← Previous day | Today | Next day → | Search | Index
All times shown according to UTC.
Time | Nick | Message |
---|---|---|
00:52 | fridolin joined #koha | |
01:13 | sonOfRa joined #koha | |
02:02 | wajasu | tcohen: thx i saw that but who would have thunk docker did its magic perm thing. just like the db and memcached without an entry in /etc/hosts |
02:03 | i hadn't used docker in 4 years and it's changed. | |
02:03 | so now i can sleep tonight. | |
02:06 | so the image has it in the koha-db template. | |
02:11 | ls | |
02:57 | tcohen | wajasu: docker does many things like routing, virtual networking, name server |
03:16 | wajasu | instead of hardcoding localhost, i made koha-create honor --dbhost, and can do: koha-create --create-db dog4 --dbhost='%' and the grant will get set correctly |
03:17 | that --dbhost could be the koha-koha-1 ip instead of a wildcard. | |
03:18 | then i shutdown the ktd, edited the .env KOHA_CONF and KOHA_INSTANCE to user dog1 and did kul | |
03:18 | it all came up, but now dog1 and not kohadev processes. | |
03:21 | that --dbhost could be the koha-koha-1 ip instead of a wildcard. | |
03:21 | definitely gives a consistent environemnt for testing. | |
03:24 | wonders if multitenant sites run an instance of apache per site, or one apache server for a chunk of sites/databases. | |
03:27 | loves all the computer related books in the test data :) | |
03:31 | tmrw i chase down my libraries' autoincrement data instances. hopes its not too bad. | |
04:01 | Rajat joined #koha | |
04:01 | Rajat | HI Team |
04:02 | is there any rest api for OPAC ? | |
05:44 | reiveune joined #koha | |
05:44 | reiveune | hello |
05:51 | dcook joined #koha | |
05:52 | dcook | allo allo #koha |
06:07 | fridolin joined #koha | |
06:23 | jajm joined #koha | |
06:41 | cait joined #koha | |
06:43 | cait1 joined #koha | |
06:50 | cait joined #koha | |
06:52 | cait1 joined #koha | |
06:53 | cait1 | good morning #koha |
07:08 | Joubu | tcohen: how long are we going to ignore jenkins' failures? |
07:08 | Oak joined #koha | |
07:16 | Joubu | ktd does not start for me this morning, only me? |
07:17 | cait1 | not going to stop and restart mine now ;) |
07:27 | ashimema | Not tried ktd yet this morning |
07:27 | Joubu | Invalid data, cannot decode metadata object (biblio_metadata.id=368, biblionumber=369, format=marcxml, schema=MARC21, decoding_error=':8: parser error : PCDATA invalid Char value 31 |
07:27 | ashimema | The dashboard isn't too terrible.. some selenium tests failing on D11 which need looking into |
07:27 | Joubu | I am getting that |
07:27 | then a weird koha_db output | |
07:27 | ashimema | Humm, interesting |
07:27 | Joubu | [Warning] Aborted connection 13 to db: 'koha_kohadev' user: 'koha_kohadev' host: '172.23.0.6' (Got an error reading communication packets) |
07:27 | but only a warning, so maybe not the cause of the crash | |
07:28 | ashimema | That sounds kinda like your Get.. to ->metadata patch |
07:28 | Oh, the first but did | |
07:28 | That second one not so much | |
07:28 | Joubu | ha! |
07:28 | I am not on master! | |
07:28 | thanks! | |
07:28 | ashimema | I can try to take a look at selenium this morning after dropping the car at the garage |
07:29 | I know Tomas and Mason have been doing stuff on Jenkins so I hadn't stepped in to help yet | |
07:30 | Humm.. is that another middle name fallout? | |
07:31 | I though we'd already fixed that | |
07:32 | * ashimema | thinks his brain may be a bit mush today, us Brits aren't used to this sort of heat |
07:32 | ashimema | It hit 40C in my little home office yesterday and today is forecast for a fair bit warmer |
07:33 | Joubu | same here... |
