Time Nick Message
00:01 davidnind It's great watching the PTFS Europe customer networking day videos - https://www.youtube.com/@PTFSEurope/videos
00:02 davidnind Bugzilla things I've learnt: use kw:academy or kw:sandbox in the Bugzilla search box
00:02 davidnind ashimema++ // for adding the sandbox keyword - bugs that can be tested in the sandboxes
03:04 Danyon Hi all, am currently having an issue with my KTD installation where trying to start it with a "ktd up" will just hang while waiting for mysql DB to connect to it, this is what the output looks like if anybody knows what's going on? I last used KTD last week with no issues and haven't changed anything I can think of in that past week
03:04 Danyon https://pastebin.com/c2y0jx1e
03:26 dcook Danyon: I'd say "The environment variable SYNC_REPO does not point to a valid Koha git repository." would be your problem?
06:30 ashimema Haha, you quick off the mark there davidnind
06:30 ashimema I've not even announced the sandbox keyword yet. I think it a good idea and want to encourage people to use it
06:31 ashimema Also more test plans written with sandboxes in mind
07:42 reiveune hello
08:05 krimsonkharne[m] morning #koha
08:06 krimsonkharne[m] cait around?
08:08 cait2 around now
08:08 cait what's up?
08:12 krimsonkharne[m] about an RFC: https://wiki.koha-community.org/wiki/OPAC_suggestions:_Put_a_hold_on_item_when_suggestion_is_received_RFC
08:12 krimsonkharne[m] do you know if there's anything about it in the pipeline for future releases?
08:25 cait oh sorry, got called away, checking now
08:25 cait I think there is ab ug report with more information/discussion
08:26 cait bug 27595
08:26 huginn` 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=27595 enhancement, P5 - low, ---, koha-bugs, NEW , Place holds for patrons on accepted purchase suggestions
08:26 cait no sponsoring yet I think
08:26 krimsonkharne[m] oh hey, cheers
08:29 cait krimsonkharne[m]: what we really need to do is fix the link between user and suggestion/order
08:30 krimsonkharne[m] sounds like another bug I'm too green to have heard of :-)
08:30 cait I am trying to find that one
08:30 krimsonkharne[m] Roger found this one: https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=19358
08:30 huginn` 04Bug 19358: enhancement, P5 - low, ---, agustinmoyano, NEW , Purchase suggestions - enhancements to linking to orders
08:30 krimsonkharne[m] probably not what you meant
08:30 cait yeah, there is another i left a lengthy comment on, give me a moment
08:31 cait bug 28844
08:31 huginn` 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=28844 major, P5 - low, ---, jonathan.druart+koha, Failed QA , Suggestion from existing title can alert patron in error
08:31 cait it shows an implementation flaw, but I didn't come up with an easy fix for the old data
08:33 krimsonkharne[m] ah
08:33 krimsonkharne[m] taking a look
08:35 krimsonkharne[m] cait++
08:43 Joubu perltidy's output can be different depending on the versions of perltidy you are using
08:44 Joubu ashimema: about the sandbox kw, I would do the reverse: announce that the patch cannot be tested on a sandbox. Otherwise testers might be afraid to test patches if the kw is not there.
08:44 Joubu there is also https://wiki.koha-community.org/wiki/Sandboxes#What_is_testable_or_not_in_a_sandbox
08:51 cait Joubu: do we have a version I should be using?
08:52 cait (perltidy)
08:52 Joubu the one in ktd
08:52 cait I had a question today about the string freeze - we are planning for the end of the week - are we still aiming for Pootle right now?
08:53 cait I'll compare then
08:53 Joubu see also last commments on 30002
08:53 cait and try to match it
08:53 cait I compare using a visual tool, so doing it in my host
08:53 Joubu 23.11 string freeze?
08:53 cait yes
08:54 cait s/compare/perltidy...
08:54 Joubu I am waiting for October releases...
