Time  Nick          Message
23:29 dcook         Might be because of the backports..
23:29 dcook         That's interesting about Stretch and would explain a lot...
21:36 tuxayo        pfiou, the whole thing has great educational value about how things can get tangled
21:31 cait          although the version for stretch is not 92 but 97 - it seems ok
21:31 cait          the last comment on the bug helps
21:31 tuxayo        Indeed, I got very confused about what factor telling an instance is expected to have the issue
21:31 tuxayo        > (thankful for that really)
21:30 lukeG1        Once it was set it quit dying on me when I 'ku'
21:29 tuxayo        Od, that shouldn't be needed
21:29 lukeG1        and my personal problem was I had no $KOHA_IMAGE set
21:29 cait          (thankful for that really)
21:28 tuxayo        [1] not asking what happened there, too scared.
21:28 cait          actually we only had one server with my demo instance on it dying :)
21:28 tuxayo        You can tell to bag that another week of holiday is an absolute necessity!
21:28 tuxayo        It's like the whole galaxy has aligned with the intention of preventing you to work after coming back from holidays [1]. Nuking part the world's Kohas if needed.
21:28 cait          not all recent...
21:28 tuxayo        (actually only recent OSes)
21:28 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653 blocker, P5 - low, ---, dcook, Signed Off , Plack stopped working - JSON validation timeout
21:28 tuxayo        and 15 min later aleisha reported what would turn out to be bug 23653 turning into a terrible mess that would affect all recent Koha instances in the world
21:28 tuxayo        - then you mistyped "kd" in the channel
21:28 tuxayo        - you had the issue with the koha container crashing for some reason
21:28 tuxayo        - you had the issue with the koha-testing-docker .env file
21:27 tuxayo        lukeG1: I laughed so much when earlier I realized this:
21:27 lukeG1        Hey tuxayo
21:25 koha-jenkins  Project Koha_19.11_U20 build #1: UNSTABLE in 46 min: https://jenkins.koha-community.org/job/Koha_19.11_U20/1/
21:23 tuxayo        lukeG1 still around?
21:21 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653 blocker, P5 - low, ---, dcook, Signed Off , Plack stopped working - JSON validation timeout
21:21 wahanui       bug 23653 is probably the Swagger-apocalypse!!!!! 🔥🔥 https://url.lesamarien.fr/elmo-fire 🔥🔥 (the rumor says it started there: https://url.lesamarien.fr/exibit-1)
21:21 tuxayo        Bug 23653!
21:21 tuxayo        And we managed to not mention:
21:21 tuxayo        > we just went over the hour
21:20 tuxayo        Somehow URLs sent ended up as if it was #info
21:19 davidnind     thanks for your patience everyone, we just went over the hour (an extra 12 minutes)!
21:19 cait          hm?
21:19 tuxayo        https://meetings.koha-community.org/2020/development_irc_meeting_1_july_2020.2020-07-01-20.03.html
21:19 tuxayo        Why do my links ended up in the minutes? :o
21:17 davidnind     :)
21:17 cait          we might just keep you on
21:17 ashimema      davidnind++
21:17 cait          might have been a mistake ;)
21:17 ashimema      community++
21:17 davidnind     first time I've chaired a development meeting I think..
21:16 davidnind     thanks everyone!
21:16 huginn        Log:            http://meetings.koha-community.org/2020/development_irc_meeting_1_july_2020.2020-07-01-20.03.log.html
21:16 huginn        Minutes (text): http://meetings.koha-community.org/2020/development_irc_meeting_1_july_2020.2020-07-01-20.03.txt
21:16 huginn        Minutes:        http://meetings.koha-community.org/2020/development_irc_meeting_1_july_2020.2020-07-01-20.03.html
21:16 huginn        Meeting ended Wed Jul  1 21:16:31 2020 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
21:16 davidnind     #endmeeting
21:16 davidnind     #info Next meeting: 15 July 2020, 14:00 UTC
21:16 tuxayo        davidnind++
21:16 tallerjoy     davidnind++
21:15 cait          davidnind++ :)
21:15 tallerjoy     yes, that works for me
21:15 ashimema      #info davidnind has been an excellent chairperson, thankyou
21:15 davidnind     UTC
21:15 * tallerjoy   will have to do some maths
21:15 tallerjoy     UTC time i assume.
21:15 cait          sounds good
21:15 ashimema      Works for me
21:14 tallerjoy     thanks!
21:14 ashimema      Worthwhile meeting everyone, thanks
21:14 davidnind     14:00 next time and two weeks time?
21:13 cait          it's also something that's asked about reasonably often
21:13 davidnind     #topic Set time of next meeting
21:13 davidnind     tuxayo++
21:13 davidnind     #action davidnind Find and update relevant places to record Perl version required, including release notes, manual and wiki (one source of the truth)
21:13 tuxayo        davidnind: I can also help, these system requirements things have been a lot in my mind lately ^^
21:12 cait          davidnind: maybe refer to release notes too? people best check that
21:12 cait          yeah, that's very hard to follow as advice :)
21:12 davidnind     #info davidnind Find and update relevant places to record Perl version required, including manual and wiki (one source of the truth)
21:12 tuxayo        oleonard should be happy
21:12 davidnind     Just pick the right page, and your good!
21:12 cait          #action cait to propose discussion about supported browsers at next meeting (IE especially)
21:11 cait          Full system recommendations can be found on the official Koha wiki along with the developer documentation: http://wiki.koha-community.org
21:11 cait          it refers to wiki, which is not really true
21:11 cait          we shoudl update that too
21:11 cait          https://koha-community.org/manual/20.05/en/html/intro.html#koha-recommendations
21:11 cait          aren#t you an hour earlier even? ;)
21:10 ashimema      /me is getting tired
21:10 cait          checking manual
21:10 tuxayo        This is very relevant for OS and DBMS, they can change in the middle of a cycle
21:10 tuxayo        > refer to release notes of the version you want to install
21:10 ashimema      😉
21:10 ashimema      Reference the definitive source
21:10 cait          sorry, late
21:09 cait          i mean, the information would still be correct
21:09 ashimema      I like that.. refer to release notes
21:09 cait          that way if we change for an upcoming veresion... things would still work
21:09 tuxayo        To have less sources of truth
21:09 cait          s
21:09 cait          for exact version
21:09 cait          we could also say: refer to release notes of the version you want to install
21:09 cait          good question
21:09 tuxayo        Or make the other places link to this?
21:09 cait          #action ashimema to update requirements on website
21:09 tuxayo        Like to somewhere else maybe?
21:08 tuxayo        Good otherwise that would another source of inconsistency ^^"
21:08 tuxayo        > hm we don't list versions there
21:08 cait          yeah we probably should
21:08 ashimema      I'm happy to add that
21:08 ashimema      We should Cait..
21:08 davidnind     #info Minimum recommended Perl version is 5.14 (until need to use a more recent version is identified)
21:08 cait          hm we don't list versions there
21:08 cait          https://koha-community.org/download-koha/
21:07 tuxayo        *5.14
21:07 cait          we should update website, manual and release notes scripts
21:07 cait          yep i think 3.14 is safe and should be done
21:07 ashimema      I reckon we've all just agreed, so let's just action it
21:07 thd           tuxayo++
21:07 tuxayo        it' just to reflect the current state of the code. There is no actual change
21:07 thd           without objection
21:07 ashimema      Cool
21:06 davidnind     is this a voting thing, or just an info and then action to document?
21:06 ashimema      No point in artificially inflating it
21:06 tuxayo        Perl seems to be very backward compatible IIUC so no issues on that side.
21:06 tuxayo        > stick with debian/ubuntu (not go higher than there?)
21:06 cait          +1
21:06 tuxayo        +1
21:06 ashimema      3.14 seems reasonable to me until we find were wanting to use something from a more recent version
21:05 tuxayo        Oh yes!
21:05 davidnind     we could create a loop, and refer it back to the mailing list for discussion :)
21:05 ashimema      I have in the past.. but I'm not of the same opinion now 😉
21:05 cait          and stick with debian/ubuntu (not go higher than there?)
21:04 cait          and i haven't seen someon eargueing for anything beyond 3.14 yet :)
21:04 ashimema      Lol
21:04 cait          the last argument seemed that we push for the version that has the features we want
21:04 tuxayo        The conclusion was to add it to the next dev meeting :P
21:04 ashimema      The conversation seemed to be ongoing on lists 😀
21:04 cait          hm it looked like 3.14?
21:04 ashimema      I reckon defer it..
21:04 ashimema      Did the next one get anywhere on the lists?
21:03 ashimema      Perfec
21:03 davidnind     defer the Perl version to the next meeting?
21:02 davidnind     we are coming up on the hour
21:02 davidnind     hope I got that right!
21:02 tuxayo        thanks davidnind
21:02 davidnind     #action tuxayo - prepare draft update to SQL11 to improve consistency for descriptions in kohastructure files
21:02 tuxayo        I'm writing the action
21:02 cait          cool
21:01 ashimema      I can help tomorrow
21:01 tuxayo        ok!
