IRC log for #koha, 2023-12-27

All times shown according to UTC.

Time S Nick Message
04:31 cait1 joined #koha
07:32 mavhc joined #koha
07:57 cait joined #koha
08:00 cait2 joined #koha
08:01 cait2 good morning #koha
08:01 cait2 anyone around?
08:03 Annelisterman[m] Physically yes, mentally.. perhaps
08:05 cait2 :)
08:05 davidnind good evening!
08:06 cait hi davidnind!
08:06 davidnind :-)
08:06 cait hope you two had nice christmas holidays!
08:09 Annelisterman[m] It was pretty quiet. Some snow clearing to do. :)
08:10 davidnind it was good - had Xmas at my niece's, and visited cousins - hope you had a nice day as well!
08:11 krimsonkharne[m] morning all!
08:13 cait joined #koha
08:18 cait2 joined #koha
08:47 cait joined #koha
08:59 cait Annelisterman[m], davidnind: sorry, got pulled away by restarts for updates... had a lovely christmas with family as well :)
09:13 ashimema morning #koha
09:40 aude_c[m] 'Morning!
09:45 saa joined #koha
09:45 saa how to enable virtualkeyboard in koha for regional language search
09:46 saa in opac
09:51 krimsonkharne[m] saa: I think this was asked once a long time ago, I found a response linking to https://greywyvern.com/code/javascript/keyboard/
09:51 krimsonkharne[m] see "Installation".. you'll probably need to put that js in the html of the opac header
09:52 krimsonkharne[m] I don't know how you would use it in Koha's search bar though
09:53 saa right thats what i am wondering
09:54 saa is there any simple way to copy paste some script and enable in OPAC so that users can search in different languages
09:55 krimsonkharne[m] the "right" way of doing it would be enabling the installed languages in the system preference OPACLanguages (and make sure opaclanguagesdisplay is enabled)
09:55 krimsonkharne[m] but you would need to install the languages first (I think, I've never done it myself :-) )
09:56 krimsonkharne[m] ah, sorry, misunderstood.. that's for display, not for searching
09:57 krimsonkharne[m] I think someone else will have to chime in on that, I don't have any experience with searching different languages
09:58 saa i hope someone will guide
10:02 saa one more query opac suggestions autofill if enabled will it help like google display words while searching
10:02 saa after enabling is it necessary to restart any service to see this feature working in opac
10:08 cait We don't have an auto-complete feature at the moment, but I believe there are some snippets in the jQuery library maybe
10:09 cait https://wiki.koha-community.or[…]ki/JQuery_Library
10:10 cait there is also no feature to enable other screen keyboards, there is a self made solution you could copy here: https://wiki.koha-community.or[…]C_screen_keyboard - but it's a bit limited and could be better looking (I wrote it, so I know)
10:11 krimsonkharne[m] oh, neat
10:11 krimsonkharne[m] cait++
10:19 cait there is also Bug 12815 - but it seems to haven't moved in a while - maybe a comment would help
10:19 cait https://bugs.koha-community.or[…]_bug.cgi?id=12815
10:19 huginn` Bug 12815: enhancement, P3, ---, indradg, ASSIGNED , Adding browser based multi-lingual jquery.ime input method support to Koha
10:30 saa thank you cait let me explore
11:17 tcohen morning
11:17 ashimema morning tcohen
11:18 ashimema hope you had a good xmas
11:27 * cait waves at ashimema and tcohen
11:28 tcohen hi cait!
11:42 cait hm why are license users limited to staff with permissions in the ERM module?
11:42 cait I had expected that I could also add patrons/members of the institution staff
11:42 cait but maybe I am misunderstanding what users is meaning here
11:43 ashimema pass
11:43 ashimema the chaps in the know here aren't around to day either I'm afraid
11:44 tcohen not on my wheelhouse
11:44 cait i found Bug 33427
11:45 cait thx for replying anyway :) at least the bug tells me I didn't miss something fundamental
11:46 tcohen bug 33427
11:46 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=33427 enhancement, P5 - low, ---, jonathan.druart+koha, NEW , Managing individual users with access to subscriptions
11:47 tcohen silly VPN issue sorted. back to work!
11:47 tcohen @later tell Joubu I don't have metrics, but it feels like the weblate server is a slow one
11:47 huginn` tcohen: The operation succeeded.
11:48 tcohen I've been looking at the logs while working on translations, the moments it got stuck the Postgres process was using CPU like crazy
11:48 Annelisterman[m] I was just about to complain that Weblate is soooooo slow. x)
11:49 tcohen Annelisterman[m]: when you translate something, it gets applied to all other versions
11:49 tcohen that's the step that takes a lot of time, sometimes :-P
11:49 tcohen and the statistics update as well
11:50 Annelisterman[m] I just changed to next translation without any changes and it has been thinking about it for at least 3-4 minutes.
11:51 tcohen get a DeLorean and move a few minutes in the future
11:51 Annelisterman[m] oh, that would be handy
11:52 tcohen server load is around 9
11:52 tcohen with 4 CPU
11:53 Annelisterman[m] 9 of 100?
11:53 cait it can be super fast too, it was in testing initially
11:53 cait maybe we need to tweak configuration/setup
11:53 tcohen 900 of 100, Annelisterman[m]
11:53 tcohen cait: maybe it needs a bigger/faster server
11:54 tcohen it has plenty of memory
