/1775719
08:20 rlefaive joined #evergreen
08:21 kmlussier By the "email on the general list", I mean the email that reports problems registering workstations on iPads.
08:23 dwgreen joined #evergreen
08:23 collum joined #evergreen
08:25 JBoyer kmlussier, I wouldn't think that should be an issue on initial registration though? I don't have one handy at work anymore but if someone can get one hooked up to a Mac and pull the console logs that should make the problem easy to spot.
08:25 kmlussier JBoyer: No, actually, I think I'm confusing it with something else.
08:25 JBoyer I can try to remember to use mine at home to see what's going on.
08:26 kmlussier There was something I was testing recently where workstation registration was not working correctly. I thought it was that patch, but I misremembered.
08:26 JBoyer Oh, was that the local -> server settings branch, maybe?
08:26 JBoyer (Obviously not their problem, but I thought I'd heard something about it too)
08:27 kmlussier JBoyer: Yes, that's the one. So it's totally unrelated.
08:27 JBoyer ++
08:28 kmlussier It just seems strange that it was working fine when they updated and then stopped working.
08:30 JBoyer Yeah, the timing makes it sound difficult to reproduce since it wasn't day one. :/
08:30 JBoyer And I don't know how easy it is to wipe out the indexdb in mobile Safari, that could play a part.
08:35 Dyrcona joined #evergreen
08:38 Dyrcona passwords--
08:42 mmorgan joined #evergreen
08:45 Ishaq29 joined #evergreen
08:49 bos20k joined #evergreen
08:54 bos20k left #evergreen
09:04 collum joined #evergreen
09:04 terran_ joined #evergreen
09:10 kmlussier We now have that handy Stored User Preferences interface where we can view/delete preferences stored in the browser, in hatch and in the database. Would it also be possible to have something that clears out the indexeddb database for those times that it needs to be cleared for troubleshooting purposes?
09:21 Sigals joined #evergreen
09:32 idjit joined #evergreen
09:40 jvwoolf joined #evergreen
09:40 JBoyer kmlussier, it would have to be automated most likely, since the idb errors kill the page load. That in turn could (potentially) cause transaction loss. I'm not sure how easy it would be to pull them out anyway at that stage, but it may be possible.
09:42 kmlussier Oh, of course! I had forgotten that you couldn't actually do anything in the client when those problems occur.
09:44 yboston joined #evergreen
09:46 JBoyer Ok, thinking about it further I know it's possible to get the transactions out (by hand) because I
09:46 JBoyer ve poked around in the Application tab of the dev tools, but I don't know how many people would ever want to bother with that.
09:48 kmlussier JBoyer: I agree. That's what I was hoping to avoid.
09:54 rjackson_isl joined #evergreen
09:58 khuckins joined #evergreen
10:04 rlefaive joined #evergreen
10:04 JBoyer Hopefully once bug 1775719 goes in it won't really be an issue anymore. Though a "refresh / cleanup" db option could be added if it looks like there's some value in that.
10:04 pinesol Launchpad bug 1775719 in Evergreen 3.0 "Multiple IndexedDB connections (via tabs) can result in data inconsistency" [High,Confirmed] https://launchpad.net/bugs/1775719
10:07 terran_ joined #evergreen
10:08 rlefaive joined #evergreen
10:11 * csharp attempted to install OpenSRF on Ubuntu 18.04 last night
10:11 bshum csharp: So much fun right?
10:11 csharp install went fine, but OpenSRF dies not being able to authenticate to ejabberd
10:11 bshum The ejabberd stuff is broken
10:11 bshum Cause of the auth changes
10:11 csharp bshum: do you remember the actual issue?
10:11 bshum I'm looking for the LP
10:12 csharp the config file still seems to think plain password auth is ok
10:12 csharp I looked for it last night but search-fu failed :-/
10:13 Dyrcona I think we're going to have to teach OpenSRF about SASL.
10:13 bshum https://bugs.launchpad.net/evergreen/+bug/1703411
10:13 pinesol Launchpad bug 1703411 in OpenSRF "OpenSRF: XMPP Non-SASL auth is being phased out" [Medium,Confirmed]
10:13 Dyrcona bshum++
10:13 bshum Yes what Dyrcona says, basically the non-SASL auth is not a choice
10:13 bshum And setting the "legacy" auth options in ejabberd config did nothing for me too
10:13 csharp bshum++
10:13 csharp thanks - I'll do some digging on that today
10:14 csharp also planning to test Angular 6 stuff (finally)
10:14 csharp JBoyer: I haven't forgotten about Firefox Hatch - just not getting to it :-/
10:15 csharp I made the mistake of waiting too long and now I'll need to relearn things I knew in April/May
10:15 rlefaive joined #evergreen
10:16 JBoyer csharp++
10:16 * Dyrcona suspects we'll have it working on Ubuntu 18.04 in time for Ubunut 20.04. :)
10:16 JBoyer technically speaking you don't *have* to build it yourself. Hatch proper hasn't changed and the installer can't automatically put it in the browser. So you can install the current version, then point FF Quantum at a the proper dir in a local checkout and it should run
10:16 Dyrcona Heh..."ubunut"
10:17 csharp @who is an Ubunut?
10:17 pinesol dluch_ is an Ubunut.
10:17 JBoyer That said, we all have many plates, don't worry that you haven't found time to juggle them all at once. :)
10:17 csharp JBoyer: thanks
10:18 JBoyer (good way to break plates!)
10:18 csharp I'll attempt that later
10:18 * csharp blows dust off Windows 10 VM that hasn't been run in weeks
10:19 * Dyrcona is using a Windows VM that hadn't been used in months.