07:33 | ashimema | Regards the GetItemInfo patch.. do we have a plan on how to help users out of the bad data hole when they hit it? |
07:34 | I can't remember what the final state of the bug was.. | |
07:34 | Joubu | bad data? |
07:34 | ashimema | That error you highlighted above |
07:34 | Isn't it caused by a Marc record failing to parse | |
07:34 | Joubu | it's bug 29697, I just let a new comment |
07:34 | ashimema | This bad data? |
07:35 | huginn | Bug https://bugs.koha-community.or[…]_bug.cgi?id=29697 enhancement, P5 - low, ---, jonathan.druart+koha, Passed QA , Replace GetMarcBiblio occurrences with $biblio->metadata->record |
07:35 | ashimema | Yeah, that one |
07:35 | Sorry.. I'm still on the mobile right now.. | |
07:35 | Joubu | biblionumber 368 is wrong in ktd, that's "expected" (for testing purpose). |
07:35 | we have the search_for_inconsistencies script, and the detail.pl view that is displaying the error | |
07:35 | no regression is expected from this patch | |
07:36 | but now there is one as ktd is failing | |
07:52 | * cait1 | waves |
09:02 | Oak joined #koha | |
09:20 | ashimema | @seen liliputech |
09:20 | huginn | ashimema: liliputech was last seen in #koha 1 week, 0 days, 21 hours, 19 minutes, and 39 seconds ago: <liliputech> back |
09:27 | ashimema | sill on leave? |
09:40 | dpk_ joined #koha | |
09:55 | cait joined #koha | |
10:24 | khall_ joined #koha | |
10:35 | oleonard | Hi #koha |
10:54 | cait1 | hi oleonard :) |
10:58 | ashimema | mornin |
11:14 | paulderscheid[m] | cait1: Slightly off-topic but does the OGND have an api? |
11:15 | cait1 | apart from SRU/Z39.50? |
11:17 | paulderscheid[m] | Yes |
11:19 | cait1 | what kidn of API are you looking for? |
11:19 | paulderscheid[m] | I want to build a tool with which I can query the main reference form of terms. |
11:19 | Maybe I can build one around Z39.50 but that sounds like a pain | |
11:20 | Does something like that already exist? | |
11:21 | cait1 | you mean the main heading? |
11:21 | can you give an example maybe? | |
11:21 | paulderscheid[m] | Yes, one second |
11:21 | cait1 | and possibly you shoudl look at the GND at DNB too |
11:23 | paulderscheid[m] | For example: If I have the term "Pirat" (in german) the actual subject term is "Seeräuber" |
11:23 | cait1 | that's a fun one |
11:23 | paulderscheid[m] | It's the most memorable one I can think of |
11:24 | And using the website is fine but also slow | |
11:24 | cait1 | maybe SRU would be a bit more fun: https://wiki.k10plus.de/pages/[…]?pageId=132874251 |
11:24 | than Z39.50 | |
11:24 | paulderscheid[m] | Ok |
11:24 | cait1 | the DNB also offers an SRU target for GND |
11:24 | paulderscheid[m] | I'll lokk into that |
11:25 | Thanks for the pointers, I think I can work with that | |
11:26 | marcelr joined #koha | |
11:26 | marcelr | hi #koha |
11:32 | cait1 | hi marcelr |
11:33 | marcelr | gutentag |
11:33 | surviving the heat? | |
11:34 | bug 31186 for QA or RM | |
11:34 | huginn | Bug https://bugs.koha-community.or[…]_bug.cgi?id=31186 normal, P5 - low, ---, m.de.rooy, Signed Off , Search result numbering in OPAC got suppressed |
11:35 | khall joined #koha | |
11:35 | ashimema | so bug 30952 |
11:35 | huginn | Bug https://bugs.koha-community.or[…]_bug.cgi?id=30952 enhancement, P5 - low, ---, solene.desvaux, Failed QA , New interface for staff client |
11:36 | ashimema | would it be a really bad idea to suggest we move the work into an entirely new theme.. push it quickly so we can work on cleaning it up, smoothing out the rough edges and then drop the old theme once we feel it's done? |
11:36 | I'm just trying to look for ways forward.. there seems a lot of issue to sort out still to me. | |
11:36 | paulderscheid[m] | That sounds like a good idea |
11:36 | ashimema | the challenge would be that for a period, we'd need bugs submitted to support both sets of templates |
11:37 | paulderscheid[m] | But as recently discussed, cutting the thing up would be even better |
11:37 | ashimema | yeah.. I'm looking at how we might achieve that right now |
11:37 | cait1 | marcelr: hiding inside helps |
11:37 | ashimema | I think people's issue with cutting it up was that you'd end up with parts in and it would look a mess as a whole |
11:38 | personally.. i'd love to have seen it submitted as a whole set of bugs in a neat tree.. and we could have worked through them over the course of the cycle | |
11:38 | paulderscheid[m] | Well it's the master branch and not a release |
11:38 | ashimema | right now.. I am not entirely optimistic anyone will sit there with them massive single patch and actually make much headway.. |
11:38 | there's no test plan, no description of any of the changes | |
11:39 | just "Here's a screenshot.. now go test it" | |
11:39 | paulderscheid[m] | I was about to say that |
11:39 | cait1 | ashimema: it hink cutting it up is the better idea |
11:39 | we had 2 opac themes for a while and it was very messy | |
11:39 | paulderscheid[m] | Without a test plan this will stay on the shelf forever |
11:39 | cait1 | it turned out really hard to maintain them both |
11:39 | ashimema | it's going to be an absolute nightmare for people upgrading |
11:39 | cait1 | and the staff interface is much more complex than the OPAC |
11:39 | ashimema | so many system preferences affect the appearance of things.. |
11:40 | cait1 | maybe we need to work towards not making it a nightmare then |
11:40 | i haven't looked yet,... but is it really new templates? | |
11:40 | ashimema | like your IntranetNav stuff.. I can't begin to imagine how much stuff our customers have put into those prefs |
11:40 | css and js too | |
11:40 | cait1 | i had the impression it was changing stuff like the header include and css |
11:40 | paulderscheid[m] | This sounds like it should be spearheaded by oleonard |
11:40 | In collaboration with BibLibre | |
11:40 | cait1 | yeah, i think a big patch might be too late now anyway |
11:41 | ashimema | it touches around 300 files |
11:41 | on a quick look | |
11:41 | paulderscheid[m] | I'd be happy to help but this patch requires guidance for testers |
11:41 | ashimema | includes, template, css, images |
11:41 | cait1 | that's a lot, but dpeendson the changes of course |
11:42 | ashimema | lots of them are rather large template re-writes in themselves |
11:42 | cait1 | yeah, they shoudl have been separate |
11:42 | Oak joined #koha | |
11:42 | ashimema | refactoring and moveing things around in the template |
11:42 | marcelr | thats problematic |
11:42 | ashimema | yup |
11:42 | cait1 | and the header area needs more work |
11:42 | ashimema | I have looked at it a couple of time time |
11:42 | it's chock full of bugs and issues | |
11:42 | marcelr | this just tells us that our templates are a nightmare |
11:42 | ashimema | but I don't even know how to report them.. |
11:42 | cait1 | hm not necessarily, but big patches are |
11:43 | ashimema | I want to try to be positive on it.. but the list of issues would end up hundreds of lines long |
11:43 | and that's before I start on the big stuff that is already being debated in the bug | |
11:43 | paulderscheid[m] | Can we create a wiki page where we lay this out and formulate tasks that individuals can pick? |
11:43 | ashimema | like the search stuff |
11:43 | I don't know | |
11:43 | oleonard | The template changes are actually fairly minor, which was the goal starting out |
11:44 | ashimema | minor they may be.. but they're a nightmare to rebase unless you're close to the patch |
11:44 | because they're often block moves.. git can't help much | |
11:44 | oleonard | Unfortunately a lot of the changes may actually be merge errors or regressions... Lots of headings which have been changed |
11:44 | ashimema | indeed |
11:44 | I think i'd have gone through it almost a template at a time to get them all uniform | |
11:45 | those would have been tiny fast moving patches | |
11:45 | then a css change | |
11:45 | then function UX improvements like moving search to the header.. making side bars into collapsible menus | |
11:45 | and stuff like that | |
11:46 | * ashimema | really wants to help but is struggling to know how |
11:47 | ashimema | There's so really big changes that were really not proposed at all too.. |
11:47 | oleonard | I feel like we need a meeting with Biblibre to discuss some of their choices |
11:47 | ashimema | like the main page change.. I find that pretty jarring to be honest |
11:47 | the strange layout | |
11:48 | oleonard | Like... Was this scrollbar a decision? https://snipboard.io/1VIsR6.jpg |
11:48 | ashimema | indeed |
11:48 | what I'd really like to have seen was a clear style guide written instead | |
11:48 | and then everyone help work towards it one piece at a time | |
11:49 | oop. lunch time.. I best feed the hungry hoards.. | |
11:49 | * ashimema | has kids home today.. schools closed due to the heat |
11:49 | Oak joined #koha | |
11:50 | paulderscheid[m] | The style guide should still be created |
11:51 | This new interface won't stay like this forever | |
12:10 | tcohen | hola #koha |
12:11 | Oak joined #koha | |
12:24 | marcelr | hola tcohen |
12:24 | tcohen: all is fine for the lists patches now | |
12:57 | huginn | News from kohagit: Bug 30905: (follow-up) Get logged in branch once <https://git.koha-community.org[…]27d3c425377fcc27a> |
12:57 | News from kohagit: Bug 30935: Holds to pull shows wrong first patron <https://git.koha-community.org[…]13d2390fd57a27a36> | |
12:57 | News from kohagit: Bug 30905: Show waiting recalls in patron account <https://git.koha-community.org[…]e7cf4eb6febf31698> | |
12:59 | wajasu | well, i think i have a solution for koha-create to get the right grant in mysql per user/instance. |
13:05 | since i don't have koha_dog1.cnf in /etc/mysql, I can grab it from DB_HOSTNAME (i.e. db or whatenver .env specifies). then if i can use 'dig $DB_HOSTNAME +short' to get the IP of the database server, and use that IP in the GRANT for the koha_dog1 user. | |
13:05 | no wildcard needed too. | |
13:05 | but we need dig added to the koha-common package deps | |
13:07 | koha-jenkins | Project Koha_Master_D11_ES6 build #23: SUCCESS in 8 min 53 sec: https://jenkins.koha-community[…]aster_D11_ES6/23/ |
13:07 | wajasu | this will let us test koha-create/koha-remove etc in ktd and support start/stopping of services, etc |
13:07 | on another note. | |
13:09 | is it feasable or worth it to create the koha-db image or a koha-db-pop image that has the database populated. | |
13:10 | so when i do kd, then kul, i don't have to wait for that long process. | |
13:10 | maybe even have zebra indexed too. | |
13:11 | then i'd just tweek .env and start/stop quicker for each bug to fix/signoff | |
13:12 | but since you guys pull images from a cental docker registry, that would make it part of your offering. | |
13:12 | i could try locally with my own private docker registry to prove it out first. | |
13:13 | or just build the image and specify my docker compose to pull formthe image ID. | |
13:14 | kulp could be the alias :) | |
13:17 | alohabot | 🎁 🦄 Koha 'master' packages pushed to 'koha-staging' repo 🍜🥭🍈 |
13:26 | Dyrcona joined #koha | |
13:31 | koha-jenkins | Project Koha_Master_D11_ES7 build #24: SUCCESS in 7 min 41 sec: https://jenkins.koha-community[…]aster_D11_ES7/24/ |
13:34 | lukeg joined #koha | |
13:37 | koha-jenkins | Project Koha_Master_U20 build #493: SUCCESS in 38 min: https://jenkins.koha-community[…]a_Master_U20/493/ |
13:40 | tcohen | wajasu: why do you want to know the IP? |
13:44 | wajasu | i can just use % wildcard. but if security is a concern, the IP might be nicer. |
13:44 | the IP of the koha-koha-1 that is. | |
13:44 | koha-jenkins | Project Koha_Master_D11_My8 build #924: SUCCESS in 46 min: https://jenkins.koha-community[…]ster_D11_My8/924/ |
13:45 | wajasu | still connecting based on "db" tht docker or the /etc/hosts may provide. |
13:47 | but granting the specific host we created from would support multiple hosts(1 to 1) (though not multiple koha-koha) | |
13:47 | i think once the multiple host scenarios kick in, we might need to use something like kubernetes. | |
13:49 | for say 3 koha web/app servers against one db server, or one web/app server supporting n sites against a specified db server. | |
13:51 | tcohen | wajasu: kubernetes is just a more complex way of doing the same thing |
13:51 | this specific deployment is using a single network (kohanet) | |
13:51 | for all services | |
13:51 | if you wanted to do something more complex, nothing prevents you from | |
13:51 | having a 'database' network | |
13:52 | granting based on the random IP that docker assigns a container | |
13:52 | could prevent you from things like 'scaling the backend' | |
13:53 | wajasu | ok. % is fine then. |
13:53 | tcohen | haha |
13:53 | this is just what we use for KTD, if you find we can do better | |
13:53 | help is always welcome! | |
13:54 | wajasu | dig koha +short just gives the IP from the web/app container for the mysql grant. but % works. |
13:54 | koha-jenkins | Project Koha_Master_D10 build #747: SUCCESS in 54 min: https://jenkins.koha-community[…]a_Master_D10/747/ |
13:54 | Project Koha_Master_D11_ES7 build #25: SUCCESS in 7 min 41 sec: https://jenkins.koha-community[…]aster_D11_ES7/25/ | |
14:04 | Project Koha_Master_U22 build #142: SUCCESS in 1 hr 7 min: https://jenkins.koha-community[…]a_Master_U22/142/ | |
14:16 | Oak joined #koha | |
14:29 | Oak joined #koha | |
14:29 | koha-jenkins | Project Koha_Master_D11_MDB_Latest build #1019: SUCCESS in 44 min: https://jenkins.koha-community[…]_MDB_Latest/1019/ |
14:30 | reiveune left #koha | |
14:48 | paulderscheid[m] | tcohen: Any plans to migrate to podman? |
14:49 | wajasu | we sure have alot of sources for db config. /etc/mysql/koha_<instance>.cnf , /etc/koha/passwd, /etc/koha/sites/<instance>/koha-conf.xml, ENV vars :) |
14:49 | tcohen | never thought about it |
14:50 | paulderscheid[m] | I want to create a containerized solution for LMSCloud as well and came to the conclusion that podman is the way to go because of better Integration with Kubernetes. |
14:50 | And it's really open-source open-source. | |
14:51 | wajasu | i am doing the same for Linode |
14:52 | paulderscheid[m] | No weird licensing stuff at a certain comany size |
14:52 | wajasu | i built images with packer, run libvirtd locally on qemu-kvm, but can also use packer to use plugin for linode and other clouds. |
14:53 | paulderscheid[m] | That's very cool wajasu |
14:53 | wajasu | i could use your docker images, or one built cloud friendly, cloud-init, or such. |
14:55 | i chained the packer images, debian, koha, so i wouldn't have to rebuild so often takes 30 minutes. not using overlays. | |
14:56 | you can mount a host directory for the koha git repo. | |
14:56 | i like the docker world though. | |
14:57 | my last inthe chain is my libraries db. | |
14:58 | your docker is just about the same. plus it supports windows. | |
14:59 | i think it might be worth generating a koha-debian-cloud.img that would work with cloud vendors. | |
15:00 | bag joined #koha | |
15:00 | wajasu | my base debian koha-common with my databse loaded is 4.8G |
15:01 | having the ability to keep the db across kul or such lets one run the updatedatabase.pl and debug failures for old sites. | |
15:02 | ktd helped me with the goal to load a 3.10 koha to 20.05, | |
15:02 | tcohen | why 20.05? |
15:03 | wajasu | because https://www.koha-community.org/ mentions it. i thought it was the latest. |
15:03 | tcohen | oh |
15:03 | upgrade to 21.11 right now :-P | |
15:04 | before they notice | |
15:04 | wajasu | i actually ran ktd which was master |
15:04 | tcohen | ah, you mean 22.05? |
15:04 | khall_ joined #koha | |
15:09 | wajasu | i told my librarian i would be deploying by friday. so i am working on auto_increment bug records now. |
15:09 | tcohen | oh |
15:09 | wajasu | 22.05 yes |
15:10 | tcohen | there's a script for that |
15:10 | wajasu | i actually got to 22.06 because thats what master was 2 weeks ago. |
15:11 | but got through FK issues, strict mariadb rules etc. | |
15:13 | i really need to do a diff between reference tables with codes that may not be present in my DB. i was missing circulation_rotating table and such. | |
15:14 | koha-jenkins | Project Koha_Master_D11_MDB_Latest build #1020: SUCCESS in 44 min: https://jenkins.koha-community[…]_MDB_Latest/1020/ |
15:14 | wajasu | i had done custom xslt, but you guys have fixed things so i can just rely on prog. |
15:15 | i also need to customize for school logo/colors. | |
15:17 | Oak joined #koha | |
15:19 | tcohen | wajasu: did you try the Galadriel plugin? |
15:47 | lukeg joined #koha | |
15:48 | lukeg | hi |
15:48 | wahanui | what's up, lukeg |
15:59 | cait1 | bye all |
16:07 | cait1 left #koha | |
16:36 | cait joined #koha | |
16:43 | khall joined #koha | |
16:51 | lukeg joined #koha | |
16:55 | Oak joined #koha | |
17:08 | wajasu | tcohen: i'll look at the Galadriel. |
17:09 | is there added benefit to using elastic search? i'm just monolithic server 1CPU 1GB ram in cloud right now. | |
17:11 | tcohen | 1GB RAM sounds like a bad thing |
17:12 | because each Plack worker will eat smth like 300MB just sitting there | |
17:19 | fridolin joined #koha | |
18:04 | wajasu | i noticed that memory growth. can koha run without plack? |
18:05 | memcached grew too | |
18:06 | tcohen | you don't want to run without Plack |
18:09 | wajasu | maybe a 2GB. |
18:09 | can plack or apache or memcached be configured with max | |
18:09 | tcohen | memcached can be configured to use, say, 64MB |
18:09 | and MariaDB needs to have the memory cache set to something reasonable | |
18:10 | yes, all of them | |
18:10 | wajasu | ok. will work it |
18:10 | tcohen | the default for Koha is to have 2 Plack workers running |
18:12 | I wouldn't go lower than that | |
18:12 | :-D | |
18:12 | @seen marcelr | |
18:12 | huginn | tcohen: marcelr was last seen in #koha 5 hours, 48 minutes, and 23 seconds ago: <marcelr> tcohen: all is fine for the lists patches now |
18:12 | tcohen | @later tell marcelr I added my two cents there, can you review it? |
18:12 | huginn | tcohen: The operation succeeded. |
18:13 | tcohen | 2 euros I'd say heh |
19:16 | Michelle joined #koha | |
19:34 | clms joined #koha | |
19:41 | fridolin1 joined #koha | |
20:57 | wajasu | tcohen: if you create a bug "add PS1 escape sequence to show current git branch in window title" we can discuss and add it to ktd. |
20:58 | somethings like: bug_2356: koha_testing_docker ... as it is today | |
21:47 | cait joined #koha |
← Previous day | Today | Next day → | Search | Index