08:54 Joubu for 2 weeks now
08:55 cait aaah, the missing bugs :(
08:55 cait I see
08:55 cait i hope we will see something happen today or tomorrow
08:55 Joubu once they are released I will create the components for the stable branches on weblate, freeze pootle, sync with weblate, push some patches to stable, have a cronjob to update koha-l10n
08:55 cait but they won't have the bugs right?
08:55 Joubu and then dns switch
08:56 cait or will they now?
08:56 cait I'll keep fingers crossed
08:56 cait have a meeting now
09:45 cait Joubu: can we walk through the acq bug maybe together?
09:45 cait I have tested multiple times on yesterday's master
09:46 cait I believe we might have been doing something different
09:46 Joubu yes sure
09:48 cait hm how best to do this?
09:48 Joubu pm
10:43 mtj hey folks
10:45 mtj im looking at bug 31729, domms patches with plackup -R seem to work ok
10:45 huginn` 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=31729 enhancement, P5 - low, ---, domm, In Discussion , Enable automatic filesystem refresh in Plack
10:46 cait I think this one is a bit too hard for me - maybe domm is around if there are any questions?
10:46 mtj ..the CPU is higher with -R, but it still looks workable
10:46 domm[m] hey
10:47 domm[m] As I've commented in the Bug, I don't think it makes sense to apply it
10:48 mtj hi domm[m], yes, i see your comment
10:48 domm[m] koha-plack is rather complex and does a lot of stuff. Hot-Reload is only needed during dev.
10:49 mtj im perfectly happy to move the Hot-Reload to another script, that is not koha-plack :)
10:49 domm[m] So instead of risking problems in production (by fiddling with koha-plack) it will be much easier to just use a different plackup command in dev (so only devs who actually want hot-reload can use it)
10:50 domm[m] (hm, do you see thumbs-up reactions in IRC? Or is everybody using matrix anyway?)
10:54 Joubu only matrix people
10:54 Joubu having a separate script sounds good
10:54 mtj domm[m]: just a small request from me... can the solution work outside of ktd too please :)
10:55 mtj (some older devs dont use ktd for everything, all the time)
10:57 domm[m] One could always come up with a plackup -R dir/to/watch .... /etc/koha/plack.psgi command and run that in any context. Any helper/wrapper scripts will just help with getting all the options to plackup right
10:57 domm[m] /usr/bin/plackup -M FindBin -R /kohadevbox/koha --workers 2 --user=${instance_user} --group ${instance_user} -E deployment --socket ${placksocket} -s Starman ${psgifile}
10:58 domm[m] If we pack something like this into a wrapper script or "just" add it to the documentation is a matter of taste :-)
11:01 mtj regarding higher CPU... https://lists.koha-community.org/pipermail/koha-devel/2016-September/043026.html
11:04 mtj plackup -R seems to be doing a reload every second, during filesystem changes (like changing git branches etc)
11:08 oleonard o/
11:10 mtj my original patchset was to find a solution that didnt burn CPU... but now, i would just be happy that any solution is accepted :)
11:12 Joubu this vim mapping and git hook idea is excellent actually
11:12 Joubu no idea why I am not using them anymore
11:19 mtj Joubu: i did try a shortcut key for a longgg while, but ultimately wanted to automate a reload
11:22 mtj hi cait, regarding... https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=31729#c52
11:22 huginn` 04Bug 31729: enhancement, P5 - low, ---, domm, In Discussion , Enable automatic filesystem refresh in Plack
11:22 mtj ..this plackup --reload option, probably is not a goos solution for a prod system
11:23 mtj s/goos/good/
11:24 mtj it uses too much extra CPU to be considered, (i think)
11:41 tcohen[m] hola #koha o/
11:45 cait mtj: makes sense, so the plugin issue remains to be solved... if at all possible?
11:46 ashimema morning cait
11:46 ashimema bug 31041 is hopefully a bit clearer to test now 🙂
11:46 huginn` 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=31041 minor, P5 - low, ---, lucas, Signed Off , Cashup summary modal printing issue
11:46 ashimema Or I can QA it.. but I did some refactoring to generalise it more.. so it may be worth a second eye
11:46 ashimema this would be another great one to have fixed at last.