11:55 Annelisterman[m] mostly it is ok and quite fast but there are times when it's just pondering something really deep and meaningful.
11:55 cait maybe Postgres needs more resources?
11:56 cait but I am guessing, it just seems that this was almost our issue with performance recently :)
11:56 tcohen asking my sysadmin to look at it
12:01 cait tcohen++ sysadmin++
12:50 tcohen ok, we did some tuning
12:50 tcohen and we have less IO load and IO wait
12:51 tcohen and Postgres uses much more memory
12:51 tcohen but it seems anytime someone enters somewhere no one was at before
12:51 tcohen it takes  a lot of resources to build a cache or smth
12:51 cait ah yeah, I noticed that - the second time it speeds up
12:52 tcohen the HDD is an old one
12:52 tcohen and there was just too much IO
12:52 tcohen gabriel++
13:10 cait anyone else seen a lot of jobs getting stuck as new in rabbitMQ?
13:11 cait Elasticsearch indexing jobs especially
13:21 ashimema Es had it's own worker now didn't it
13:30 cait it should yes
13:31 cait from what I gathered so far there are 3 processes handling the task queue
13:31 cait the "normal" the "long tasks" and the elastic one
13:40 cait ashimema: do your libraries run with RabbitMQ? do you have any issues with jobs not getting done?
13:52 ashimema We turn it off
13:52 cait ok thx
13:53 cait did you notice any disadvantages without?
13:53 cait and how do you turn it off actually? is there a switch somewhere?
13:55 ashimema Just kill rabbit
13:56 ashimema The workers will then fall back to polling the database
14:00 cait since introduction of rabbitMQ this channel got more violent
14:08 Dyrcona joined #koha
14:13 ashimema lol
14:13 ashimema i don't know who is using rabbit honestly
14:14 cait I know someone who is trying
14:15 ashimema I think in theory if your going to use multiple workers for the same queue to process jobs in parallel the rabbit is better
14:15 ashimema but so far I've not really seen any need to do that
14:15 ashimema and really, not many of our jobs are actually designed for such running anyway yet
14:16 ashimema I think I'd be pro putting a switch in rather than the current behaviour of filling the logs and putting a warning in the about oage
14:17 ashimema yup
14:17 cait can you explain that last bit?
14:22 ashimema currently database polling is a fallback. if you uninstall or stop the rabbit service, SQL will take over.
14:22 ashimema but
14:22 ashimema we'll spit out a warning in the logs forever
14:23 ashimema and put a message on the about page
14:24 cait oh that's not realyl nice
14:24 cait yeah, a proper switch would do better especially if a lot of people are running with SQL
14:24 ashimema yup
14:24 cait do you want me to file a bug for that?
14:25 cait .... and bad news on weblate - down for anyone but me?
14:25 cait 502 bad gateway on submitting a translation
14:26 ashimema feel free to submit a bug cait
14:26 ashimema I don't think it would be all that hard to achieve
14:26 ashimema but I also know some people are keen to rip out the sql polling support entirely instead
14:27 ashimema I don't know the answer
14:27 cait you mean people might disagree... I can live with that
14:31 cait bug 35655
14:31 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=35655 normal, P5 - low, ---, koha-bugs, NEW , Make it possible to switch off RabbitMQ without any warns in logs/about page
14:32 cait tcohen: could you check on the weblate server again?
15:10 cait it's not working at all for me right now
16:00 bag joined #koha
16:28 * cait waves
16:42 tuxayo 260 working runs of the tests for Bug 35506 so far :)
16:42 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=35506 normal, P5 - low, ---, victor, ASSIGNED , selenium/regressions.t is failing randomly
17:01 cait ashimema or tcohen still around?
17:02 cait RM question about a patch
17:02 ashimema Phone only now I'm afraid
17:02 cait maybe it's easy
17:02 cait I am reviewing 31297
17:02 ashimema Fire away
17:02 cait you changed the database structure
17:02 ashimema Bug 31279
17:02 cait i need to run dbic right?
17:02 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=31279 critical, P5 - low, ---, koha-bugs, CLOSED INVALID, Getting error 500 while doing a search in koha catalogue
17:02 cait h
17:02 cait 31297
17:03 ashimema Bug 31297
17:03 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=31297 normal, P5 - low, ---, nick, Passed QA , Cannot add new subscription patterns from edit subscription page
17:03 ashimema Lol
17:03 cait double checked, I got the number right in my DBRev commit
17:03 ashimema Okies
17:04 ashimema Yes, you do need to run dbic
17:04 cait so i ran dbic and i have a change in SubscriptionNumberpattern.pm
17:04 ashimema And commit the result 🙂
17:04 cait ok cool :)
17:04 cait on the right path then
17:04 ashimema Coolioes
17:04 cait should i just rwap it in the DBRev patch or better keep separate?
17:05 cait hm looks like we keep them separate from history
17:07 cait got it; I think :)
17:10 ashimema Nice one cait
17:12 cait another first - but getting there
17:15 cait ashimema++ # thx!
17:20 cait2 joined #koha
17:23 cait3 joined #koha
18:16 cait joined #koha
18:17 cait2 joined #koha
18:18 cait4 joined #koha
18:21 cait4 left #koha

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary