Time  Nick             Message
00:16 dcook            paulderscheid: I didn't realize you were a librarian by education and software engineer by profession
00:17 dcook            Just like me :D
00:35 dcook            Didn't realize it's only early evening at Kohacon..
01:18 paulderscheid[m] dcook: Yeah, that’s life, I guess ^^
01:20 dcook            Well I wanted to go into computers when I was a kid, so I just took a roundabout trip to get there..
01:20 dcook            Did you go to Kansas from Germany?
01:21 dcook            I briefly thought about coming from Australia but it's not a great time atm..
01:34 tuxayo           paulderscheid: «Yeah, we bought the CO2 certs but you‘re still right»
01:34 tuxayo           Most CO2 compensations are fishy so it's quite the work to know if it's reliable but there is definitely a chance that it was something that work. Since they aren't much costly it's worth a try.
01:35 tuxayo           dcook: «librarian by education and software engineer by profession»
01:35 tuxayo           That looks like and often case in Koha :D
01:35 tuxayo           And hi ^^
01:35 tuxayo           *an often
01:37 dcook            Really? I think we're still fairly rare
01:37 dcook            Lots of programmers who have never worked in libraries heh
01:37 * dcook          worked his first library job back in 2006 O_O
01:37 dcook            and this koha-z3950-responder is driving me crazy..
01:37 dcook            Got it working on one server but not the other..
01:39 paulderscheid[m] dcook: yeah, with a transfer in Washington Dulles
01:40 tuxayo           dcook: Nick for sure and Andrew FH I think started librarians
01:41 dcook            I think Andrew might still be in a librarian type role at BWS
01:42 dcook            I do recall Nick's transformation though heh
01:47 tuxayo           dcook: Andrew went back to work in a library. I don't know what kind of work though$
01:47 tuxayo           *though.
01:51 dcook            This koha-z3959-responder is doing my head in..
01:51 dcook            I remember it was tough to get working the first time but this time is much worse..
01:51 dcook            Nothing inthe logs of course..
01:58 tuxayo           dcook: So it doesn't crash. But it doesn't do what is should?
01:58 dcook            Exactement
01:59 dcook            Looks like it's probably an issue with Net::Z3950::SimpleServer though..
01:59 tuxayo           Good luck with digging in a lib...
01:59 dcook            ikr
01:59 dcook            A lot of the server libraries on CPAN are awful in a lot of ways..
02:01 tuxayo           Just add some warns in the lib files in /usr/share/perl5 😵‍💫
02:02 dcook            Not so bad when you're in koha-testing-docker, but trying to troubleshoot a prod server.. :|
02:02 dcook            Cool when I run the script by hand, it just crashes silently
02:06 dcook            And now when I use koha-z3950-responder it fails to start too
02:06 dcook            Coooool
02:06 dcook            Because apparently it can't handle its own pid file..
02:06 dcook            Well I don't have time for this garbage. No z3950 today I geuss
02:15 dcook            I am stubborn though... and I think I might know why it wasn't logging..
02:21 dcook            bug 28364
02:21 huginn`          Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=28364 major, P5 - low, ---, tomascohen, CLOSED FIXED, koha-z3950-responder breaks because of log4perl.conf permissions
02:22 dcook            Ahhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh haha
02:22 wajasu           wooo
02:24 wajasu           i was hitting some of this a month ago.  one time i was deleting my log files for my image but one of the servers didn't create it's log file, so no i just 'truncate the log files'
02:25 dcook            wajasu: I bet you must've run into a lot of frustrations. We don't take good enough care of the upgrade path...
02:25 dcook            While I'm on a version where bug 28364 is "fixed", it's only fixed for new instances. Not upgraded ones.. :/
02:27 wajasu           i had to dig through some needles in a haystack, but getting to 22.05 from 3.10 and fixing the mysql seqno stuff was the time consuming part.
02:27 wajasu           i'l just work signoffs, but it's hard to tell what direction you guys are going in.
02:28 wajasu           i was going to work LDAP stuff, but I am not sure if the effort will be accepted.
02:33 dcook            wajasu: Hehe "direction"
02:33 dcook            I don't think there is a direction
02:34 dcook            Unless it's a small patch, it's really hard to get things accepted
02:34 dcook            I'm wishing luck to BibLibre at the moment with the staff interface redesign
02:35 dcook            I think they're arguably the second biggest Koha company, and if they're struggling to make changes, I don't know what hope an individual has
02:35 dcook            But... the collective effort is still worthwhile
02:35 dcook            No way that one or two people could handle as many bug fixes as the community does together
02:36 wajasu           yea.  i did a bunch a bunch of years ago, and even though i wasn't around, i still got all the bugzilla updates via email when they got push through over the years.
02:36 dcook            Yeah, I recall you from years ago heh
02:36 dcook            It used to be easier
02:36 wajasu           not a two week dev cycle for sure.
02:36 dcook            I've given up on quite a few things over the years
02:37 dcook            These days, I'm more interested in making it easier to customize Koha without upstreaming, which is unfortunate really...
02:37 wajasu           i see the workflow with QA, RM, etc now.  I didn't know that it exsited years ago and expected things to go out.
02:37 dcook            Yeah, we still had it years ago, but the overall process was less rigid
02:37 dcook            That said, there was a lot of bad code back in the day too
02:37 dcook            But bugs still get through anyway
02:38 dcook            At the end of the day, it's all about time
02:38 dcook            I know I also have less time now than I used to as I have more projects..
02:38 dcook            More clients, more projects, and less time
02:39 wajasu           i had one bug work, but it didn't pass perltidy formatting.  I would expect a QA person could run that and test.  Because by the time I go the fail, it was a year or so later.
02:40 wajasu           i did a bunch of XSLT for my library, but this new version was sufficient with template toolkit.
02:41 dcook            Yeah, I just had one fail in part due to perltidy formatting apparently, although it had passed the QA tools so I don't know why it wasn't good enough...
02:42 * dcook          is just a big grump these days
02:42 dcook            When I do QAing, I'll usually fix the small issues I find, as it just seems more efficient than sending it back
02:42 dcook            But yeah unfortunately no direction for Koha
02:43 dcook            I've suggested for many years that we choose one, but no go
02:43 wajasu           i'm glad they got the KTD.   I have my packer/virtlib working now with all your pieces (mounted koha, etc)
02:43 dcook            I admire the DSpace and Fedora Commons teams which are able to choose a direction and collectively work towards it
02:43 dcook            Yeah KTD is pretty sweet
02:44 wajasu           i made my koha kvm image cloud-init friendly, because I was pushing to linode.
02:45 wajasu           i think someone should write a  kohactl  (for all those separate scripts).
02:45 wajasu           kohactl create ...
02:47 wajasu           i also used selenium-ide to create a script to walk through the isntaller/onboarding, etc with my defaults.
02:48 wajasu           then i just run it on new ktd  or my images  to save time.
02:49 wajasu           if something like that could be in the dev or maintenance scripts directory, one could run a consistent environment for tests.
02:49 wajasu           from the commandline.
02:52 wajasu           also the misc4dev magic that loads the superlibrarian, and test data, could use that.  but you guys would have to agree to provide those scripts to pass tests when someone changes the installer.
02:53 wajasu           with ktd we should be able to have 3 libraries/branches and standard patrons, and such to test complex ILL scenarios.
02:54 dcook            You can talk to Joubu and tcohen about that
02:54 wajasu           i was going to doit in one image, but maybe a docker compose with multiple branhces running in the "docker" network would be useful.
02:54 dcook            Not sure what you mean in regards to kohactl
02:56 wajasu           well i didn't know about koha-common starting up and shutting down all the services, so i edited koha-create to do that, adding in the SIP, and z3950.
02:56 wajasu           if they were configured.
02:56 dcook            O_o
02:56 dcook            koha-create?
02:57 dcook            Maybe kohactl would be better than the hodgepodge of scripts that we have currently..
02:57 wajasu           yes
02:58 wajasu           thats the trend
02:58 wajasu           plus i go in a fix because multiple scripts need to share code.  i.e. the shell functions
03:01 wajasu           i got confused becaue i needed to run cp_debian_files after every script edit.
03:07 wajasu           well, it was nice to see the kohacon this week.
03:09 wajasu           i don't know what the atomic update stuff is about?  maybe when they get the new wiki going i can get up to speed.
03:10 tuxayo           dcook: «But yeah unfortunately no direction for Koha»
03:10 tuxayo           What do you mean? Like people throw patches and hope for them to be accepted? Without real planning like we see with ERM or staff redesign
03:11 tuxayo           Or even that doesn't show direction?
03:11 dcook            The ERM and staff redesign are the efforts of just one person/company
03:12 dcook            It's not like we all sat down, had a vote, and decided where we wanted Koha to be in the next 1-5 years
03:12 dcook            And then worked together to get there
03:12 dcook            The argument is that everyone has their own interests, so they pursue them, and that's that
03:12 dcook            And that does make a certain amount of sense..
03:13 dcook            If everyone voted that they wanted to spend the next year changing every colour to a different shade of pink, I probably wouldn't invest time in helping
03:13 wajasu           you guys are privy with the day to day.  i guess i could go read old meetings over the year.  so i will just do signoffs to speed up the workflow.
03:14 dcook            Well, I'm not in the most ideal timezone heh
03:14 dcook            Signoffs are good :D
03:14 tuxayo           It looks rare that a company has moments where we can invest in some change in a given direction. So if we we all sat down, had a vote, and decided where we wanted Koha to be in the next 1-5 years then several companies need to have at the same time the free ressources to make happen the decisions.
03:14 tuxayo           That's very theorical ↑
03:15 tuxayo           > The argument is that everyone has their own interests, so they pursue them, and that's that
03:15 tuxayo           Very organic and doesn't require much hierarchy and coordinated ressources.
03:16 dcook            It's true and it's gotten us this far
03:16 dcook            And really there are a few companies that do operate behind the scenes to coordinate their resources/time
03:16 tuxayo           > you guys are privy with the day to day.
03:16 tuxayo           What do you mean wajasu ? That's out of my english knowledge
03:16 dcook            tuxayo: We're au courant heh
03:17 wajasu           have knowledge of the day to day discussions/efforts
03:17 tuxayo           Ok, thanks ^^
03:17 tuxayo           Merci beaucoup.
03:18 tuxayo           wajasu: «so i will just do signoffs to speed up the workflow.» That's the biggest bottleneck with QA. And if it wasn't so much a bottleneck, there would be more care of the less obvious proposals that get stuck.
03:19 wajasu           that's what i thought.
03:19 dcook            Btw if you're not familiar with it already... https://dashboard.koha-community.org/
03:20 wajasu           yes. i got into that.  love the build section.
03:20 dcook            At the moment, it looks like the biggest count is Failed QA heh
03:20 dcook            Yeah the build section is cool
03:20 dcook            bug 31614
03:20 huginn`          Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=31614 enhancement, P5 - low, ---, koha-bugs, NEW , Add configurable timeout for Zebra connections
03:20 dcook            Search may be slow but this at least lets it finish..
03:20 wajasu           once you get LTS in there, people can work backports with that one in mind.
03:21 tuxayo           What do you mean?
03:22 wajasu           well, if you have somone backporting a fix to a prior version, they would prioritize the LTS version.
03:23 tuxayo           Backports work in cascade. When it's merged in main, the RMaints of stable picks it and only then the RMaint of oldstable. And than oldoldstable. And then LTS
03:23 wajasu           i see.
03:24 tuxayo           The introduction of LTS won't change that. And it make the work easier since you don't get 2 years of conflicts. Each backporting step get part of the changes?
03:24 tuxayo           *changes.
03:25 tuxayo           > And really there are a few companies that do operate behind the scenes to coordinate their resources/time
03:25 tuxayo           dcook, About that, since two months I'm being paid for my time on QA and Signoff by some of the support companies here. ^^
03:25 dcook            tuxayo: Oh nice! I was wondering about that!
03:26 dcook            Glad to hear you're part of the cabal!
03:26 dcook            Hehe. But seriously. Good to hear that you're getting paid for your efforts.
03:27 wajasu           if work a signoff, and really get into the bug, but it fails.  can i supply the patch to help fix.  then let someone else sign it off?
03:28 wajasu           since you have QA as followup, could i signoff if i contributed a patch?
03:28 wajasu           don't want to step on any ones toes that worked a patch though.
03:29 dcook            wajasu: Yeah, typically you can sign off the first patch, you supply your patch, then someone else (including the original author) can sign off your patch
03:29 dcook            It's a good way of moving things along
03:29 wajasu           great
03:30 tuxayo           > Oh nice! I was wondering about that!
03:30 tuxayo           RMaint of oldoldstable + bringing back from the dead the hold rules checker (bug 23732) and the trainings I'm getting for managing my company and fact that i'm working part time causes that I'm not putting much work on QA&SO though :(
03:30 huginn`          Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23732 enhancement, P5 - low, ---, victor, Patch doesn't apply , Hold rules checker: show matched rules and syspref values to help understand why a hold is possible or not
03:30 tuxayo           Maybe one of the KohaCon talk is about this funding.
03:30 tuxayo           «Funding QA and Joint Development»
03:30 tuxayo           «Brendan Gallagher (ByWater Solutions), Paul Poulain (BibLibre), Andrew Auld (PTFS Europe Limited), Jonathan Field (PTFS Europe Limited), and Nate Curulla (ByWater Solutions)»
03:30 tuxayo           These people are very suspect :)
03:30 dcook            hehe
03:31 dcook            I'll have to watch any recordings they do
03:31 tuxayo           KohaCon day 2 timestamp 7:27:00
03:31 tuxayo           Yet to watch it.
03:33 dcook            Hmm I"m at that timestamp but don't see anything..
03:33 dcook            Ah there we are
03:33 dcook            https://youtu.be/zjL4LHEH74g?t=26861
03:34 dcook            Time for me to go do other tings though I think..
03:35 tuxayo           > Hehe. But seriously. Good to hear that you're getting paid for your efforts.
03:35 tuxayo           Thanks ^^ Even with not enough time to work on that, that's still the minimum to be able not need to use savings anymore. So I can work forever on Koha now ♾️ :D (until the end of abundant oil that will screw things up ^^")
03:37 tuxayo           > don't want to step on any ones toes that worked a patch though.
03:37 tuxayo           Don't worry about that. Adding follow-up is very welcome. That's a very nice thing of our workflow with bugzilla, anyone can add patches to any submission. With merge requests, that would be a barrier.
03:38 dcook            (good point about the merge requests)
03:38 dcook            And I hear you on working on Koha until the apocalypse
03:39 tuxayo           🔥💃🔥
03:41 wajasu           in my case i still use mt ancient email account for bugzilla.  but gitlab won't accept my email to signup. the admin for gitlab for koha could allow my email domain. so i can't do merge requests until i get a new email one day.
03:42 tuxayo           > the admin for gitlab for koha could allow my email domain
03:42 tuxayo           That's a global GitLab.com issue. Only they can allow your domain.
03:46 tuxayo           Going to bed finally, tomorrow patch testing session with french speaking librarians :)
03:46 tuxayo           And Monday with English speaking ones.
03:46 tuxayo           I say librarians but anyone can come, it's just that I try to get new workforce in the community, not converting existing one to testing.
03:47 tuxayo           Even though if for every patch submitted, the dev would signoff two other patches, that would solves the bottlenecks but slow new patch submissions rates. But time not wasted in not doing endless rebases should balance that of almost. But anyway, getting a lot of librarians to test would be great also.
03:54 wajasu           https://docs.gitlab.com/ee/user/admin_area/settings/sign_up_restrictions.html#allow-or-deny-sign-ups-using-specific-email-domains
03:57 wajasu           say goodnight gracie
06:03 marcelr          hi #koha
06:54 reiveune         hello
06:54 wahanui          hi, reiveune
07:02 alex_            Bonjour
07:02 wahanui          que tal, alex_
07:16 dcook            hola Europeans
07:16 dcook            time for me to go hooooome
07:44 magnuse_         kia ora dcook
07:47 magnuse_         i have a tt file of 1300++ lines and suspect a problem with the nesting of the tt "tags". is there a good way to debug tt? like syntax checking/highlighting or similar?
07:49 magnuse_         logs say: Template process failed: file error - parse error - /usr/share/koha/opac/htdocs/opac-tmpl/bootstrap/en/modules/opac-memberentry.tt line 1372: unexpected end of input at /usr/share/koha/lib/C4/Templates.pm line 122.
08:51 mtj              hi magnuse_, try the debug option http://www.template-toolkit.org/docs/manual/Directives.html#section_DEBUG
08:53 magnuse_         mtj: thanks, i have missed that!
09:01 mtj              magnuse_: tpage could be useful for debugging syntax errors
09:01 mtj              http://www.template-toolkit.org/docs/tools/tpage.html
09:08 magnuse_         interesting!
09:09 magnuse_         i did get lucky though, and found some improperly nested tt tags quite fast
11:29 tcohen           hola #koha!
11:48 tcohen           happy friday
11:51 tcohen           can anyone reproduce the t/Auth_with_shibboleth.t failure locally? (I cannot)
11:59 paulderscheid[m] I think I‘m going to write a new tt extension for VSCode this year.
12:07 tuxayo           hi #koha :)
12:11 tcohen           paulderscheid[m]: \o/
12:14 * tcohen         needs coffee, typed a bug number in 3 different ways on each patch
12:14 tcohen           paulderscheid[m]: what do you want to change from the current TT extension?
12:15 magnuse_         happy equinox, #koha!
12:15 tcohen           <3
12:47 mtj              hi tcohen, i did make a little progress on the failing sele patrons.t test
12:48 mtj              sele seems to have a general problem with the 'processing' alert box, when updating a table
12:57 tcohen           oh
12:57 tcohen           we need a sleep
12:57 tcohen           ?
12:58 tcohen           I really need help with Auth_with_shibboleth.t
12:58 tcohen           seems like a really easy fix, but I cannot reproduce
12:59 tcohen           please: try locally, if it fails, call me :-D
13:09 pastebot         "mtj" at 127.0.0.1 pasted "Auth_with_shibboleth.t" (16 lines) at https://paste.koha-community.org/15167
13:11 pastebot         "mtj" at 127.0.0.1 pasted "Auth_with_shibboleth.t 2" (84 lines) at https://paste.koha-community.org/15168
13:12 mtj              hi tcohen: i get that failure ^
13:12 mtj              hmm, but i havent pulled recently
13:12 tcohen           can you share that dump?
13:12 tcohen           on pm
13:13 tcohen           and also, where are you running it?
13:15 paulderscheid[m] tcohen: I feel like the autocompletions and suggestions could be better
13:16 tcohen           right, it could offer closing things, etc
13:19 mtj              tcohen: the dump was made just after 'ku' finished booting...
13:19 mtj              $ mysqldump  --host=db --user=koha_kohadev --password=password koha_kohadev > back.sql
13:20 mtj              then...
13:20 mtj              mysql  --host=db --user=koha_kohadev --password=password koha_kohadev < back.sql ; perl ./t/db/sele/patrons.t
13:20 tcohen           ok, running as root makes it fail
13:20 tcohen           thanks!
13:21 mtj              loading the db before starting a test, was a good method for debugging the sele/patrons.t tests
13:22 mtj              85030fc38a (HEAD -> z11, origin/master) Bug 7021: DBRev 22.06.00.051
13:22 huginn`          Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=7021 enhancement, P5 - low, ---, emmi.takkinen, Pushed to master , Add patron category to the statistics table
13:23 mtj              ^ ive been testing on that older commit
13:24 tcohen           @later tell marcelr you owe me a beer
13:24 huginn`          tcohen: The operation succeeded.
13:25 tcohen           097b2afb17da42878c8fffaf7ebdab1ad0cb49fa is the first bad commit
13:25 tcohen           commit 097b2afb17da42878c8fffaf7ebdab1ad0cb49fa
13:25 tcohen           Author: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
13:25 tcohen           Date:   Fri Sep 16 07:07:00 2022 +0000
13:25 tcohen           git_bisect++
13:25 ashimema         lol
13:26 tcohen           mtj++
13:33 tcohen           regression_tests++
13:34 tcohen           a minor refactoring broke a feature, and we detected it
13:34 tcohen           that's great :-D
13:35 tuxayo           Ah great
13:43 mtj              ill paste a bit more info about the failing sele/patrons.t soon
13:50 oleonard         o/
13:52 paulderscheid[m] I just had a fun thought: Why don't we put animated clocks of the different timezones on the dashboard?
13:53 oleonard         That would be fun
14:05 * tcohen         likes
14:27 wizzyrea         hi friends - i have a question and I've been going around and around and it's probably very easy and i just can't find how to do it - how does one set up the password protection on the zebra z3950 service? Not the incoming records - the z3950 access from outside the library
14:34 tcohen           you need to have the relevant authentication section on the publicserver definition
14:35 tcohen           for reference, for Koha's internal use we generate /etc/koha/sites/<instance>/zebra.passwd and that info can also be found in koha-conf.xml for authorityserver and biblioserver
14:35 mtj              hi wizzyrea
14:41 mtj              i pmed you
14:42 tcohen           rmaints: please take a careful look at your rel_<version>_candidate keywords on bugzilla. It seems there are lots of bugfixes that never get backported
14:42 tcohen           ^^ tuxayo lukeg liliputech
14:42 tcohen           example: bug 30730
14:42 huginn`          Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=30730 major, P5 - low, ---, shi-yao.wang, Pushed to master , Holds to Pull should not list items with a notforloan status
14:43 milktoast_       morning all...
14:43 milktoast_       I've exported the data from our "handy library manager" and it has it's own names for areas...
14:44 milktoast_       can someone look at these names and see what MARC21 catalog number they match so I can change them before importing into Koha?
14:44 milktoast_       IDNO	TITLE	AUTHOR	TYPE	COAUTHORS	PUBLISHER	PUBLISHED	PLACE	SERIES	EDITION	CATEGORY	LEVEL	FORMAT	EDITOR	ILLUSTRATOR	PAGES	PHYSICAL	DEWEY
14:45 milktoast_       each one is at the top of a column in libreoffice calc
14:46 tcohen           you need to talk to a librarian or a migrations expert
14:46 milktoast_       :-)
14:46 oleonard         MARC21 documentation should help too: https://www.loc.gov/marc/bibliographic/
14:47 oleonard         milktoast_ : IDNO be not be relevant if it was an internal identifier in your previous software
14:48 oleonard         milktoast_ And category may be something you want to convert to an itemtype or authorized value...I'm guessing
14:48 milktoast_       Thanks, I found that .. like <tcohen> said,  I need to  talk to a librarian...
14:50 milktoast_       ok, I'll print a paper and take toa Library and ask them what they think each one should be
14:50 milktoast_       it is 14K books in that system
14:53 tcohen           IDNO is usually put on 001
14:53 tuxayo           tcohen: I never had in my workflow to check rel_<version>_candidate 😱
14:53 tcohen           OMFG
14:53 tcohen           haha
14:53 mtj              tcohen, wizzyrea: i updated bz17035 regarding the public z3950 question
14:54 tuxayo           tcohen: to be fair, every patch should be examined by RMaint. And have a comment left if skipped.
14:54 reiveune         bye
14:55 tuxayo           And I check every BZ notification for if someone request a backport that I skipped
14:55 tuxayo           So that should be fool proof.
14:57 tuxayo           However I wouldn't bet the RMaints for stable and oldstable can stay on top of all there BZ notifications. Because they touch more tickets than me with oldoldstable. And the time to spend backporting and testing is much more. So less time left to stay on top of more notifications.
15:11 tcohen           tuxayo: yeah, we should team up for lowering the weight on our shuoldrs
15:11 tcohen           I try to add the keyword myself when I'm pretty sure things should be backported
15:15 tuxayo           Good, I have nothing in  rel_21_05_candidate :)
15:17 tuxayo           > we should team up for lowering the weight on our shuoldrs
15:17 tuxayo           There can totally be multiple people RMainting a given branch a split the backport & test load. Once or twice stable or oldstable had two people from bywater.
15:17 tuxayo           *and split
15:18 wizzyrea         hi mtj :)
15:19 wizzyrea         tcohen - thank you I was hoping it was literally that easy but didn't dream it actually was
15:21 mtj              hiya wizzyrea
15:23 mtj              tcohen: perhaps we allow a bit of time for support people to update systems, before pushing bz17035
15:36 ashimema         splitting does mean maintaining a queue very carefully though..
15:36 ashimema         but it is doable
15:37 ashimema         I think one of the biggest issues is really just that rmaints aren't always given the time that the job requires
15:38 ashimema         when I was an rmaint I checked my queue every morning and worked through it doing backports.. I made sure that on a friday my queue was always empty.. that can take a significant time
15:38 ashimema         especially for stable
15:45 mtj              hi ashimema
15:46 ashimema         hi
15:46 ashimema         I had a question for you....
15:46 * ashimema       digs through notes
15:46 mtj              tcohen: git-bisect says bug 31565 broke the sele/patrons.t test
15:46 huginn`          Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=31565 normal, P5 - low, ---, lari.taskula, Pushed to master , Patron search filter by category code with special character returns no results
15:46 mtj              ... i havent click why yet tho :)
15:48 ashimema         found it.. do you think it's viable to add a simple option in the packages to set up letsencrypt certs for koha instances
15:49 ashimema         Our infra team here are suggesting it's a bit of a pain getting Certbot to play nicely with the koha vhosts we ship with packages out of the box...
15:49 ashimema         honestly.. I've not looked into it at all myself
15:50 ashimema         there's money in the pot if it's something you think as packaging guru you could chomp through?
15:50 mtj              hmm, yes and yes
15:51 ashimema         excellent
15:51 wahanui          darn tootin' it is.
15:51 mtj              i have a bit of experience setting up lets-encrypt stuff with koha
15:52 ashimema         had a feeling you might have
15:52 ashimema         feel free to punt me a suggestion of what it would entail and a quote when you have a moment..
15:53 mtj              ok, ill have a little look to refresh my memory
15:55 mtj              i recall an annoying gotcha was the cert renewal process, it wanted to run a webserver on port 80 to do the update
15:55 mtj              that was the easy/default option
15:56 tcohen           school drop
15:56 mtj              ..but to do that would require apache to be shutdown :/
15:57 mtj              good news is that there are a bunch of other ways to do the renewal
15:58 ashimema         excellent
15:58 wahanui          darn tootin' it is.
15:58 ashimema         I know Mojo has support for it these days.. though I've not looked into the details..
15:59 ashimema         we may be able to do renewals inside our koha code..
16:01 ashimema         ho.. when going to add a bug I just found https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=15303
16:01 huginn`          Bug 15303: enhancement, P5 - low, ---, mirko, CLOSED FIXED, Letsencrypt option for Debian package installations
16:04 mtj              ok, that looks like a great start
16:04 ashimema         I'm confused now.. I have no idea what that does or how my colleagues missed it
16:04 ashimema         lol
16:10 mtj              well, its been committed
16:13 mtj              i confess to setting up my koha/certbot stuff manually, (bofh sysadmin)
18:11 tcohen           mtj[m]: could you please upgrade libmojolicious-plugin-oauth2-perl to 2.02? thanks!
18:22 mtj              tcohen: try the 'koha-staging dev' repo
19:37 mtj              tcohen: i fixed the sele/patrons.t bug, ill send a patch later
19:37 mtj              its a silly bug :/