10:19 csharp honestly our office move BS kind of blew the summer away - finally in a more stable situation now
10:19 * Dyrcona forgot his laptop power cord at home, so...
10:19 Dyrcona Well, moves will do that.
10:19 csharp Dyrcona: rage, RAGE, against the dying of your laptop battery!
10:20 Dyrcona It gets good life, but don't think it will last 8 hours.
10:20 Dyrcona I figure I'll just slum it today.
10:20 Dyrcona :)
10:20 JBoyer I imagine you're both about to be @Oprah'd with updates too. ("You get an update, and you get an update! Everybody updates!")
10:21 Dyrcona My VM has been running and getting updates.
10:21 Dyrcona We have these HP thin client things.
10:21 JBoyer Ah, I mistook unused for powered off.
10:22 csharp mine has been powered off
10:22 Dyrcona Sure. It's not always obvious.
10:22 Dyrcona English is like that. :)
10:22 * Dyrcona ponders adding a comment to a bug about terminology..... :)
10:33 pinesol [evergreen|Bill Erickson] LP#1768947 Offline DB runs in shared web worker - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=071a549>
10:33 pinesol [evergreen|Bill Erickson] LP#1768947 Offline xact presence is cached; show date - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=3256fd4>
10:33 pinesol [evergreen|Bill Erickson] LP#1768947 Disable offline download button, improve logging - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=457d7f6>
10:34 berick kmlussier++
10:35 csharp kmlussier++
10:35 csharp berick++
10:44 berick @breakfast
10:44 pinesol berick: Dyrcona probably has a script for that.
10:44 berick excellent
10:45 kmlussier berick: I just added a comment to the bug. There were conflicts when trying to backport it to 3.0, and I'm not comfortable with resolving them. Would you be able to take a look?
10:45 berick kmlussier: yes, can do
10:48 csharp berick: building angular6 branch - is it better for testing to build in dev or prod mode? (or does it matter?)
10:48 berick csharp: dev mode is good for testing, especially if you want to experiment with code changes
10:48 berick (in --watch mode)
10:48 csharp ok - will do
10:49 idjit i've been poking around in some of the angular6 stuff as well. i'm still trying to learn my way around, but is there anything useful i can be doing?
10:50 berick idjit: the main goal for now is making sure the app in general can be used alongside the angularjs app.  and that simple things like authentication, navigation, etc. generally work as expected.
10:50 berick csharp: idjit: as a reminder, log out or clear cookies if you previously used the angularjs app on your test client.  the cookie path changes for auth tokens.
10:51 * berick should probably add a auto-migration step for that
10:52 idjit noted, thank you.
10:52 idjit berick++
11:02 JBoyer kmlussier++
11:02 JBoyer berick++
11:02 JBoyer (just catching up on my lp email)
11:07 csharp berick: with 'ng build --watch', it appears to have finished, but is sitting there now without returning to the prompt - is that expected?
11:07 Dyrcona csharp: I believe that is expected.
11:07 csharp ok
11:07 berick yeah, if you open and save a file, you should see it do a quick recompile
11:08 berick a .ts file
11:08 csharp oh, I understand
11:08 Dyrcona You can always stop it and then restart it in a screen or tmux session.
11:08 berick yeah, exactly, i like to run a separate terminal window for that
11:09 csharp ok - makes sense
11:10 berick csharp: also, if you do make changes and don't want to keep reinstalling / copying files, I do this on my dev machine:  ln -s /path/to/Evergreen/Open-ILS/web/eg2 /openils/var/web/eg2
11:10 berick that plus --watch means all changes are immediately available to the browser
11:10 berick speeds up dev quite a bit
11:11 csharp makes sense
11:16 bshum systemd--
11:27 abneiman this may be obvious to some of you, but, if I discovered if you're voiced but not registered to services freenode won't let you PM other users (or maybe just certain users).  JFYI.
11:28 bshum Just certain users
11:29 abneiman thanks bshum, just wanted to drop it in channel in case anyone else was temporarily mystified :)
11:30 bshum I think users who recently authenticated as registered users with freenode services get set with +R mode, which only allows other registered users to message them
11:30 csharp correct
11:31 bshum You can turn that off by keying something like "/mode -R" (but of course that means you might see spam sent to you direct)
11:31 bshum spammers--
11:31 rhamby_ spammers--
11:37 csharp @blame the spammers
11:37 pinesol csharp: the spammers was monkeying around too much on the prod servers!
11:37 vamiry13 joined #evergreen
11:44 Metacity5 joined #evergreen
11:45 csharp berick: something busted on my box:
11:45 csharp AH00125: Request exceeded the limit of 10 subrequest nesting levels due to probable configuration error. Use 'LimitInternalRecursion' to increase the limit if necessary. Use 'LogLevel debug' to get a backtrace.
11:45 andshelbya joined #evergreen
11:45 csharp apparently related to "FallbackResource /eg2/index.html"
11:46 csharp that file doesn't exist on the filesystem, so I gues that's no surprise
11:46 berick yeah, that'll do it
11:46 berick the ng build step creates the file
11:46 csharp oh - hmm
11:46 berick sounds like it still need copying into place (or the symlink I mentioned)
11:47 kmlussier spammers--
11:47 csharp berick: ok - I'll keep at it :_)
11:47 berick make install also works, fwiw
11:52 pinesol [evergreen|a. bellenir] LP#1785333: Right-click can clear grid selection - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=536e5e8>
11:53 berick csharp: also, ICYMI, I pushed i18n changes recently.  see bottom of Open-ILS/examples/apache_24/eg_vhost.conf.in
11:54 berick csharp: arg, that's probably the issue you're having, sorry forgot about that
11:54 csharp berick: yeah, I see "en-US"
11:54 berick locale is in the URL now (and optionally angjs cookie)
11:54 csharp ah
11:55 berick and actually, I need to un-comment the section just above the first <Directory>
11:55 berick so that having no locale still redirects correctly, even when only 1 locale is supported
11:56 * berick does that and updates docs
11:57 rlefaive joined #evergreen
12:00 berick ok, docs updated, asking user to look at the sample config instead of maintaining a separate copy
12:01 idjit miker++ # you didn't like a commit message that used the phrase "more gooder"? :-)
12:01 miker :) ... one of the committer duties
12:05 kmlussier miker++ # Grammar matters! :)
12:06 idjit bah. capitalization--
12:06 terran_ lol
12:10 jihpringle joined #evergreen
12:11 Dyrcona Well, I was considering signing off on the "more gooder" branch and squashing the two commits, since the second only changes a comment.