21:01 ashimema      Fancy investigating that side of it and drafting a guideline tuxayo?
21:00 davidnind     so, looks like an updated guideline is required - who would like to do that?
21:00 tuxayo        It that the same for the table comment?
21:00 davidnind     sounds sensible to me, but I know nothing about this :)
20:59 ashimema      As apposed to mixing and matching
20:59 ashimema      Struggling to follow in phone.. but it sounds like the COMMENT version works for both.. so I think I'd advocate moving to that consistently ?
20:57 tuxayo        Seems right
20:57 tuxayo        > not sure if it looks right (sparse comments?)
20:57 tuxayo        http://schema.koha-community.org/master/tables/search_field.html
20:57 tuxayo        For search field it's ok
20:56 cait          not sure if it looks right (sparse comments?)
20:56 tuxayo        `name` varchar(255) NOT NULL COMMENT 'the name of the field as it will be stored in the search engine',
20:56 tuxayo        Ok, so it's like that in SQL
20:56 cait          http://schema.koha-community.org/master/tables/collections_tracking.html
20:56 ashimema      I also don't know if you need different options when dumping schema via schemaspy to pick them up if we don't already..
20:55 ashimema      How does it appear in the schemaspy schema
20:55 ashimema      collections_tracking has COMMENT at the field level
20:55 koha-jenkins  Project Koha_19.11_U18 build #192: FIXED in 44 min: https://jenkins.koha-community.org/job/Koha_19.11_U18/192/
20:55 wahanui       Congratulations!
20:55 koha-jenkins  Yippee, build fixed!
20:54 tuxayo        CREATE TABLE `biblio` ( -- table that stores bibliographic information
20:54 tuxayo        * https://gitlab.com/koha-community/Koha/-/blob/master/installer/data/mysql/kohastructure.sql#L132
20:54 tuxayo        whoops
20:53 tuxayo        https://gitlab.com/koha-community/Koha/-/blob/master/installer/data/mysql/kohastructure.sql
20:53 tuxayo        ashimema: this is an example of the alternative?
20:53 ashimema      In the dbic context it's just the addition of information in the POD.. so describes the fields use in the code rather than having to refer to schemaspy or the kohastructure to find out.
20:53 cait          maybe we can find something that works for both
20:53 cait          I'd hate to lose the comments on schema
20:52 ashimema      The -- version only appears in the dumpfile so can't get picked up by dbic dump.. hense why only 'COMMENT' works in the dbic context..
20:52 tuxayo        schemaspy uses it for the table description
20:52 tuxayo        And it's on tables
20:52 tuxayo        Two instance of "COMMENT="
20:52 ashimema      I would hope schemaspy would grab both
20:52 tuxayo        https://gitlab.com/koha-community/Koha/-/blob/master/installer/data/mysql/kohastructure.sql
20:51 cait          We have updated Schema Spy a few times since - not sure if there was a change
20:51 cait          https://wiki.koha-community.org/wiki/Coding_Guidelines#SQL11:_All_fields_added_to_the_database_must_be_documented_in_kohastructure.sql
20:50 tuxayo        I'm searching
20:50 koha-jenkins  Project Koha_19.11_D8 build #197: FIXED in 40 min: https://jenkins.koha-community.org/job/Koha_19.11_D8/197/
20:50 wahanui       Congratulations!
20:50 koha-jenkins  Yippee, build fixed!
20:50 cait          it's been a whie, do we have an example for comment?
20:50 davidnind     I like consistency so ...
20:49 tuxayo        I suppose both show? Otherwise part of them is much less useful :o
20:49 cait          Example: http://schema.koha-community.org/master/tables/items.html
20:49 cait          schema?
20:49 cait          I tihnk we picked what will show up on schema
20:48 tuxayo        https://wiki.koha-community.org/wiki/Development_IRC_meeting_1_July_2020
20:48 * cait        lost the wiki page again
20:48 ashimema      Back.
20:48 tuxayo        Is this good?
20:48 tuxayo        About item 2: «the former appears in schema classes at dump time.»
20:47 davidnind     was any general consensus reached on this with developers on the minimum version (I sort of followed some of the discussion, but....)
20:46 davidnind     we'll tackle the Perl version until ashimema is back
20:46 ashimema      Brb
20:46 ashimema      I was just surprised to see the two varieties in the dB structure and wondered if there was a reason to go for one or the other
20:45 ashimema      This next one was me too
20:45 ashimema      Hehe
20:45 tuxayo        "Seeking cosponsors"
20:44 thd           #info Thomas Dukleth, Agogme, New York City
20:44 davidnind     volunteered!
20:44 ashimema      Lol
20:44 ashimema      Joubu.. for his absence 😛
20:44 davidnind     #action ashimema - add a draft coding guideline for direct notation rather than indirect object notation calls
20:42 davidnind     so who gets this one? :)
20:42 tuxayo        ++
20:42 davidnind     #info  start removing 'indirect object notation' calls - bug 25898 and add as a coding guideline
20:42 cait          especially the ones new to the project
20:41 ashimema      Ack.. I sense more homework coming
20:41 cait          it always helps if you can point to something written to help people
20:41 ashimema      Yup.. we should guideline it and add it to QA tools
20:41 cait          davidnind: i don't think so, because code base is big and undocumented expectations are hard
20:41 davidnind     (not sure whether developers need a guideline for everything, or they automagically pick these things up)
20:41 cait          for these kind of things if we are serious about it
20:41 cait          always good to have a coding guideline and qa tools updated
20:40 ashimema      Yup
20:40 tuxayo        The manual equivalent of git blame but for the wiki blames ashimema for this meeting item :P
20:40 cait          it looks like it would be reaosnably safe to replace?
20:40 davidnind     or is it just something that needs to be done?
20:39 ashimema      The latter is more modern and is where perl appears to be headed in general
20:39 davidnind     does this need to be a coding guideline?
20:39 ashimema      It's 'new thing' Vs 'thing->new' syntactically in code
20:38 ashimema      The first one is about advocating more modern best practice.. we mostly do it already
20:38 tcohen        +1
20:38 davidnind     any takers for this one?
20:38 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=25898 enhancement, P5 - low, ---, julian.maurice, Needs Signoff , Prohibit indirect object notation
20:38 davidnind     start removing 'indirect object notation' calls - bug 25898
20:38 koha-jenkins  Project Koha_19.11_D9 build #195: SUCCESS in 43 min: https://jenkins.koha-community.org/job/Koha_19.11_D9/195/
20:38 davidnind     there are no names beside the agenda items...
20:37 davidnind     thanks tallerjoy!
20:37 davidnind     #topic General development discussion (trends, ideas, ...)
20:37 davidnind     agree, always important that the rationale for a guideline or rule is identified/documented
20:36 tallerjoy     good point!
20:36 cait          yeah i think translators coudl refer to the rule as well
20:36 tuxayo        yep
20:36 tallerjoy     and also to see what other languages have different inclusive language concerns as well.
20:35 tuxayo        That depend of the culture and background: the POV of what kind of language changes are desirable.
20:35 tuxayo        Indeed it was at the proposal stage, so a draft submitted at a meeting would help more people more of us understand the reasons.
20:35 tuxayo        «why the change was proposed»
20:34 davidnind     Once that is done I guess we will need some bugs for any changes required, but that will come later
20:34 ashimema      Hehe
20:34 tallerjoy     heh
20:33 ashimema      Toomanyjoys++
20:33 davidnind     #action tallerjoy - Draft TERM3 for for inclusive language
20:33 ashimema      Lol
20:33 tcohen        tallerjoy++
20:33 tallerjoy     :D
20:33 cait          :P
20:33 cait          tallestjoy++
20:33 bag           tallerjoy++
20:32 tcohen        talljoy++
20:32 tuxayo        +1
20:32 ashimema      Perfect
20:32 davidnind     tallerjoy++
20:32 cait          context
20:32 tallerjoy     thanks cait.
20:32 cait          yep
20:32 tallerjoy     It would be easy to draft something for a TERM3 section.  I can do that.  I think also folks wanted some context as well about why the change was proposed, and I can include that as well.
20:31 cait          if that makes sense?
20:31 cait          explicit and implicit
20:31 cait          also a terminology list and stating that you want to use inclusive language are different things
20:30 cait          so someone hands something in... we think it's bad... QA could refer to general rule about inclusive language
20:30 cait          only if you have already listed it
20:30 tallerjoy     i had that though, but also tuxayo's thought
20:30 tuxayo        Doesn't TERM1 work?
20:30 cait          that would be a good spot
20:29 cait          TERM3 maybe, after gender neutral pronouns
20:29 cait          i think having a guideline as we find more would be helfpul
20:29 tallerjoy     this is a terminology issue mainly, so putting it there seems most appropriate.
20:29 cait          we only deal with deny list etc now
20:29 cait          should we have a more general rule?
20:29 tuxayo        https://wiki.koha-community.org/wiki/Coding_Guidelines#TERM1:_The_agreed_set_of_library_terms_must_be_used
20:29 davidnind     #info Terminology list updated for more inclusive language for https://wiki.koha-community.org/wiki/Terminology
20:29 cait          was looking at coding guidelines
20:29 kidclamp      bye all. gotta run
20:29 tallerjoy     that link?
20:29 cait          ah ok, the terminology
20:28 tallerjoy     https://wiki.koha-community.org/wiki/Terminology
20:28 cait          do you have the link maybe?
20:28 tallerjoy     so, it didn't seem relevant to propose it?  But i can still write one if you'd like it.
20:27 tallerjoy     Well, i was going to, but it appears that someone already updated the coding guidelines terminology page.
20:27 ashimema      You were drafting a language guideline
20:27 davidnind     the action point for inclusive language
20:26 tallerjoy     yes?
20:26 wahanui       well, talljoy is here :)
20:26 davidnind     talljoy?
20:26 davidnind     #info "skeleton" template we could copy for new pages - deferred until Boostrap update merged
20:26 cait          ;)
20:25 ashimema      😀
20:25 davidnind     I'll manually update the agenda for next time to remove it
20:25 davidnind     okay, that sounds sensible
20:24 ashimema      I think so too cait
20:24 tuxayo        cait: Indeed that was mentioned. Not sure about the outcome.
20:24 tuxayo        Lot of homework for ashimema
20:24 cait          I think oleonard had suggested to wait with the template until we got the bootstrap update in? (iirc?)
20:24 davidnind     #action ashimema - write a guideline about mandatory 'maincontent' class (OPAC)
20:23 davidnind     #action ashimema - to draft the first 'Accessibility' guideline; All OPAC pages require a single 'maincontent' classed block.
20:23 ashimema      Thanks davidnind
20:23 ashimema      I did start drafting, but got distracted..
20:23 davidnind     #action ashimema - open a bug report to introduce the idea of having a "skeleton" template we could copy for new pages
20:23 tuxayo        Thanks davidnind  I was searching the thing
20:22 davidnind     #action tuxayo - write a proposal for handing over stable branches
20:22 tuxayo        yep
20:22 davidnind     with action points being picked up by the meeting script, I guess I just add as actions again..
20:22 cait          and you wrote the action reminder thing :)
20:21 ashimema      Yikes.. I failed on both my actions
20:21 davidnind     thanks tuxayo!
20:21 cait          #info testing day turned into a 2 week thing, maybe will try a different approach next time :)
20:20 tuxayo        Late sorry, not drafted for now
20:20 tuxayo        «tuxayo write a proposal for handing over stable branches»
20:20 davidnind     as there are a few items, maybe just a quick comment from whomever is responsible
20:19 davidnind     #topic Actions from last meeting
20:18 davidnind     thanks cait and the QA team!
20:18 ashimema      +1
20:18 cait          yep, that's the thing, one bug at a time :)
20:17 davidnind     one bug at  a time :-D
20:17 tuxayo        No worries about numbers, it's the big stuff window now!
20:17 ashimema      Nice to see more names regularly on the QA leaderboards
20:17 cait          but we will keep at it
20:17 cait          at the moment the queue has a lot of bigger patches, so getting the numbers down is not so easy
20:17 ashimema      Trying to keep up cait.  But also catching up on work
20:16 cait          looking for something to add :9
20:16 cait          #info QA team is bigger, list is too! :)
20:15 davidnind     #topic Updates from the QA team
20:14 ashimema      Well done guys and gals
20:14 wahanui       darn tootin' it is.
20:14 davidnind     Excellent!
20:14 tuxayo        Thanks to various people here on the channel that helped me! :D <3
20:13 tuxayo        Joubu++
20:13 tuxayo        19.05.x: first release was done. I'm looking forward to next ones to less struggle ^^
20:13 lukeG1        First maintain release was last week for the new team. Things seem to be going well. Thanks to Joubu for rolling the 20.05.01 release for me while I was away!
20:13 davidnind     #info Congratulations to the release maintainers aleisha and tuxayo for their first releases
20:13 cait          thx :)
20:13 cait          aleisha++ tuxayo++
20:13 tcohen        yes m'am
20:13 cait          tcohen: could you add what you tested here maybe tcohen? https://lite.framacalc.org/9hdw-bug_20271
20:13 davidnind     congratulations to the release maintainers aleisha and tuxayo for their first releases!
20:12 wahanui       lukeG, aleisha, and tuxayo, I SUMMON THEE!!
20:12 tuxayo        rmaints?
20:12 ashimema      #info Martin Renvoize, PTFS-E
20:12 tcohen        there's an issue with teh atomicupdate
20:12 tcohen        Javier says hi to the keyboard
20:12 davidnind     thanks tcohen!
20:12 tcohen        d a 0bit t0oday
20:12 tcohen        I've successfully tested yesterday an
20:11 davidnind     #topic Updates from the Release Maintainers
20:11 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20271 enhancement, P1 - high, ---, jonathan.druart, Signed Off , Merge deleted biblio, biblioitems, biblio_metadata, and items tables
20:11 tcohen        he's offline
20:11 tuxayo        https://lists.koha-community.org/pipermail/koha-devel/2020-June/045810.html
20:11 tuxayo        For bug 20271, a summary is here:
20:11 davidnind     Possible a bit late for Joubu... so moving on.
20:10 davidnind     #topic Update from the Release manager (20.11)
20:10 davidnind     Thanks cait!
20:10 davidnind     #info Update on Kohacon20 https://www.catalyst.net.nz/blog/keeping-kohacon20-%E2%80%93-episode-1
20:10 cait          #info Please help testing Bug 20271  - Merge deleted biblio, biblioitems, biblio_metadata, and items tables! (also see Call for Testing on koha-devel)
20:09 davidnind     Nice update on kohacon20 is here https://www.catalyst.net.nz/blog/keeping-kohacon20-%E2%80%93-episode-1
20:09 cait          i got one
20:09 cait          oh
20:08 davidnind     Has anyone got any announcements to share?
20:08 davidnind     #topic Announcements
20:07 tuxayo        #info tuxayo/Victor Grousset, France
20:06 kidclamp      #info Nick Clemens ByWater
20:06 bag           #info Brendan Gallagher ByWater
20:05 tcohen        #info Tomas Cohen Arazi, Theke Solutions
20:05 lukeG1        #info Lucas Gass ByWater Solutions
20:05 davidnind     :)
20:05 davidnind     #info David Nind, Wellington, New Zealand
20:04 tallerjoy     #info Joy Nelson Bywater Solutions
20:04 cait          oops, too early
20:04 davidnind     #topic Introductions
20:04 cait          #info Katrin Fischer, BSZ, Germany
20:04 davidnind     #info Agenda: https://wiki.koha-community.org/wiki/Development_IRC_meeting_1_July_2020
20:04 huginn        Current chairs: ashimema cait davidnind
20:04 davidnind     #chair cait ashimema
20:03 huginn        The meeting name has been set to 'development_irc_meeting_1_july_2020'
20:03 huginn        Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:03 huginn        Meeting started Wed Jul  1 20:03:53 2020 UTC.  The chair is davidnind. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:03 davidnind     #startmeeting Development IRC meeting 1 July 2020
20:02 * tuxayo      runs to get water and food
20:00 * cait        just had espresso in hope to stay awake
20:00 ashimema      I'm coming from my phone so can't volunteer to run it tonight
20:00 wahanui       Meeting is probably over :)
20:00 ashimema      Meeting?
19:59 ashimema      Certainly had some positive affects.. got to showcase so more recent features and show off positive relationships.
19:59 cait          but i think it got lost anyway :)
19:59 * cait        tried to stick with facts...
19:59 * ashimema    doesn't like people spreading FUD so tried hard to be the friendly but informative positive spin person
19:56 ashimema      Evening
19:49 koha-jenkins  Project Koha_20.05_U18 build #18: STILL UNSTABLE in 57 min: https://jenkins.koha-community.org/job/Koha_20.05_U18/18/
19:49 koha-jenkins  Project Koha_19.11_D10 build #28: SUCCESS in 34 min: https://jenkins.koha-community.org/job/Koha_19.11_D10/28/
19:32 cait          yep there
19:31 davidnind     tcohen: https://twitter.com/librarythingtim/status/1277999574054043648
19:29 tcohen        cait where do I look for that conversation
19:12 davidnind     I personally don't mind the OPAC design - it is straight forward, functional and easy to use (maybe I've become a bit too familiar with it)
19:08 * tcohen      catches one
19:07 davidnind     and to all the Koha contributors, continually making Koha better
19:06 davidnind     sends oleonard virtual cookies/chocolate/beverage of choice
19:05 cait          i am very doubtful
19:04 davidnind     (for replying to the tweet)
19:04 davidnind     cait++
19:04 davidnind     ashimema++
18:47 cait          I lost that fight :(
18:46 oleonard      I am certainly working hard to not engage that Library Thing troll
18:45 cait          kind of, i want to avoid getting more upset and writing things
18:43 oleonard      cait does your Twitter need to be put in a time-out?
18:43 cait          can someone take my twitter?
18:20 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=25351 enhancement, P5 - low, ---, oleonard, Needs Signoff , Move cart-related strings out of opac-bottom.inc and into basket.js
18:20 oleonard      tcohen: Because no one has signed off on Bug 25351 yet :)
18:19 tcohen        how does it work on staff and not in the OPAC
18:19 tcohen        that's sad :-D
18:19 tcohen        ah
18:18 oleonard      http://git.koha-community.org/gitweb/?p=koha.git;a=commitdiff;h=6f5dc3dc0dd9fa02d6d11e92a6e466e545b8eccc#patch2
18:18 oleonard      tcohen, unfortunately the single-underscore function won't do anything because it doesn't work inside a JS file. We have to go back to passing the MSG_DT_* variables
18:16 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20936 enhancement, P5 - low, ---, agustinmoyano, Signed Off , Holds History for patrons in OPAC
18:16 tcohen        that OPAC portion of the bug was supposed to be used by bug 20936 that sunk and we went the-old-way to avoid it being blocked by the discussion about the API implementation details
18:15 tcohen        no worries, I understood the difference between _ and __ already and submitted a patch
18:15 oleonard      Sorry tcohen, I had to be afk for a bit
17:43 tcohen        https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=105922
17:43 tcohen        I think it was Joubu's follow-up
17:42 tcohen        I always forget
17:42 tcohen        is _ wrong? and the right one __?
17:42 tcohen        oleonard what is the difference between _ and __
17:31 * oleonard    resists the urge to suggest the xpm format in reply to the Webp question on the mailing list
17:29 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=25909 enhancement, P5 - low, ---, koha-bugs, NEW , Recent change to datatables JS in the OPAC causes errors
17:29 oleonard      See Bug 25909
17:27 tcohen        oleonard yes
17:15 oleonard      tcohen still around?
16:17 koha-jenkins  Project Koha_20.05_U18 build #17: UNSTABLE in 40 min: https://jenkins.koha-community.org/job/Koha_20.05_U18/17/
16:00 tuxayo        That's seems to be moving forward in the bug report. With the solution via patch and the one via a package from the Koha repo.
16:00 tuxayo        > we need coordination for a proper solution to this in the long term
15:58 tuxayo        (Ok it's fine Debian 9 that's why)
15:58 tuxayo        I'm just catching up and giving some info buried in the bug report ^^
15:54 tcohen        (i.e. so it doesn't bite us again)
15:54 tcohen        tuxayo, please communicate with mtj, we need coordination for a proper solution to this in the long term
15:52 huginn        Bug 23653: blocker, P5 - low, ---, dcook, Signed Off , Plack stopped working - JSON validation timeout
15:52 tuxayo        https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653#c16
15:52 tuxayo        Here is the test that showed that version 19.05 works:
15:49 tuxayo        Oh, maybe something else makes my 19.05 Koha on Debian work.
15:49 tuxayo        «I think anything with the Mojolicious::OpenAPI plugin will be affected»
15:45 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653 blocker, P5 - low, ---, dcook, Signed Off , Plack stopped working - JSON validation timeout
15:45 tuxayo        cait2 : bug 23653 is 19.11+ so no worries on 18.11
15:45 tuxayo        dcook: «I'm not sure which versions of Koha. I'd say definitely 19.11+ but maybe 19.05+»
15:42 koha-jenkins  Project Koha_20.05_U20 build #16: SUCCESS in 55 min: https://jenkins.koha-community.org/job/Koha_20.05_U20/16/
15:41 koha-jenkins  Project Koha_20.05_U16 build #16: FIXED in 1 hr 12 min: https://jenkins.koha-community.org/job/Koha_20.05_U16/16/
15:41 wahanui       Congratulations!
15:41 koha-jenkins  Yippee, build fixed!
15:36 koha-jenkins  Project Koha_20.05_D9 build #25: SUCCESS in 1 hr 8 min: https://jenkins.koha-community.org/job/Koha_20.05_D9/25/
15:11 reiveune      bye
14:44 nlegrand      ahlan
14:41 koha-jenkins  Project Koha_20.05_D10 build #21: SUCCESS in 56 min: https://jenkins.koha-community.org/job/Koha_20.05_D10/21/
14:11 koha-jenkins  Project Koha_20.05_D9 build #24: ABORTED in 12 min: https://jenkins.koha-community.org/job/Koha_20.05_D9/24/
13:42 mtj           swagger.io is back up too -> http://swagger.io/v2/schema.json
13:39 mtj           tests pending... but it looks promising
13:39 tcohen        mtj++
13:39 mtj           hmm, i think ive found a combo of json-validator and libmojolicious-plugin-openapi that works for all versions
13:30 koha-jenkins  Project Koha_20.05_U20 build #15: FIXED in 49 min: https://jenkins.koha-community.org/job/Koha_20.05_U20/15/
13:30 wahanui       Congratulations!
13:30 koha-jenkins  Yippee, build fixed!
13:21 koha-jenkins  Project Koha_20.05_U18 build #16: FIXED in 57 min: https://jenkins.koha-community.org/job/Koha_20.05_U18/16/
13:21 wahanui       Congratulations!
13:21 koha-jenkins  Yippee, build fixed!
13:10 mtj           hmm, let me test a few ideas...
13:08 mtj           i think short answer is no - that wont work for u20 and u19, they will use higer pkg versions
13:06 tcohen        can't we keep the older library version, but with the cached file?
13:05 mtj           i think 'uniq' was added to libscalar-list-utils-perl in 1.47 - and u16 has 1.43 :'(
13:04 mtj           hmm, im not quite sure what you mean tcohen?
13:00 tcohen        doesn't Ubuntu 16 rely on a different repo schema?
12:56 mtj           ...i havent quite got to a fix for it yet :/
12:54 mtj           https://jenkins.koha-community.org/view/20.05/job/Koha_20.05_U16/15/console
12:54 mtj           00:04:19.051 koha_1       | #     Error:  "uniq" is not exported by the List::Util module
12:54 mtj           00:04:19.051 koha_1       | #     Tried to use 'Koha::REST::V1'.
12:54 mtj           hmm, i think its because u16's perl version is older, and causes a problem..
12:52 tcohen        because it is a newer version?
12:52 mtj           it looks like it might work for ubuntu 20,19,18 - but breaks on ubuntu 16
12:51 tcohen        mtj cool
12:51 mtj           hi tcohen, ive been testing a new package for 23653
12:49 tcohen        cait2 any doubts about my position on that field name?
12:47 tcohen        I mistakenly thought they had +o perms on the channel
12:43 wahanui       i guess aaru_swartz is facing an issue in uploading bulk data in Koha
12:43 tcohen        who's aaru_swartz?
12:41 koha-jenkins  Project Koha_20.05_U16 build #15: STILL FAILING in 4 min 39 sec: https://jenkins.koha-community.org/job/Koha_20.05_U16/15/
12:40 koha-jenkins  Project Koha_20.05_D9 build #23: FIXED in 43 min: https://jenkins.koha-community.org/job/Koha_20.05_D9/23/
12:40 wahanui       Congratulations!
12:40 koha-jenkins  Yippee, build fixed!
12:22 koha-jenkins  Project Koha_20.05_D10 build #20: FIXED in 44 min: https://jenkins.koha-community.org/job/Koha_20.05_D10/20/
12:22 wahanui       Congratulations!
12:22 koha-jenkins  Yippee, build fixed!
12:10 koha-jenkins  Project Koha_20.05_U20 build #14: FAILURE in 1 min 42 sec: https://jenkins.koha-community.org/job/Koha_20.05_U20/14/
12:07 koha-jenkins  Project Koha_20.05_U16 build #14: ABORTED in 1 min 42 sec: https://jenkins.koha-community.org/job/Koha_20.05_U16/14/
12:05 tcohen        morning
11:55 khall         cait2 what's the bug number?
11:55 cait2         khall: could yougive your opinion on the pathological holds patch maybe?
11:54 khall         oleonard I'll prolly be working on that one too, but I need to talk to talljoy about it
11:54 huginn        Bug 25907: minor, P5 - low, ---, kyle, Needs Signoff , When cancelling a waiting hold on returns.pl, looks for new hold to fill without rescanning barcode
11:54 khall         oleonard https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=25907
11:48 oleonard      It's too bad it wasn't just right for 4948 because fixing a 10-year-old bug with one line is pretty cool 8)
11:47 khall         oleonard I'll take care of that right now then!
11:47 oleonard      I've got my signoff ready :)
11:47 oleonard      I guess a new bug would be appropriate
11:46 khall         oleonard the patch I just submitted doesn't quite fit the description. I'm considering moving it to a separate bug. What do you think? This one operates on the checkin page, not the holds over page
11:44 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=4948 enhancement, P5 - low, ---, kyle.m.hall, NEW , When cancelling a waiting hold, allow to trigger hold for next patron
11:44 oleonard      khall should Bug 4948 be "needs signoff?"
11:42 oleonard      khall++
11:35 * oleonard    needs to stop
11:35 oleonard      librarytechnology.org lists 5 libraries using LibraryThing, so we do have some catching up to do.
11:12 Null404       other Holds seems to work: Holds to pull, Holds awaiting pickup for your library and Hold ratios
11:11 koha-jenkins  Project Koha_19.11_D8 build #196: UNSTABLE in 1 hr 30 min: https://jenkins.koha-community.org/job/Koha_19.11_D8/196/
11:11 Null404       Hi, I got strange bug on Koha 18.11.16.000 - Hold queue won't updates properly
10:58 oleonard      [off] Someone should reply by asking for a one-tweet explanation for why LibraryThing lost so badly to GoodReads
10:57 eythian       Wow, librarything. Turns out I last logged into that in 2007.
10:56 oleonard      [off] And they've never submitted a patch to LibraryThing integration. Clearly not interested in engaging with the Open Source process
10:52 oleonard      [off] And if he's complaining about designs looking dated... LibraryThing has redesigned never.
10:52 ashimema      Fair points
10:52 ashimema      Those that are really keen just aren't using the opac in my experience.. they bolt WordPress or discovery system on top instead.
10:51 ashimema      As for looking dated.. people round here have been on about redesigning to opac for ages but I've really not seen a clear way to do it which kept allot of the features people are using
10:51 oleonard      I don't think it's worth engaging with Tim Spalding because he seems to have some kind of grudge
10:51 koha-jenkins  Project Koha_19.11_U18 build #191: STILL UNSTABLE in 1 hr 33 min: https://jenkins.koha-community.org/job/Koha_19.11_U18/191/
10:51 ashimema      Yeah, I've noticed that
10:50 oleonard      ashimema: Library Thing has had it in for Koha for a long time, don't know why
10:21 eythian       Anything else is necessarily going to be incremental.
10:21 eythian       ashimema: re: looking dated, I think the only way to avoid that is to schedule a complete redesign/rewrite every three years to keep up with current fashions. Which isn't a bad idea, but likely much more work than it's worth.
10:01 koha-jenkins  Project Koha_19.11_D10 build #27: SUCCESS in 44 min: https://jenkins.koha-community.org/job/Koha_19.11_D10/27/
09:59 sophie_m      wrong chan :-[
09:59 wahanui       i already had it that way, huginn.
09:59 huginn        sophie_m: I'll give you the answer just as soon as RDA is ready
09:59 sophie_m      @paul_p ton message sur koha-info va semer la panique, chez nous seuls les clients sous bionic sont concernés.
09:51 davidnind     I saw this this morning, but wasn't sure how to respond... I see you've jumped in!
09:49 ashimema      Anyone fancy joining that discussion.. I'm always interested to see Koha from different view points and would love to see a good debate
09:49 ashimema      https://twitter.com/librarythingtim/status/1277999574054043648
09:40 koha-jenkins  Project Koha_19.11_D9 build #194: FIXED in 51 min: https://jenkins.koha-community.org/job/Koha_19.11_D9/194/
09:40 wahanui       Congratulations!
09:40 koha-jenkins  Yippee, build fixed!
08:53 huginn        Bug 25906: enhancement, P5 - low, ---, oleonard, NEW , In opac-serial-issues.pl, when looking at full history, the search facet form should look good.
08:53 TimothyAlexis https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=25906
08:48 koha-jenkins  Project Koha_20.05_D9 build #22: UNSTABLE in 40 min: https://jenkins.koha-community.org/job/Koha_20.05_D9/22/
08:40 koha-jenkins  Project Koha_20.05_D10 build #19: STILL UNSTABLE in 44 min: https://jenkins.koha-community.org/job/Koha_20.05_D10/19/
08:36 TimothyAlexis I will report it as a bug.
08:23 TimothyAlexis http://www.freealization.com/files/how-it-should-look.png
08:23 TimothyAlexis http://www.freealization.com/files/what-it-looks-like.png
08:23 TimothyAlexis In opac-serial-issues.pl, when looking at full history, the search facets look strange. Is there a patch for this?
08:19 fridolin      oups wrong chan ;)
08:19 fridolin      https://wiki.biblibre.com/pole:maintenance:suivi:cr:2020-07-01:biblibre chris_n
08:16 mtj           hi cait, yes.. 0.97 is ok :)
08:16 cait          ok, I think i found the answer: confirmed libjson-validator-perl (0.97-1~kohadev1) has the cache dir  = so we are good
08:15 cait          mtj maybe?
08:14 TimothyAlexis Hello. :)
08:10 cait          sorry to ask again... but checkout about.pl i see we have 0.97 - later than listed on the bug. still no need to worry?
08:06 cait          thx dcook - latest comment clears it up :)
08:05 koha-jenkins  Project Koha_20.05_U18 build #15: UNSTABLE in 57 min: https://jenkins.koha-community.org/job/Koha_20.05_U18/15/
08:02 dcook         Wishing everyone luck with their day
08:02 dcook         Anyway, I think it's time for me to call it a night
08:02 dcook         Although that's when oleonard mentioned it I think..
08:02 dcook         Or maybe I was imagining the issue in koha-testing-docker..
08:01 * dcook       scratches his head
08:01 dcook         Hmm or not..
08:01 dcook         Oh probably because it uses the staging repo..
08:01 dcook         But it had the problem...
08:01 dcook         koha-testing-docker is Debian Stretch 9
08:00 dcook         Although now that I think abotu it...
08:00 dcook         Hmm
08:00 dcook         Not that I have many of those left anymore
08:00 dcook         And my non-Debian/Ubuntu instances are fine
07:59 dcook         But all my Debian Jessie instances are fine
07:59 dcook         I have a lot of instances on Ubuntu 18.04 which is why we got hit by it so much
07:58 koha-jenkins  Project Koha_20.05_U20 build #13: SUCCESS in 38 min: https://jenkins.koha-community.org/job/Koha_20.05_U20/13/
07:58 dcook         Just remembered Debian Jessie 8 was EOL yesterday
07:58 dcook         cait: Ok I hope comment 40 is better
07:56 dcook         I'll add a follow-up comment
07:56 dcook         You should be fine on Debian Stretch
07:56 cait          hm got disconnected sorry (vpn)
07:53 cait1         so... if we are building our own packageson debian stretch... what am I going to do?
07:53 wahanui       rumour has it help is appreciated for bugs and keeping jenkins happy
07:53 cait1         help?
07:52 cait1         i read it three tiems and i still not get it :)
07:52 cait1         i have troubel reading comment 39 on the bug
07:51 dcook         Just confirmed a Debian Jessie library is up and all good
07:51 koha-jenkins  Project Koha_20.05_D10 build #18: ABORTED in 4 min 43 sec: https://jenkins.koha-community.org/job/Koha_20.05_D10/18/
07:51 koha-jenkins  Project Koha_20.05_D9 build #21: ABORTED in 3 min 45 sec: https://jenkins.koha-community.org/job/Koha_20.05_D9/21/
07:47 koha-jenkins  Project Koha_20.05_D9 build #20: ABORTED in 8 min 56 sec: https://jenkins.koha-community.org/job/Koha_20.05_D9/20/
07:45 dcook         I was going to go home tonight to work on mobile games for my kid, but maybe no more coding for me tonight lol
07:44 ashimema      :)
07:40 dcook         Although kind of glad that it is to expose this problem at least
07:40 dcook         Yeah I wonder why it is still offline too
07:40 huginn        Bug 23653: blocker, P5 - low, ---, dcook, Signed Off , Plack stopped working - JSON validation timeout
07:40 dcook         cait2: This table should be useful: https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653#c39
07:39 cait2         that was bad decsision number one... but also wondering why the file is still offline :(
07:37 * dcook       goes back to making his table instead of letting his mind be blown on the definition of cache
07:37 dcook         Actually probably the latter
07:37 dcook         Or maybe I've been misunderstanding the definition all this time...
07:36 dcook         Even if I'd argue "cache" is a bit of a misnomer in this case heh
07:36 dcook         Yeah no kidding
07:36 ashimema      I think I was the one that argued for shipping with a cached version of that file in the first place over in the openapi camp.. the is just proves how worthwhile that was
07:34 dcook         Yeah I think I saw the Bugzilla report a week ago and thought nothing of it as it hadn't hit us
07:34 ashimema      Wasn't aware Debian had removed the cached file
07:34 dcook         But took a while to completely turn the code inside out
07:34 ashimema      Thanks for jumped Ng on that one.. it's one I was aware of in the periphery but turned out way worse than I thought
07:33 dcook         Had all my servers fixed by 9:30am
07:33 cait          mtj++ and dcook++ btw
07:33 dcook         Oh yeah lol
07:33 ashimema      Sounds like you had a busy day dcook
07:33 dcook         heyo ashimema
07:33 ashimema      Mornin'
07:33 dcook         I'll add a little summary to bugzilla
07:32 dcook         And Ubuntu 18.04+ for sure
07:32 cait1         I guess we will get asked a fair bit today, so I can help
07:31 dcook         Buster
07:31 cait1         so which is the broken one?
07:31 dcook         As Stretch comes with 0.92
07:31 dcook         https://packages.debian.org/stretch/libjson-validator-perl
07:31 dcook         Stretch should be ok too
07:31 cait1         Debian 4.9.210 on about page... stretch?
07:30 dcook         Confirmed it too just now
07:30 dcook         Yep
07:30 cait1         thx for your help dcook and mtj
07:30 mtj           confirmed libjson-validator-perl (0.97-1~kohadev1) has the cache dir
07:29 dcook         I guess one way to make sure..
07:29 mtj           yep ^
07:28 cait1         so 18.11 on jessie should be ok?
07:28 dcook         Yeah just noticed the kohadev1. That's promising
07:28 mtj           ^ that was built from cpan, so has cache dir (i think)
07:28 dcook         That's interesting
07:27 dcook         Oh is that one OK?
07:27 dcook         Yeah 18.11 should be vulnerable
07:27 cait1         older Koha yes, but our Debian is not so old i think
07:27 mtj           if you are on 18.11 with jessie , you will have libjson-validator-perl (0.97-1~kohadev1)
07:27 dcook         I think I'll do a quick check now to see if 18.11 will be bitten by it
07:26 mtj           cait1: id say you are running an older koha and/or debian - so you havent hit the problem
07:25 dcook         cait1: That's a tough one to troubleshoot remotely :/
07:24 dcook         Didn't realize koha-testing-docker used the staging repo
07:24 dcook         mtj: Oh oops
07:24 mtj           dcook: hiya, just testing 3.23 on the staging repo...
07:23 cait1         so at the moment i am a little puzzled
07:23 cait1         the issue is, my coworker says there is no JSON error in the logs
07:22 dcook         cait1: ^
07:18 dcook         So yeah you'd probably be fine doing an "apt-get update && apt-get install libjson-validator-perl"
07:18 dcook         I can see 3.23+dfsg-1+koha1 available
07:17 dcook         Oh I think he has 3.23 up now too
07:17 dcook         Yeah, I thought that he'd updated Bugzilla. He has it on IRC further up.
07:17 cait1         it woudl be good to communicate that
07:16 dcook         Version 3.06+dfsg-1+koha1 that is
07:16 dcook         I tested it on Debian Jessie and it worked
07:16 dcook         It was but then mtj found some dependency issues for newer versions of Ubuntu, so had to redo it with a newer version
07:13 cait1         is this fix working? https://lists.koha-community.org/pipermail/koha-devel/2020-July/045842.html
07:13 cait1         I don't have server access so have to forward the information
07:12 cait1         i thought the package was done?
07:12 dcook         Ok it's a lot quieter than I thought it would be
07:05 * dcook       might hang around for a bit to help folk even though it's home time
07:05 dcook         And mtj is almost done preparing a new libjson-validator-perl package which will also resolve the problem
07:04 dcook         cait1: I also have a patch that can fix the problem available on that Bugzilla report
07:04 dcook         although you'd have to manually create the /usr/share/perl5/JSON/Validator/cache directory
07:04 dcook         That would be the fastest way to do it
07:04 dcook         You could copy https://raw.githubusercontent.com/OAI/OpenAPI-Specification/master/schemas/v2.0/schema.json to /usr/share/perl5/JSON/Validator/cache/36d1bd12eeed51e86c8695bd8876a9df
07:03 dcook         The relief for now is to create this file: /usr/share/perl5/JSON/Validator/cache/36d1bd12eeed51e86c8695bd8876a9df
07:03 dcook         Oh we have lots of solutions already in the works heh
07:03 cait1         we will have tof ix the broken link, but that woudl at least give some relief
07:02 dcook         yeah tuxayo reported it to them
07:02 cait1         strange  question... but has someone tried to get them to fix that?
07:02 dcook         If you're able to pop on, there is a quick workaround that can save you
07:02 cait1         and we are in the middle of an update on one of our servers too :(
07:02 dcook         I'm guessing unintentional since they're still showing v2 docs at https://swagger.io/docs/specification/2-0/basic-structure/
07:02 dcook         Still no idea whether the unavailability is intentional or not
07:01 dcook         Due to http://swagger.io/v2/schema.json being unavailable currently
07:01 dcook         The Debian team excluded some files which turned out to be quite essential
07:01 dcook         ^
07:01 dcook         That uses the Debian package for libjson-validator-perl
07:01 mtj           cait1: the plack bug depends on which version of libjson-validator-perl you have
07:00 dcook         So anything running the REST API I would think
07:00 dcook         I think anything with the Mojolicious::OpenAPI plugin will be affected
07:00 dcook         Hmm probably still affected
07:00 cait1         we are 18.11
06:59 dcook         Yep makes sense
06:59 cait1         so ... they have been busy earlier than us
06:59 huginn        Bug 23653: blocker, P5 - low, ---, tomascohen, Signed Off , Plack stopped working - JSON validation timeout
06:59 cait1         yea the first report is from jakarta
06:59 dcook         cait1: One of the easiest solutions is here https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653#c11
06:59 dcook         mtj: heh
06:59 mtj           dcook: ^ thanks, its makes a nice change from breaking things :p
06:59 dcook         And you'll notice a loop of failures in your plack-error.log
06:58 dcook         But as new ones are created, they'll fail
06:58 dcook         Because it's early in the day for you, you may still have some long-lived Plack processes going
06:58 dcook         I'd say chances are you're affected, cait1
06:58 alex_a        Bonjour
06:58 dcook         I'm not sure which versions of Koha. I'd say definitely 19.11+ but maybe 19.05+
06:58 cait1         we are on Debian
06:58 dcook         I didn't have the issue on OpenSUSE as we're using a CPAN module there which doesn't have the problem
06:57 dcook         Debian/Ubuntu for sure
06:57 cait1         is there any rule on which versions / OS are affected?
06:57 cait1         so far we have one report for one instance on a server with others that is slow...
06:57 dcook         Sure
06:56 cait1         can someone help me on how to diagnose the Plack issue?
06:55 cait1         morning all
06:50 koha-jenkins  Project Koha_20.05_U16 build #13: FAILURE in 3 min 24 sec: https://jenkins.koha-community.org/job/Koha_20.05_U16/13/
06:44 dcook         Or Days because of timezones heh
06:44 dcook         mtj is certainly the Koha hero of the day I reckon
06:44 dcook         mtj: I was wondering where you were at. That's good to know.
06:36 mtj           i have built a json-validator-3.23 ready - just testing it on jenkins before golive
06:29 dcook         If Europeans aren't experiencing problems yet, I imagine they will soon
06:28 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653 blocker, P5 - low, ---, tomascohen, Signed Off , Plack stopped working - JSON validation timeout
06:28 * dcook       refers people to Bug 23653
06:28 dcook         heya reiveune
06:28 huginn        dcook: The operation succeeded.
06:28 dcook         @later tell aleisha Hmm no idea
06:19 reiveune      hello
05:33 aleisha       im heading home now but will come to this tomorrow :)
05:32 aleisha       im not see the errors that others have seen with strict_sql_modes enabled
05:31 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=24488 enhancement, P5 - low, ---, aleisha, ASSIGNED , Holds to Pull sometimes shows the wrong 'first patron' details
05:31 aleisha       bug 24488
05:30 dcook         You were expecting errors?
05:29 aleisha       strangely no errors :/
05:29 dcook         Ahh that might explain why I get different results from the mysql CLI and Koha...
05:29 aleisha       yes i restarted it and mysqld
05:29 dcook         That'll bite you every time O_O
05:28 dcook         aleisha: After you updated etc/koha-conf.xml, did you clear your Memcached?
05:28 dcook         Ah I think I know
05:27 dcook         aleisha: what are you trying to test?
05:08 aleisha       is there something else i need to do?
05:07 aleisha       i've enabled the flag in etc/koha-conf.xml but that doesn't seem to have changed anything
05:07 aleisha       hi can someone please tell me how to enable strict_sql_modes on my dev environment?
04:38 dcook         Thanks :D
04:32 mtj           yep, will do ^
04:31 dcook         At the moment, I'm saying the updated package is all good
04:31 huginn        Bug 23653: blocker, P5 - low, ---, tomascohen, Signed Off , Plack stopped working - JSON validation timeout
04:31 dcook         mtj: Can you update https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653 with your notes?
04:31 dcook         mtj++ I'd have pulled out hair if it weren't for your workaround this morning
04:30 dcook         Glad that I'm so busy I hadn't tried deploying it too widely
04:30 dcook         Yikes! Thanks for the heads up
04:29 mtj           dcook++ many thanks for the swagger patch
04:27 mtj           js-val 3.25 has some problematic deps, like libsereal-encoder-perl (>= 4.00)
04:27 mtj           i might have to use libjson-validator-perl 3.23, over 3.06
04:23 mtj           .. so they choose those versions over the 3.06+koha patched version :/
04:22 mtj           hmm, ive clicked that both ubuntu eoan and focal have higer versions of libjson-validator-perl than 3.06
04:02 dcook         Ah yep. It was just messed up
04:01 dcook         Or maybe my test instance is just a bit messed up..
04:01 dcook         Interesting... it has a number of additional dependencies
03:58 dcook         Oh it looks like Mason has updated the package
03:56 dcook         I mean I guess the Intranet and maybe the OPAC increasingly depend on the API so maybe that's not a bad thing per se...
03:56 dcook         Hmm you know something else to keep in mind is that the API brought down the whole of Koha...
03:51 dcook         Too right
03:36 tuxayo        muahahaha
03:35 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653 blocker, P5 - low, ---, tomascohen, Signed Off , Plack stopped working - JSON validation timeout
03:35 wahanui       bug 23653 is the Swagger-apocalypse!!!!! https://i.ytimg.com/vi/m0h7ESBEf3U/hqdefault.jpg
03:35 tuxayo        bug 23653
03:07 dcook         bbiab
03:06 dcook         Speaking of sinking my teeth into things, I'm going to go get some lunch O_O
03:06 dcook         Thanks. I'm like a bulldog with problems I guess. Have to understand and fix it >_>
03:02 tuxayo        Well done dcook o/
02:59 dcook         Cheers :)
02:49 tuxayo        Quite small patch indeed!
02:48 tuxayo        I'm signing off while I'm at it.
02:47 tuxayo        ok, reading the patch
02:47 tuxayo        Indeed, but I guess it's for the same reasons of the current issue required to run the tests.
02:47 tuxayo        > I don't think we refer to swagger.io in the tests?
02:45 dcook         tuxayo: Check out the patch if you have a few minutes. It wound up being a lot easier than I anticipated
02:45 dcook         Anyway, probably time for me to get some lunch and decompress O_O
02:44 dcook         I don't think we refer to swagger.io in the tests? At least I hope not
02:44 dcook         I still am amazed we managed to cooperate so well
02:44 dcook         Yeah, it was pretty amazing
02:43 tuxayo        congrats to whole chain that it worked
02:43 dcook         That was a special day for me lol
02:43 tuxayo        wooooooooooooow
02:43 tuxayo        Until know I only though the dependence on swagger.io was for tests. But nope ^^"
02:43 dcook         They had some networking troubles, but weren't Internet connected, so I remotely instructed them on how to use the CLI in English... where the person working was an English speaker with a Russian translator and the Debian was in Russian... heh
02:42 tuxayo        I didn't get that ^^"
02:42 tuxayo        > I guess a lot of people did just pay their bills in June so that's handy
02:42 tuxayo        wow :D
02:42 tuxayo        Yeah, WiFi with directional antennas can do that
02:42 dcook         That one is in Kyrgyzstan
02:42 dcook         I think it's provided using a microwave beam to a tower several kilometres away
02:41 dcook         For instance, we helped one school that is physically quite distant from any Internet service
02:41 dcook         >_>
02:41 dcook         I guess a lot of people did just pay their bills in June so that's handy
02:41 tuxayo        I heard about a few of them also
02:41 tuxayo        > non-Internet connected Koha installations
02:41 dcook         Well, it's winter here and the end of the financial year heh
02:40 tuxayo        It's a good moment to happen in summer ^^
02:40 tuxayo        yep
02:40 tuxayo        > grateful for it in a way, since it exposes this fairly unacceptable reliance on the outside world
02:40 dcook         No reason for Koha to require the Internet
02:40 dcook         I've set up a number of non-Internet connected Koha installations over the years
02:40 dcook         Although I'm kind of grateful for it in a way, since it exposes this fairly unacceptable reliance on the outside world
02:39 tuxayo        done
02:39 dcook         Oh yeah that would be a good one heh
02:37 tuxayo        Doing that now
02:37 tuxayo        Wait, I forgot to report to swagger.io that there schema hosting is down! :o
02:37 tuxayo        he he
02:37 tuxayo        > The other pro is mine is available in about 10 seconds from now heh
02:37 tuxayo        yep
02:32 dcook         I'm happy with either a Debian dependency fix or this code fix. Figure it's nice to have options.
02:32 huginn        Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653 blocker, P5 - low, ---, tomascohen, Needs Signoff , Plack stopped working - JSON validation timeout
02:32 dcook         Bug 23653
02:27 dcook         The other pro is mine is available in about 10 seconds from now heh
02:24 dcook         I mean arguably a code change would be too, although it would be explicit - rather than hidden
02:23 dcook         Although then that's another thing for us to maintain forever, if Debian doesn't change its package
02:21 tuxayo        Indeed, that way, deployers only have to update the perl-deps, no patch to apply.
02:21 tuxayo        > the best solution is probably to do a new package that includes the JSON cache
02:20 tuxayo        dcook++
02:20 dcook         But figure I may as well include this code fix
02:20 dcook         On Bugzilla, tcohen said the best solution is probably to do a new package that includes the JSON cache. That's probably true.
02:15 dcook         Woohoo, created a code fix
02:14 tuxayo        And I just realized that I have a Debian running on my computer: my Koha dev env! :D
02:13 tuxayo        I'm fighting since too much time with the packages on EndeavourOS (Arch Linux actually) (I even have to install apt!).
02:13 tuxayo        I'm trying to report a bug to the Debian package bug tracker but it's very recommended to use their "reportbug" program for that.
02:11 dcook         Ahhh because it doesn't have the validator..
02:11 dcook         Hmm looks like maybe I was wrong about PluginRoutes.pm being the problem..
01:49 dcook         tcohen: Looks like you're awake. Are you on IRC too?
01:40 dcook         Great :)
01:40 tuxayo        Good news, PluginRoutes.pm has seen little changes between 19.11, 20.05 and master
01:34 dcook         Due to the way Koha/REST/Plugin/PluginRoutes.pm is written there is a lot of refactoring to do..
01:34 dcook         Well "simple" from a sysadmin point of view
01:32 tuxayo        ^^
01:32 dcook         Actually yeah... yeah
01:32 dcook         Well...
01:32 dcook         tuxayo: Looks like I'll have a very simple fix coming soon I think
01:30 dcook         Which means it only affects plugin-enabled instances
01:20 dcook         Bundle is working in /usr/share/koha/lib/Koha/REST/V1.pm, although it's fallback would fail..
01:20 dcook         It's /usr/share/koha/lib/Koha/REST/Plugin/PluginRoutes.pm
01:19 dcook         Ahh I see it now..
01:19 dcook         Unless..
01:12 dcook         So if there are failures it's beacuse bundle isn't working..
01:12 dcook         It looks like $validator->bundle is using a massively simpler schema file
01:11 dcook         Interesting...
01:09 tuxayo        👌
01:09 dcook         Just trying to focus
01:09 dcook         Still around heh
01:05 tuxayo        (actually I think there are 1/2 chances that it's not that XD)
01:05 tuxayo        See you, good luck.
01:05 tuxayo        IIUC, that means you are leaving o/
01:05 tuxayo        «Cheers, tuxayo»
01:03 * tuxayo      tries the find the Debian issue tracker for the package
01:02 tuxayo        Indeed lol, one line diff
01:02 dcook         I'll go with this latest one you've provided
01:02 dcook         Interestingly there is a 1 line difference between them
01:02 dcook         Cheers, tuxayo
01:02 tuxayo        *https://github.com/OAI/OpenAPI-Specification/blob/master/LICENSE
01:02 tuxayo        So under the Apache 2 license of the repo Apache License 2.0
01:01 tuxayo        dcook: the schema file is also here: https://github.com/OAI/OpenAPI-Specification/blob/master/schemas/v2.0/schema.json
00:58 tuxayo        Good find, maybe
00:55 dcook         Hmm tcohen pointed out the licence at https://github.com/OAI/OpenAPI-Specification/blob/master/LICENSE
00:55 dcook         I think this will probably affect anyone using the API but I haven't checked
00:53 tuxayo        Hopefully they aren't on >= Koha 19.11 yet
00:53 tuxayo        > And installs that don't have reactive/proactive support
00:52 dcook         Anyway, just about to start on this patch
00:52 tuxayo        Ok I get it now. Good nothing worse then ^^"
00:51 dcook         And installs that don't have reactive/proactive support
00:51 dcook         But any new installs will have the problem
00:51 dcook         Well, it'll fix the problem for anyone that does it
00:50 tuxayo        You mean the current workaround puts a lot of people at risk?
00:50 tuxayo        > But that puts a lot of people at risk still
00:43 dcook         I definitely need some sugar/caffeine heh
00:43 dcook         Because JSON::Validator just has a copy of it
00:43 dcook         Silly me
00:43 dcook         My mistake
00:43 dcook         Ah right
00:43 tuxayo        JSON::Validator can't decide the license of this file. It's from Swagger
00:43 dcook         So I'm going to do the workaround on... two more servers and then I'll write a patch
00:43 tuxayo        Oh, it's 3 level! (Swagger, JSON::Validator, Debian)
00:43 tuxayo        > Right, but swagger.io isn't part of that
00:43 dcook         But that puts a lot of people at risk still
00:42 dcook         Of course, the current workaround will solve this problem "permanently" for anyone who puts it in
00:42 dcook         Excellent point
00:42 dcook         Good point
00:42 dcook         ^
00:42 tuxayo        IIUC only Debian/Ubuntu users would have this issue. Well, other distros could have found this lack of license and exclude this file
00:42 dcook         As that's the source of that cache
00:41 dcook         https://github.com/mojolicious/json-validator
00:41 tuxayo        > The only issue will be for people not using Debian packages
00:41 tuxayo        > Using the Debian packages we could automate it fairly easily
00:41 dcook         Or wait I alrady shared that other github..
00:41 dcook         Maybe it's this one..https://metacpan.org/pod/JSON::Validator::OpenAPI::Mojolicious
00:41 dcook         Wait not that one..
00:41 dcook         We probably want to open an issue at https://metacpan.org/pod/JSON::Validator
00:41 dcook         Right, but swagger.io isn't part of that
00:40 tuxayo        *cause of
00:40 tuxayo        It's the cause Debian not shipping it
00:40 dcook         Jan Henning Thorsen
00:40 dcook         It's made by...
00:40 dcook         The JSON/Validator/cache isn't made by them
00:40 tuxayo        lib/JSON/Validator directory, where the license is unknown.
00:40 tuxayo        Comment: The +dfsg version is created by removing the cache files from the
00:40 tuxayo        https://salsa.debian.org/perl-team/modules/packages/libjson-validator-perl/-/raw/master/debian/copyright
00:40 dcook         Yeah but you raised an issue against the swagger.io website
00:39 huginn        Bug 23653: blocker, P5 - low, ---, tomascohen, In Discussion , Plack stopped working - JSON validation timeout
00:39 dcook         Well, we can code it to work for them too if they do a proper upgrade..
00:39 tuxayo        dcook: https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653#c13
00:39 dcook         But I guess that's the downside of them not using them..
00:39 dcook         The only issue will be for people not using Debian packages
00:39 dcook         Using the Debian packages we could automate it fairly easily
00:39 tuxayo        dcook: If there is a way to have it directly from a file without changing much stuff that would be great
00:38 dcook         Hmm tuxayo, what made you think to raise the issue at https://github.com/swagger-api/swagger.io/issues/281?
00:35 dcook         I don't love that kind of dependency either
00:35 dcook         :/
00:35 dcook         So it would be fetching it from Apache
00:35 dcook         Plack fetches this file at startup
00:35 dcook         I guess it would
00:35 dcook         Well...
00:34 dcook         Actually that won't work either
00:34 dcook         Wait...
00:34 dcook         Unless I guess if you can rely on the base url being set..
00:34 dcook         Although there would still be some server config
00:34 dcook         I suppose there's no harm in exposing it to end users...
00:33 dcook         Oh right.. yeah I see what you're saying
00:33 dcook         Oh interesting... JSON_VALIDATOR_CACHE_PATH environmental variable..
00:32 dcook         This "cache" isn't a real cache
00:31 dcook         oleonard: On the plus side, this workaround is sort of a permanent workaround
00:31 dcook         JSON::Validator can read it as a local file
00:31 dcook         We don't need to load it over HTTP
00:31 dcook         It's a backend asset. Doesn't need to be shown to end users.
00:31 dcook         Why woudl you put it in the opac-tmpl path?
00:27 rangi         anywhere in the opac-tmpl path would do tho
00:27 rangi         thats where i would put it
00:26 tuxayo        Is it okay with our JS and CSS assets? It ensures it's reachable without server config.
00:26 tuxayo        dcook: «Where to make this file live»
00:25 tuxayo        dcook: it's just a check. This page isn't 100% JSON
00:24 dcook         Although it's still an online option
00:24 dcook         I suppose that's an option yeah
00:24 tuxayo        So it's good
00:24 tuxayo        https://web.archive.org/web/20181127092010/http://23.22.16.221/v2/schema.json
00:24 tuxayo        dcook: What you have is perfect match to this archive
00:24 rangi         the l is too long :)
00:23 rangi         its the problem with the lts releases
00:23 dcook         Where to make this file live
00:23 dcook         So now quick question for people
00:23 dcook         Yeah, I was being imprecise
00:23 rangi         for that module
00:23 rangi         cos sid now is 4.0x something
00:23 dcook         True
00:23 dcook         heh
00:23 dcook         Yeah buster before buster was buster
00:23 rangi         so you can't really say that, unless you say sid as of april 2018 :)
00:22 dcook         Whoa haven't seen rangi here in ages O_O
00:22 rangi         and testing
00:22 rangi         buster/sid is unstable
00:20 dcook         So I'll use that for now
00:20 dcook         https://github.com/mojolicious/json-validator/blob/3.06/lib/JSON/Validator/cache/36d1bd12eeed51e86c8695bd8876a9df
00:19 tuxayo        https://askubuntu.com/questions/445487/what-debian-version-are-the-different-ubuntu-versions-based-on
00:19 tuxayo        Indeed
00:19 tuxayo        > 18.04 which is based off buster
00:18 dcook         Well I think we know 3.06 is working right now so I'll look at including that version
00:18 dcook         Hmm..
00:17 dcook         buster/sid that is
00:17 dcook         libjson-validator-perl is 1.08 on 18.04 which is based off buster I think..
00:17 dcook         Hmm interesting
00:17 mtj           hi dcook, 3.06 is the version available for buster and stretch-backports, other than that - there is nothing special about the version
00:16 dcook         In Ubuntu, I'm seeing version 1.08
00:15 dcook         mtj: In your workaround, how did you know to use 3.06?
00:13 dcook         Ok I think I've worked out the logic. Now just to do a patch and test it...
00:10 dcook         I know I got it in a lot of production boxes today
00:10 oleonard      tuxayo just in my dev machine
00:09 tuxayo        oleonard: so you got that in production without restarting plack?
00:09 dcook         I've almost worked this out..
00:09 dcook         Right?
00:09 tuxayo        And now it just 504 timeout!!!!
00:09 tuxayo        No way, it's real. I just opened more than 100 pages on a demo instance (2 workers, 50 pages for each if my spies are right)
00:09 tuxayo        > Got to love a ticking time bomb
00:08 oleonard      Thanks y'all, that workaround works great
00:07 tuxayo        Yes!
00:07 dcook         By embedding the schema.json in Koha
00:07 dcook         I'm hoping within the next 60 minutes to have a more permanent solution
00:06 hayley        yes very stressful!
00:06 hayley        done!
00:06 dcook         I think I aged 10 years in an hour this morning :p
00:06 tuxayo        +1
00:06 dcook         hayley: Oh yeah I would
00:06 hayley        thoughts on moving it to critical/blocker?
00:06 dcook         mtj++
00:05 dcook         Comment 11 specifically
00:05 dcook         That works for the time being
00:05 huginn        Bug 23653: major, P5 - low, ---, tomascohen, In Discussion , Plack stopped working - JSON validation timeout
00:05 dcook         oleonard: https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23653#c11
00:05 tuxayo        Indeed
00:05 tuxayo        «dcook just thinks it's crazy to be forced to load data from a third-party URL»
00:05 dcook         oleonard: Yep
00:04 dcook         The schema argument goes to Mojo::JSON::Pointer, which I bet does some auto magic..
00:04 oleonard      I'm just jumping back in after taking a break... Do we have a workaround, even if it's temporary?
00:03 dcook         Alternatively we could put it on koha-community.org but I still don't like that external dependency, especially as there are Koha instances which are not Internet connected...
00:03 dcook         Well I'd need to work out the logic
00:02 dcook         So we could just put that file into Koha, and then have load_and_validate_schema() read it from Koha
00:02 dcook         "/usr/share/perl5/JSON/Validator/cache/36d1bd12eeed51e86c8695bd8876a9df" shoudl provide us a copy of "http://swagger.io/v2/schema.json"
00:01 dcook         JSON::Validator defaults to http://json-schema.org/draft-04/schema# right... makes sense
00:01 dcook         Errr JSON::Validator::OpenAPI rather, which is the parent of  JSON::Validator::OpenAPI::Mojolicious
00:01 dcook         Ok... JSON::Validator::OpenAPI::Mojolicious uses "http://swagger.io/v2/schema.json" if a "schema" argument isn't provided to load_and_validate_schema()