11:46 cait I can try, but i have some ohter issues to deal with first
11:47 cait acq..
11:47 cait and currently in a call and no lunch yet :)
11:48 ashimema no worries
12:07 davewood ES problem once again. if I delete&rebuild the index for one specific biblionumber it works. if I do it for the entire catalogue i see errors when rebuild_elasticsearch.pl and no biblios will be indexed. afterwards if i try to re-index a single biblio i get a warning that "Index "biblios" has status "recreate required", suggesting it should be recreated at /usr/share/koha/bin/search_tools/rebuild_elasti
12:07 davewood csearch.pl line 270.
12:08 davewood using Search::Elasticsearch 7.717 and 6.80 with the same error message.
12:10 davewood maybe i should restart koha services after installing a different version. ill give that a try.
12:10 oleonard davewood: I'm guessing that is required
12:14 davewood i restarted koha-plack and koha-common ... did I miss anything?
12:14 davewood because i still get the same error.
12:20 davewood i have another koha instance where I dont have this problem. both are using the same ES instance 7.17.6 ... the instance that is working uses Search::Elasticsearch 7.30 ... so i used that on my non working instance but still see the same error.
12:20 davewood could the data in the koha database be erroneous?
12:41 davewood koha only says that an error occurred but not what the error is.
12:46 davewood i managed to narrow it down to a specific biblionumber that makes rebuild_elasticsearch.pl fail and the ES index is corrupted afterwards.
12:47 davewood question: should I not do whatever it is im doing here right now? spamming you guys with my ES issue?
12:50 oleonard Sorry davewood you're just not catching anyone around who knows how to answer your question
12:50 davewood I dont mind that part, i just dont want to annoy anyone.
12:58 oleonard davewood: It's not like you're talking over anyone :D
13:22 davewood i started digging and got to this bit ... [TOO_MANY_REQUESTS/12/disk u
13:22 davewood sage exceeded flood-stage watermark, index has read-only-allow-delete block];'
13:22 davewood but koha isnt reporting this error. just carp'ing a hardcoded error message instead
13:24 davewood https://github.com/Koha-Community/Koha/blob/master/Koha/SearchEngine/Elasticsearch/Indexer.pm#L140
13:30 Joubu davewood: are you able to confirm that a relevant error is in $response->{errors}?
13:31 davewood so the actual error is probably that our FS is full.
13:32 davewood this is $response -> http://paste.scsys.co.uk/18562
13:34 Joubu that would be interesting to display it instead of hiding it behind a generic error
13:41 davewood now to check why our monitoring didnt report the full disk.
13:58 cait sorry, was in a meeting - reading back now
13:59 cait davewood++ please don't feel shy to talk around here :)
14:07 cait if nothing else, we can be rubber ducks (listening to help solve the problem)
14:07 cait hi caroline
14:07 caroline hi!
14:08 * cait hands the day over to caroline and runs away
14:08 caroline will be on and off for the next three days, it's the big library conference but we have wifi)
14:09 caroline uh oh, what's goin on?
14:09 caroline if cait flees, there must be something wrong lol
14:10 cait just a lot :)
14:10 cait it as worth a try
14:10 cait do you have a booth or sitting in talks?
14:12 caroline both, booths start tomorrow, but eric lets me roam around in talks also
14:13 cait nice, have a good time :)
14:13 lukeg Hi
14:13 cait hi lukeg
14:13 cait kidclamp: around?
14:14 cait got a question about the rabbit if you have a moment...
14:43 kidclamp Sure
14:47 cait pm
14:47 ashimema bug 34704 back at you cait 😜
14:47 huginn` 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=34704 normal, P5 - low, ---, matt.blenkinsop, Signed Off , Print templates are formatted incorrectly
15:14 Joubu @later tell mtj ping me (again) where you are around please :)
15:14 huginn` Joubu: The operation succeeded.
15:17 tuxayo Hi :)
15:17 tuxayo Does this looks good UI wise? https://pic.infini.fr/gallery#rTRO8lJI/mLFEXEiQ.png
15:17 tuxayo It's from a patch to add a hold cancelation reason when checking out an item that has a waiting hold.
15:18 tuxayo And where is it possible to find if the cancelation reason was effective? Without setting up emails to check the notification sent to the patron.
15:27 ashimema looks a bit all over the place to me
15:27 ashimema but the core info appears to mostly be there to me
15:32 reiveune bye
16:12 logk[m] Hello, I would like to understand CPU requirements for a possible Koha installation. Can anyone share what CPU you are using, CPU usage, approx. number of bibliographic records and user activity? Or point me to such info? Is there a centralized list of hardware being used? Thanks.
16:19 davewood his is what I found after googling "koha system requirement" => https://wiki.koha-community.org/wiki/System_requirements_and_recommendations
16:19 davewood hmm no HW info though
16:19 logk[m] Thank you. I have seen this, but it does not answer my question.
16:20 logk[m] I have found a few random examples from users, but I'm looking for more. For example: https://www.mail-archive.com/koha@lists.katipo.co.nz/msg30629.html
16:20 davewood https://kohageek.blogspot.com/2019/09/hardware-selection-for-koha-installation.html
16:22 logk[m] Thank you, unfortunately that does not list actual hardware in use. The recommendation of "i5" by itself doesn't really answer the question.
16:22 ashimema it depends on a lot of factors logk
16:23 ashimema we have it running on a raspberry pi 2
16:23 ashimema and we have it running on some mega hardware
16:23 ashimema are you going to run zebra or elastic
16:23 ashimema are you expecting hudnreds of requests or millions
16:23 ashimema are you exposing the API's and using them heavily or not
16:23 ashimema etc
16:24 ashimema there's not really a central base of knowledge for that as Koha is hugely flexible and can run in a whole host of different configurations
16:26 logk[m] Zebra, probably. Very low number of requests. 100-200k bibliographic records. Very low API usage. Limited patron usage, main users will be a few librarians. I suspect CPU usage will be quite low, but looking for examples to try to confirm this. Or, I might set up a test installation at a cloud provider and vary the hardware to see real performance. However, final installation will not be on local hardware.
16:27 ashimema that all sounds like fairly low usage to me, so you should be able to get away with some fairly barebone specs
16:27 logk[m] Meant: final installation will be on local hardware.
16:28 ashimema as you suggest.. might be worth giving it a go with a few different configurations to play.
16:29 ashimema caching is important.. we use a persistent daemon (plack) and memcached..
16:29 logk[m] Yes, from what I've read, it sounds like RAM is much more likely to be a bottleneck than CPU.
16:30 ashimema I gave up access to our server stats a while ago I'm afraid, preferring to hand that part of my job off to our infrastructure team here.. else I'd try to be a bit more helpful
16:30 logk[m] No worries. Thanks very much for your comments.
16:38 cait logk[m]: it will give you a good performance gain to make sure your DBMS has enough RAM
16:38 cait there are tools like mysqltuner that can give hints
16:39 cait it can make a huge difference as we have seen recently ona project
16:47 logk[m] <cait> "logk: it will give you a good..." <- Thanks.
17:54 cait1 dcook: around?
17:55 davidnind about 5am where he is, so probably not 8-)
17:55 cait1 was just tryingto reply here to the email (in case I am wrong ;) )
18:21 JasonGreene[m] hello all I'm trying to convert our current library system to Koha. We are using Primasoft's "Handy Library Manager" Has anyone made this conversion?