12:12 Dyrcona But, not using my usual machine today.
12:13 Dyrcona Basically would have done what miker did. :)
12:13 Dyrcona miker++
12:18 rlefaive joined #evergreen
12:38 Dyrcona Bmagic: Re Evergreen and iPads: You do know that all browsers on iOS must use the Safari engine?
12:39 * Dyrcona thinks we should just flat out say we don't support iOS devices, or we do, and then make an effort to actually support them.
12:40 Dyrcona browsers--
12:41 jonadab Supporting iOS devices well is highly problematic for open-source projects that use a share-alike license.  Because you can't submit anything share-alike licensed to their app store without violating their terms of service for developers.
12:41 bshum "Community supported" remains a very broad, undefined thing to me
12:42 Christineb joined #evergreen
12:44 jonadab Permissive licenses, like MIT, BSD, X11, Apache, etc., are ok, though.
12:44 Dyrcona Yeah, this is why I stopped using Apple products a couple of years ago.
12:45 Dyrcona Weird how they got more control freaky after Jobs died....
12:54 submain joined #evergreen
12:55 circ-user-dSZ7a joined #evergreen
13:02 dpearl joined #evergreen
13:04 Carlos061113 joined #evergreen
13:05 and0 joined #evergreen
13:08 csharp berick: ok, finally in, but basically everywhere I go is pointing back to angularJS pages - is there something I should test?
13:10 csharp nm - I see your response to idjit earlier...
13:12 nfburton joined #evergreen
13:12 csharp so far so goo
13:12 csharp d
13:12 berick csharp: awesome.  the ang6 stuff is the experimental catalog and admin -> server admin
13:13 berick and change operator
13:13 csharp berick: so how do I get to those? (might be missing something obvious)
13:13 csharp login/ws registration was clearly ang6
13:14 berick csharp: Cataloging -> Staff Catalog (experimental)
13:14 csharp oh - I see it now
13:14 berick adin admin -> server admin
13:15 berick s/adin/and/
13:16 csharp nice!
13:16 berick csharp++ # glad it's working
13:16 idjit are console errors expected? most pages i visit seem to function, but there's a "TypeError: Cannot read property 'code' of undefined" from core.js:1598
13:18 berick idjit: generally, no, not expected.
13:18 berick idjit: that's in the ang6 app?
13:19 idjit yeah, happens when the page loads. doesn't seem to matter which screen.
13:19 idjit doesn't seem to break anything, either. i'd just been ignoring it.
13:19 berick huh, that's not an error I recall seeing
13:20 pastebot "idjit" at 64.57.241.14 pasted "ERROR TypeError: Cannot read p" (11 lines) at http://paste.evergreen-ils.org/13885
13:22 berick thanks idjit
13:22 idjit if it only happens on my test install, i'm willing to bet i goofed it up, rather than anything actually being broken.
13:22 berick related to newest code... idjit what's your URL path?
13:22 idjit https://egdev1.grpl.org/eg2/staff/splash
13:23 berick idjit: try https://egdev1.grpl.org/eg2/en-US/staff/splash
13:23 berick same issue that csharp was having.  just pushed a doc update this morning
13:23 berick and an apache config update
13:23 idjit i'll have to pull your latest
13:24 * berick nods
13:28 csharp not seeing errors here fwiw
13:29 berick woohoo
13:29 * berick looks for food
13:35 kmlussier @dessert berick
13:35 * pinesol grabs some chocolate-marshmallow-ice-cream root beer float for berick
13:35 kmlussier Ewwww.
13:36 Dyrcona Doesn't sound that bad....bit on the sweet side.
13:36 Dyrcona Unfortunately, I don't have a script for lunch. :)
13:37 kmlussier Apparently, I'm the one who added that dessert. My memory isn't what it used to be.
13:38 jvwoolf Hey, if I'm finding some bugs in the ebook API, should I open a new bug ticket? Or add comments to LP # 1673870?
13:39 Dyrcona jvwoolf: Open a new bug.
13:39 Dyrcona Or, bugs if there is more than 1.
13:40 jvwoolf Dyrcona: Will do. Thanks!
13:40 jvwoolf Dyrcona++
13:40 idjit does eg_vhost.conf still need <Directory "/openils/var/web/eg2">FallbackResource /eg2/index.html</Directory>, or was this supplanted by <Directory "/openils/vra/web/eg2/en-US">... ?
13:40 * Dyrcona doesn't deserve karma for that.
13:41 csharp idjit: supplanted, yes
13:42 idjit huh. whale, i bork'd it. i should probably go back and actually read the docs.
13:43 Dyrcona @dunno
13:43 pinesol Dyrcona: pretty sure there's already a bug report about that.
13:47 idjit /openils/var/web/eg2 should contain en-US and other enabled locales? should it contain main.js, index.html, and so on, or should all of these just live in eg2/en-US/? i'm getting "Error: Cannot match any routes. URL Segment: 'en-US/staff/splash'" and a rather grumpy stacktrace.
13:48 csharp idjit: in my case /openils/var/web/eg2/ just contains en-US and en-US contains the other files
13:48 Dyrcona @band add Grumpy Stacktrace
13:48 pinesol Dyrcona: Band 'Grumpy Stacktrace' added to list
13:48 JBoyer That sounds kind of like the ang 1.X app getting a route with a locale in it that it wasn't expecting.
13:48 csharp idjit: also, make sure to clear your file cache
13:50 Dyrcona When all else fails, destroy the VM and start over.
13:50 Bmagic Dyrcona: no, I was not aware - I thought Google Chrome was on the app store. Even so, maybe Apple followed suit?
13:50 idjit http://evergreen-ils.org/~​erickson/EG_Ang6_App.html says to run `ng build --deploy-url /eg2 --base-href /eg2 ...`. will that take care of the localization for me, or should that get updated to include the en-US part of the path?
13:51 Dyrcona Bmagic: Yes, it's "Google Chrome" in name with some of Chrome's features, but the render/HTML/JS engine is Safari.
13:51 Dyrcona I don't know that means for SSL handling.
13:51 csharp idjit: I was following http://git.evergreen-ils.org/?p=working/Evergree​n.git;a=blob_plain;f=Open-ILS/src/eg2/README.ado​c;h=3dc9b43890ff28ffaa73634142061202b06d2a1c;hb=​refs/heads/collab/berick/lp1775466-ang6-base-app but I didn't do locale building
13:52 idjit ok, cool. i'll tinker a bit more, and if doesn't work i'll nuke the VM and start over :-)
13:54 collum_ joined #evergreen
13:55 mmorgan1 joined #evergreen
13:57 * kmlussier would in favor of supporting iOS devices because so many of our libraries have them.
13:59 Xlbrag_ joined #evergreen
14:00 berick idjit: only en-US is supported in dev mode.
14:00 berick there's fr-CA support on --prod mode, but no translations.
14:00 Bmagic Dyrcona: The timing of the failure made me wonder if it was related to the SSL time bomb, just a thought
14:02 Immune joined #evergreen
14:04 collum joined #evergreen
14:04 * Dyrcona shrugs.
14:05 collum joined #evergreen
14:06 csharp berick: I'm really keen on this making it into 3.2... what sort of testing rigor do you think is needed for that?  So far, everything I'm doing works without trouble - not testing super deeply, but UIs are loading fine, etc.
14:08 berick csharp: great, i've been giving it a good workout too, so I'm also pretty confident merge-safe.  I probably need to roll back a few changes to be absolutely safe, though.
14:08 csharp legacy angular UIs work fine too
14:08 csharp ok
14:08 berick e.g. the server admin page needs a ton of testing if we're going to use it now.  I turned it on for testing/review, but we probably don't want to replace that UI wholesale
14:08 csharp right, I think having it available with a fallback is ideal
14:09 berick i could roll that switch back, then re-deploy to master after 3.2
14:09 berick I suspect we'll want to hide the experimental catalog for 3.2 as well
14:09 berick since it's not fully functional
14:09 csharp sure
14:09 * berick can make those changes in minutes
14:09 Dyrcona If you're hiding everything, then what's the point of putting it in 3.2?
14:10 berick Dyrcona: fair point.  we could expose a few admin UI's...
14:10 * berick also ported the ACQ admin -- forgot about that
14:10 csharp anything that allows us to use dojo less is a win in my book
14:11 csharp oh - cool - I'll look at those too
14:11 * Dyrcona agrees with getting rid of dojo.
14:11 csharp and make my testbox avialable to fellow PINES staff for pokage
14:12 Dyrcona I'm for putting it in 3.2 if it does something useful.
14:12 berick there are quite a few dojo interfaces replaced.  question is how many can we sign off on before 3.2
14:12 Dyrcona If it's just for the sake of having the code there, then I'm not so enthusiastic.
14:13 berick many of them are dead simple, of course
14:13 berick just a grid, create, edit, and delete actions
14:13 berick well, most of them
14:13 Dyrcona Should we put this on the meeting agenda?
14:14 berick Dyrcona: you read my mind
14:14 csharp ugh - I'm gonna miss most of the dev meeting - my daughter's school pickup collides with that timeslot
14:14 berick heh, found a bug I just created in the admin UI's.  /me fixes
14:15 Dyrcona Speaking of finding bugs...Tests I wrote for a new feature actually found a bug in my code.
14:15 Dyrcona I thought about signing into IRC on Sunday just to say, "Yay, tests!"
14:16 kmlussier tests++
14:28 * JBoyer won't be here for the meeting either.
14:29 JBoyer Though I'm +1 for including ang6 in 3.2. The more dojo admin interfaces removed, the better.
14:30 collum joined #evergreen
14:32 nfburton joined #evergreen
14:33 nfburton Quick Q - Trying to find the setting for default OPAC OU. It defaults to Consortium instead of System right now. I changed it on my test server but can't seem to find the setting again
14:35 Dyrcona nfburton: Do you mean for a user or for a domain?
14:36 nfburton When not signed in, the default for the opac.
14:38 csharp nfburton: do you mean having the OPAC default to a particular library?
14:38 nfburton Yeah
14:39 Bmagic we do it with an Apache config "RedirectMatch ^/$ https://yourdomain.org/eg/opac/home?locg=127"
14:39 kmlussier nfburton: It's not an org unit setting. There is a way you can configure your domain to include an locg parameter to default to a certain search scope.
14:39 csharp nfburton: if so, that's done via setting the OU name and IP address range in /openils/conf/lib_ips.txt, then making sure apache knows about it
14:39 csharp or that
14:39 kmlussier Yes, what Bmagic said.
14:39 Dyrcona Well, depends on how many domains you have.
14:40 Dyrcona If just one, Bmagic's trick works.
14:40 Bmagic also - "SetEnv physical_loc 127" is good for other things like self registration
14:40 nfburton yeah, but for one page
14:40 kmlussier The IP file is useful for in-library access that defaults to a search library.
14:40 csharp /etc/apache2/eg_startup has the redirection lines - uncomment those if you want to use IP redirection
14:40 Dyrcona If you have multiple domains, you're better off using something more complex.
14:41 nfburton I'm sure I found this in Apache or something. The locg parameter that's fed to get_library to create the ddl
14:41 csharp in other words, "there's more than one way to do it" ;-)
14:41 csharp a lot of our libraries just hard-code the logc parameter in the browser and call it a day
14:41 nfburton hmm
14:41 Dyrcona yeah, locg should get passed around once set. If you find a page where it gets dropped, that's likely a bug.
14:42 nfburton SetEnv physical_loc 127 - Thats what I am looking for!
14:43 nfburton Which file is that in?
14:43 berick csharp: mind adding a simple comment to bug 1775466 re: your testing?
14:43 pinesol Launchpad bug 1775466 in Evergreen "Angular6 Base Application" [Undecided,New] https://launchpad.net/bugs/1775466 - Assigned to Bill Erickson (berick)
14:43 Bmagic if you just have one domain, then probably eg.conf
14:44 * Dyrcona switches to the laptop for the remainder of the day. BRB.
14:46 nfburton Bmagic++ Thank you! That was going to bug me all day
14:46 Bmagic nfburton: yw
14:46 Dyrcona joined #evergreen
14:49 khuckins joined #evergreen
14:57 gmcharlt meeting warning in 3 minutes
14:57 gmcharlt or well, you get the warning now
15:00 gmcharlt #startmeeting Evergreen Development Meeting 2018-08-08
15:00 pinesol Meeting started Wed Aug  8 15:00:45 2018 US/Eastern.  The chair is gmcharlt. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00 pinesol Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00 Topic for #evergreen is now  (Meeting topic: Evergreen Development Meeting 2018-08-08)
15:00 pinesol The meeting name has been set to 'evergreen_development_meeting_2018_08_08'
15:00 gmcharlt #info Agenda is https://wiki.evergreen-ils.org/do​ku.php?id=dev:meetings:2018-08-08
15:01 gmcharlt #topic Introductions
15:01 Topic for #evergreen is now Introductions (Meeting topic: Evergreen Development Meeting 2018-08-08)
15:01 Dyrcona #info Dyrcona is Jason Stephenson CW MARS
15:01 gmcharlt #info gmcharlt = Galen Charlton, Equinox
15:01 miker #info miker == Mike Rylander, EOLI
15:01 dpearl #info dpearl is Dan Pearl (C/W MARS Inc.)
15:01 berick #info berick = Bill Erickson, KCLS
15:02 miker ha! pinesol, hush
15:02 abneiman #info abneiman is Andrea Neiman, EOLI
15:02 remingtron #info remingtron is Remington Steed, Hekman Library (Calvin College)
15:02 dbwells #info dbwells is Dan Wells, Hekman Library (Calvin College)
15:03 gmcharlt #topic Action items from previous meeting
15:03 Topic for #evergreen is now Action items from previous meeting (Meeting topic: Evergreen Development Meeting 2018-08-08)
15:04 gmcharlt one was csharp loading an update for the Hatch FF add-on
15:04 gmcharlt csharp: did that happen?
15:05 Dyrcona csharp may not be here. He said earlier he has a schedule conflict.
15:05 berick i think csharp said he'd be missing the meeting
15:05 gmcharlt ah, OK
15:05 gmcharlt so, next was in regards to OpenSRF 3.0.2
15:05 gmcharlt and since testing for bug 1774703 in production has not been promising, and it looks like we'll likely be changing websocket servers, no bugfix release
15:05 pinesol Launchpad bug 1774703 in OpenSRF "Websockets processes locked at 100% CPU" [Undecided,Confirmed] https://launchpad.net/bugs/1774703
15:06 gmcharlt but I have issued a call for roadmap entries for OpenSRF 3.1 today
15:06 berick gmcharlt++
15:06 gmcharlt #info Roadmap entries for OpenSRF 3.1 can be added here: https://wiki.evergreen-ils.org/​doku.php?id=dev:opensrf_roadmap
15:06 gmcharlt with, at present, websocketd being the biggest one targetted
15:07 * Dyrcona ponders adding SASL support.
15:08 gmcharlt there's also https://bugs.launchpad.net/opensrf/+bug/1702978, which is just a bugfix, but the full fix adds a small change to the ABI
15:08 pinesol Launchpad bug 1702978 in OpenSRF "memcache keys containing % fail" [High,Confirmed]
15:08 dbs #info dbs = Dan Scott, Laurentian
15:08 dbs win 31
15:08 tktech16 joined #evergreen
15:08 gmcharlt dbs: ancient history man, ancient history
15:08 kmlussier #info kmlussier is Kathy Lussier, MassLNC
15:09 gmcharlt #action gmcharlt will open and work on bugs for documentation changes for better ejabberd configuration during installation of OpenSRF
15:09 gmcharlt but for another update
15:09 gmcharlt #info berick and gmcharlt have scheduled a presentation to discuss new Angular at the hack-a-way
15:10 gmcharlt it will be on Monday 11/5
15:10 gmcharlt so y'all will be expected to be fresh! and full! of vim! and vigor!
15:10 gmcharlt (or be full of emacs, if you must)
15:10 Dyrcona heh
15:11 * bshum has a soft spot for nano
15:11 gmcharlt #info kmlussier and JBoyer finished their review of https://bugs.launchpad.net/evergreen/+bug/1750894
15:11 pinesol Launchpad bug 1750894 in Evergreen "Wishlist: Store web staff workstation settings on the server" [Wishlist,Fix committed]
15:11 gmcharlt and that does it for action items
15:11 gmcharlt and for that matter, the OpenSRF release update
15:11 gmcharlt so, without further ado
15:11 gmcharlt #topic Evergreen release
15:11 Topic for #evergreen is now Evergreen release (Meeting topic: Evergreen Development Meeting 2018-08-08)
15:11 gmcharlt berick?
15:12 berick jump into 3.2 stuff?
15:12 berick alright..
15:12 gmcharlt sure
15:12 berick buildmasters wanted.
15:12 berick first builds will be early next month
15:12 berick #info https://docs.google.com/spreadsheets/d/1gZayHfF​7qK0zwLMEAXt-PbKBMiAM_F6EZguqzIYceBY/edit#gid=0
15:13 berick i realized I have some travel plans this month
15:13 berick so I propose we push the feature slush deadline back from the 17th to the 22nd
15:13 gmcharlt +1
15:14 berick not critical, but I'd prefer to be around for the deadline
15:14 berick (though I'll be away for the days leading up to it)
15:14 Dyrcona +q
15:14 Dyrcona +1
15:14 Dyrcona :)
15:14 berick that may mean pushing back the feature freeze, but I was going to see how things go
15:14 kmlussier I'm always in favor of later feature freeze deadlines. :)
15:15 berick :) one in the plus column there
15:16 berick and finally, csharp had some good luck testing the ang6 code today
15:16 gmcharlt cool
15:16 berick there's some momentum toward including it in 3.2
15:16 berick if we do, we need to decide soon what in the 3.2 code should be available to staff
15:16 berick so we can test those bits specifically
15:16 berick we're limited to some admin UI's at this point
15:17 Dyrcona Is it all of the admin uis or just some of them?
15:17 berick also, good everyone understands that means adding a new build step when installing from source.
15:17 miker I'd prefer just dojo-replacing code ATM (and infrastructure and static-ish pages)
15:17 berick Dyrcona: just some of them
15:17 berick miker: agreed
15:17 * Dyrcona agrees with miker.
15:18 kmlussier +1
15:18 berick it's about 2/3 of the server->admin pages and acq admin pages
15:18 berick i can propose a list of pages to thumbs up/down if that helps
15:18 Dyrcona I can probably arrange some testing of those soon, even though I'll be out most of next week.
15:18 berick then we can test them
15:19 berick ok, so I'll get the ang6 branch into final merge-able state and..
15:20 berick put together a list of admin UI's that should be suitable for using in 3.2 as replacements for the dojo versions.
15:20 berick that will also mean modifying the links in the angjs app to point to the same new UI's
15:21 berick otherwise, re: 3.2, we've closed a few more blockers in the past ~week
15:21 Dyrcona berick++
15:22 berick and of course, those are not bound by the feature slush, but sooner the better, obviously
15:22 berick any questions for me?
15:23 Dyrcona Well, I have a question for "room" generally.
15:23 Dyrcona What do we think the odds of being able to remove XUL in 3.2 are at this point?
15:23 * kmlussier was about to ask a similar question.
15:23 berick good question
15:24 * gmcharlt can commit to a patch to disable it sufficiently for release purposes
15:24 Dyrcona We've actually had at least one library "give up" and go back to XUL on 3.0.
15:24 gmcharlt but that's kinda the much easier question
15:24 andshelbya left #evergreen
15:24 miker the room asks back: what's the bar for removing XUL? all blockers gone? something more (or less)?
15:24 kmlussier We still have a handful of blockers - https://bit.ly/2MvbmsT
15:25 Dyrcona Well, I guess that's part of my question, too, because I don't have a definitive answer on what the bar should be.
15:25 kmlussier My thought was that blockers are blockers that prevent people from moving to the web client.
15:25 * miker suspects that bug 1768947 might be gone now, with the new offline code in
15:25 pinesol Launchpad bug 1768947 in Evergreen "web client: non-hatch white screen issues" [Medium,Confirmed] https://launchpad.net/bugs/1768947
15:26 berick that's the hope
15:26 berick and some of those have pullrequests...
15:26 dbwells My preference would be that we remove XUL no matter what from 3.2, and that we commit to supporting 3.1 until all blockers are gone.  I can do my part for the second piece of that.
15:26 kmlussier miker: The code just went in today. I thought it would be good to keep it open until people start using EG with the fix.
15:26 berick and some of those i18n bugs might need re-verifying after the IDL fix
15:26 miker kmlussier: totes agree
15:27 kmlussier My preference is to wait on xul removal until the blockers are gone. Otherwise, we could be supporting 3.1 for a very long time. It's extra motivation.
15:27 berick i'd be a little surprised if we haven't broken some stuff in XUL by now
15:28 gmcharlt I share dbwells' view
15:28 gmcharlt I'd rather an extended 3.1 support period (which I can also help with) than an indefinite postponement of the cutoff
15:29 berick kmlussier: from another perspective, removing XUL is motivation to fix the blockers
15:29 dbwells I think having XUL removed from 3.2 is a really strong motivation to work on blockers.
15:29 Dyrcona It also does no good to keep XUL if much of it is broken.
15:30 miker the biggest "breakage" AFAIK is copy alert msg vs alert matrix
15:30 miker but that's a known quantity
15:30 kmlussier In any case, I would be willing to take an action item to test any webstaffblockers with a pullrequest.