18:27 davidnind I had not heard of the software until now, but I'm sure others have
18:28 davidnind the main challenge I think (with any system) is with getting your record, item, and other data out, and then getting the MARC records ready for import into Koha
18:29 davidnind many of the support providers use scripts to work with the data, but others use MARCEdit as well
18:30 davidnind I would suggest posting to the general mailing list to see if you get any response
18:34 davidnind we don't have a comprehensive implementation guide, but I've just started work on one - there are some check lists in the manual (https://koha-community.org/manual/latest/en/html/implementation_checklist.html) and some material on the Wiki (https://wiki.koha-community.org/wiki/Newbie_guide)
18:38 davidnind There are some good resources at koha-US (https://koha-us.org/learn/), ByWater Solutions (https://koha.bywatersolutions.com/) and inLibro (https://inlibro.com/en/products/koha-open-source-lms/knowledge-base/)
18:39 davidnind Also, ByWater Solutions recently posted a video about their implementation process https://www.youtube.com/watch?v=0MVb2pk7EcU
18:49 ashimema davidnind++
20:23 cait1 JasonGreene[m]: Haven't heard of it either, but I am in Germany - where are you located?
21:21 cait1 oleonard-away: are you still around somewhere?
21:21 cait1 or davidnind: i have trouble reproducing the issue on bug 35280
21:21 huginn` 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=35280 normal, P5 - low, ---, oleonard, Signed Off , OPAC patron entry form: Patron attributes "clear" link broken
21:21 cait1 ooh opac.
21:21 cait1 i anwered my own quesiton, but I can report it works really well in staff
21:24 davidnind I was too slow - glad you figured it out!
21:25 cait1 thx for the testing :)
21:32 davidnind sorry about filing up the signed off queue again 8-) (mainly concentrating on bugs rather the enhancements or new features)
21:38 cait1 that's just the right thing, thanks
22:18 * dcook looks at scrollback
22:26 davidnind dcook: cait1 was looking for you earlier 8-)
22:27 cait1 oh?
22:27 cait1 ah
22:28 cait1 i was just going to say that I think you need to use ktd --es7 up
22:28 cait1 if you want elasticsearch - then ti will set itself up alright
22:34 dcook Hmm interesting
22:34 dcook I tend to just use docker-compose directly and I think it worked all right without that..
22:35 dcook Actually maybe I do use some aliases like "kd" and "ku"
22:35 dcook I have no idea how to use "ktd" since it doesn't seem to be an alias I can find..
22:37 dcook Ah ./bin/ktd
22:37 cait1 you need to set it up once
22:37 cait1 it was added a longer whiel ago, but if your ktd is older it migh tnot be there
22:37 cait1 the ktd aliases are very useful
22:38 lukeg yes, very very useful!
22:38 dcook I recall other people using it but I figure I'm one of the only people doing things on Windows, so I usually ignore what other people are doing heh
22:38 cait1 you are not the only one I think
22:38 dcook No, I think I might be the first. A few other have followed in my footsteps hehe
22:39 dcook Looks like I use a modified "ku" anyway
22:39 dcook Also hi lukeg :)
22:39 lukeg Hello dcook! :)
22:40 dcook And now I notice Jonathan's email explaining how to use ktd heh
22:40 dcook I was going to say I haven't followed these instrutions and it is working, but it must be for working out of the box..
22:40 dcook But I'm not so concerned about ktd
22:40 dcook More interested in running Elasticsearch in prod for some big Koha libraries
22:42 dcook koha-es-indexer --start --quiet $(koha-list --enabled --elasticsearch)
22:43 dcook Oh interesting...
22:43 dcook search_engine=$(koha-shell $instancename -c "/usr/share/koha/bin/admin/koha-preferences get SearchEngine 2> /dev/null")
22:43 dcook That's nifty
22:44 dcook Koha is so big you don't even realize half of what it can do...
23:24 tuxayo logk: "Very low number of requests" How much concurrent usage on daily or weekly busy hours can you expect?
23:26 tuxayo @later tell ashimema "looks a bit all over the place": about the button being on two line now? Or is that part good enough?
23:26 huginn` tuxayo: The operation succeeded.
23:29 tuxayo Anyone else to tell if the bottom on that UI is good enough despite the buttons going on two lines? https://pic.infini.fr/rTRO8lJI/mLFEXEiQ.png
23:29 tuxayo After the addition of hold cancellation reason when checking out an item that has a waiting hold. (that's a proposed patch I'm reviewing)