15:31 kmlussier There are a couple of bugs there without pullrequest tags that I know are big concerns for our libraries.
15:32 berick if anyone can describe a way to reproduce bug 1724029, i'll happily fix that one.
15:32 pinesol Launchpad bug 1724029 in Evergreen "Web Client: Patron Search sorted by last name not working as expected" [High,Confirmed] https://launchpad.net/bugs/1724029
15:32 berick or give me SSH access to their prod servers!
15:32 dbwells berick: right on
15:34 berick we have a little over a month before RC is cut.
15:34 berick if it looks like we're really close on the blockers, that can also be pushed back.
15:34 berick i'd rather not yet, just to avoid procastination
15:35 kmlussier Here is my concern. Way back in the jspac removal days, we had a bunch of bugs with blocker tags on them. But we decided to remove jspac without addressing the blockers, and a lot of them sat in LP without any love.
15:35 kmlussier I really don't want to delay removing the XUL client, but I would hate to see something like that happen again.
15:37 gmcharlt an increasingly broken XUL client also imposes costs on users
15:37 berick and new blockers will be opened during the 3.2 cycle :|
15:38 kmlussier berick: Of course! :)
15:40 gmcharlt so... I think it's fair to say that we have an open question - but one that maybe we can punt until closer to release?
15:40 rlefaive joined #evergreen
15:41 berick my only request is the XUL removal code be merged by beta cutting day (1 month from today) so it can get some real testing
15:41 berick so ideally decide then-ish
15:41 gmcharlt sounds like a plan, or a guideline, anyway
15:42 gmcharlt any other questions regarding 3.2?
15:42 gmcharlt or any questions or updates regarding maintenance releases?
15:45 kmlussier Nope
15:45 gmcharlt ok
15:45 gmcharlt #topic Hatch updates
15:45 Topic for #evergreen is now Hatch updates (Meeting topic: Evergreen Development Meeting 2018-08-08)
15:45 * gmcharlt summons the spirit of csharp-past
15:45 gmcharlt #info No new releases since last meeting
15:45 gmcharlt onward
15:46 gmcharlt #topic Feedback for stuff under development
15:46 Topic for #evergreen is now Feedback for stuff under development (Meeting topic: Evergreen Development Meeting 2018-08-08)
15:46 gmcharlt #info Dyrcona would like eyes on the patch for bug 1780660
15:46 pinesol Launchpad bug 1780660 in Evergreen "Add More Workstation Functions to OpenILS::Utils::TestUtils" [Wishlist,Confirmed] https://launchpad.net/bugs/1780660
15:46 gmcharlt Dyrcona: any comments to add?
15:46 Dyrcona Just a bit of me being selfish and cleaning up some test code.
15:47 Dyrcona When writing Perl tests recently, I noticed we had at least 3 different implementations of code to add workstations for testing.
15:47 Dyrcona So, thought I'd come up with a single version of the necessary functions and add them to TestUtils.
15:47 gmcharlt Dyrcona++
15:48 miker +1
15:48 Dyrcona I'm also using the changes in the tests in my working branch, so I'd like the changes to go in before I rebase.
15:48 Dyrcona Should be simple to test. Just check out the branch cd to perlmods and run make livecheck
15:49 Dyrcona I rebased the branch after someone updated a test this week, too.
15:49 kmlussier Dyrcona++
15:50 * miker is willling to simply push it having read the commit
15:50 Dyrcona So, that's my shameless plug. :)
15:50 miker the CI tester will tell us if it is unhappy
15:50 Dyrcona :)
15:50 gmcharlt DRY: applies to cubicle farms and workstations?
15:50 * gmcharlt ducks
15:51 miker DONE
15:51 berick miker: heh, oh yeah, let's put that machine to work
15:52 gmcharlt ok
15:52 gmcharlt so any other pull requests that folks would like to draw attention to?
15:52 * miker dusts of his committing finger
15:53 pinesol [evergreen|Jason Stephenson] LP 1780660: Add more workstation functions to OpenILS::Utils::TestUtils. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=913237b>
15:53 miker I'd like to see the %-in-usernames patch go in ASAP
15:54 miker there's one for each of EG and OpenSRF
15:54 kmlussier miker: Which bug number is that?
15:54 * miker looks... terrible LP searching... grumble
15:55 Dyrcona Is that bug 1702978 ?
15:55 pinesol Launchpad bug 1702978 in OpenSRF "memcache keys containing % fail" [High,Confirmed] https://launchpad.net/bugs/1702978
15:55 miker Dyrcona: yeppers
15:55 miker there's a non-abi-breaking version posted currently
15:56 Dyrcona Are there any advantages to the abi-breaking version?
15:57 miker it would keep someone from accidentally implementing a similar bug
15:57 gmcharlt but that part can be reserved for OpenSRF 3.1 IMO
15:57 miker but requires tighter coordination between EG and OSRF
15:58 abneiman regarding things in need of testing, several cataloging bugs have PRs -- bug 1732761 bug 1675882 bug 1739460 bug 1739286 -- a couple of which are flagged as blockers
15:58 Dyrcona OK. I'll see about having a look at it.
15:58 pinesol Launchpad bug 1732761 in Evergreen "Web Client - When adding vols/copies to multiple branches, circ library does not populate" [High,Confirmed] https://launchpad.net/bugs/1732761
15:58 pinesol Launchpad bug 1675882 in Evergreen "webclient: adding individual copies defaults circ OU to workstation OU" [Medium,Confirmed] https://launchpad.net/bugs/1675882
15:58 pinesol Launchpad bug 1739460 in Evergreen "Web client: branch library shelving locations not visible to HQ workstation" [High,Confirmed] https://launchpad.net/bugs/1739460
15:58 pinesol Launchpad bug 1739286 in Evergreen "web client: Cannot set default search box in Z39.50" [Low,Confirmed] https://launchpad.net/bugs/1739286
15:58 miker abneiman++
15:59 berick what's the status of 1732761 since it references the omnibus?
15:59 berick which was already merged
16:00 berick ditto bug 1675882
16:00 pinesol Launchpad bug 1675882 in Evergreen "webclient: adding individual copies defaults circ OU to workstation OU" [Medium,Confirmed] https://launchpad.net/bugs/1675882
16:00 berick are they branches built atop the omnibus, but not merged w/ the omnibus?
16:00 miker 1732761 is in master
16:01 * miker will mark it dup and close
16:01 berick awesome
16:01 kmlussier miker++
16:01 Dyrcona Speaking of the omnibus, should that be backported to 3.0?
16:01 Dyrcona We've tested it here with 3.0.10 and it works. There's a conflict, but I know how to resolve it.
16:02 miker berick: checking the other ... Dyrcona: I'm for it
16:02 abneiman +1 to backporting omnibus
16:02 berick miker: bug 1739460 too please, unclear what to merge
16:02 pinesol Launchpad bug 1739460 in Evergreen "Web client: branch library shelving locations not visible to HQ workstation" [High,Confirmed] https://launchpad.net/bugs/1739460
16:02 miker berick: 1675882 is already marked dup of obnibus, actually
16:03 berick oh, so it is.
16:04 kmlussier For bug 1739460, I thought people were no longer able to replicate the original problem. Is it still needed?
16:04 pinesol Launchpad bug 1739460 in Evergreen "Web client: branch library shelving locations not visible to HQ workstation" [High,Confirmed] https://launchpad.net/bugs/1739460
16:04 miker berick: well, 1739460 is working as intended -- filtering locations based on selected copies. I don't know if that's what people want, though...
16:05 miker the actual code is belt-and-suspenders for filtering by numeric values, really
16:05 miker so... maybe it doesn't hurt to merge?
16:06 berick that would be user/miker/lp-1739460-all-relevant-locations ?
16:06 miker berick: yessir
16:06 berick OK, i can merge that, I've already tested it once
16:06 * berick grabs
16:07 gmcharlt and.. that takes us past the hour
16:07 gmcharlt thanks, folks!
16:07 gmcharlt #endmeeting
16:07 Topic for #evergreen is now Welcome to the #evergreen library system channel! | We are publicly logged: http://irc.evergreen-ils.org/evergreen | Large pastes at http://paste.evergreen-ils.org | Can't speak? Make sure your nickname is registered and that you are identified to freenode services: https://freenode.net/kb/answer/registration
16:07 pinesol Meeting ended Wed Aug  8 16:07:34 2018 US/Eastern.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
16:07 pinesol Minutes:        http://evergreen-ils.org/meetings/evergr​een/2018/evergreen.2018-08-08-15.00.html
16:07 pinesol Minutes (text): http://evergreen-ils.org/meetings/evergr​een/2018/evergreen.2018-08-08-15.00.txt
16:07 pinesol Log:            http://evergreen-ils.org/meetings/evergree​n/2018/evergreen.2018-08-08-15.00.log.html
16:07 kmlussier gmcharlt++
16:07 * miker suspects gmcharlt held the meeting open to avoid scattering devs to the wind :)
16:07 berick thanks gmcharlt
16:07 miker gmcharlt++
16:08 Dyrcona gmcharlt++
16:08 abneiman gmcharlt++
16:08 Dyrcona And, me has to flee. Will back port omnibus to 3.0 tonight or tomorrow morning.
16:10 abowling joined #evergreen
16:58 mmorgan1 left #evergreen
17:05 jwoodard joined #evergreen
17:05 pinesol [evergreen|Mike Rylander] LP#1739460: Be more careful when gathering shelving locations - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=5b5b9cc>
17:33 pinesol [evergreen|Mike Rylander] LP#621459: Discard overlay target after merge in Z39.50 interface - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=abb3510>
17:33 pinesol [evergreen|Bill Erickson] LP#621459 Clear Z39 overlay target message after overlay - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=1fb2209>
17:39 berick late in the day to ask, but I'll toss it out there.  anyone know why the fine generator only looks at fines created after the circ due date instead of looking at all fines on the circ?  specifically to see if the max fine amount has already been billed.
17:39 berick looks like changing a circ due date can allow fines to exceed max fines, because of the above
17:39 * berick will open a bug tomorrow unless that's known / exptected
17:39 miker berick: is this related to the shifted fine dates? .. oh, it's not
17:40 miker sounds like staff are trying to game the system ;)
17:41 berick well, using the buttons in the staff client, but sure
17:42 miker what I mean is, that's what renew is for... if it's because of a closing, and they use the new emergency closing thing ... that clears those fines
17:46 berick yeah, all renewals were used up.  possibly staff didin't have perm to extend renewal count, but did have perm to change due date.  *shrug*.
17:47 berick in any event, i think it's reasonable to assume max fines is a hard limit on a circ
18:03 miker seems fair
18:31 pinesol News from qatests: Failed Running Evergreen browser client build/test - Expected 6 errors but encountered 7. <http://testing.evergreen-ils.org/~live>
18:37 Levex joined #evergreen
18:56 BigRig joined #evergreen
18:56 foobarrel joined #evergreen
19:05 jeff joined #evergreen
21:02 swoolley12 joined #evergreen
21:15 ReimuHakurei6 joined #evergreen
21:31 sandbergja joined #evergreen
21:47 Bkid3 joined #evergreen
23:18 Auctus3 joined #evergreen
23:58 Slade14 joined #evergreen

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat