| Time |
S |
Nick |
Message |
| 01:19 |
|
|
dcook_ joined #koha |
| 01:53 |
|
|
cait1 joined #koha |
| 03:20 |
|
|
alohabot joined #koha |
| 03:39 |
|
|
bag joined #koha |
| 04:55 |
|
|
alohabot joined #koha |
| 06:04 |
|
|
thibaud_g joined #koha |
| 06:06 |
|
|
reiveune joined #koha |
| 06:07 |
|
reiveune |
hello |
| 06:07 |
|
wahanui |
hello, reiveune |
| 06:35 |
|
krimsonkharne[m] |
morning #koha |
| 06:47 |
|
magnuse |
huh, it looks like oai-pmh is only showing deleted records? |
| 07:07 |
|
|
cait joined #koha |
| 07:35 |
|
ashimema |
Morning |
| 07:38 |
|
* ashimema |
looks at KITs firstv submission |
| 08:07 |
|
ashimema |
Sorry it took me a week or so to get to Michaela Sieber (BIB) 🙂 |
| 08:07 |
|
ashimema |
code looks good, just running through the test plan |
| 08:09 |
|
NikolayGospodinov[m] |
Hello! I have a question? Do you know if there is a plugin that export lists or statistics in doc format? |
| 08:20 |
|
magnuse |
nah, oai-pmh just shows deleted records first |
| 08:20 |
|
magnuse |
kit++ |
| 08:20 |
|
MichaelaSieberBIB[m] |
<ashimema> "code looks good, just running..." <- thanks martin |
| 08:21 |
|
magnuse |
which bug is it? |
| 08:22 |
|
MichaelaSieberBIB[m] |
https://bugs.koha-community.or[…]_bug.cgi?id=20755 |
| 08:22 |
|
huginn |
Bug 20755: enhancement, P5 - low, ---, raphael.straub, Needs Signoff , Allow separate email configuration for acquisition and serials emails |
| 08:24 |
|
ashimema |
hmm |
| 08:24 |
|
ashimema |
is ktd dead in the water today or is it just me? |
| 08:28 |
|
ashimema |
signed off. code looks solid, should be a quick QA.. I imagine you have a pet QA person lined up for that 😜 |
| 08:31 |
|
MatthewBlenkinsop[m] |
<ashimema> "is ktd dead in the water today..." <- I've just had to restart it, kept freezing whenever I tried a command in the shell |
| 08:31 |
|
ashimema |
I can't get it to start |
| 08:31 |
|
ashimema |
gets to almost there then crashes |
| 08:31 |
|
ashimema |
weird |
| 08:31 |
|
MatthewBlenkinsop[m] |
Mine is rebooting now, I'll see where it gets to |
| 08:32 |
|
Joubu |
what's the error? |
| 08:32 |
|
wahanui |
the error is a silent one which is not visible unless i enable verbose logging for zebra |
| 08:32 |
|
ashimema |
just scrolling back to find it |
| 08:32 |
|
ashimema |
ooh |
| 08:32 |
|
ashimema |
looks cypress related again |
| 08:33 |
|
MatthewBlenkinsop[m] |
Mine crashed |
| 08:33 |
|
ashimema |
ERROR in /kohadevbox/koha/cypress.config.ts(12,9) |
| 08:33 |
|
MatthewBlenkinsop[m] |
koha-koha-1 | Database is not empty! at /kohadevbox/misc4dev/populate_db.pl line 103. |
| 08:33 |
|
MatthewBlenkinsop[m] |
koha-koha-1 exited with code 1 |
| 08:34 |
|
pastebot |
"ashimema" at 127.0.0.1 pasted "ktd errors" (34 lines) at https://paste.koha-community.org/27986 |
| 08:34 |
|
MichaelaSieberBIB[m] |
<ashimema> "signed off. code looks solid..." <- Thank you so much martin! |
| 08:35 |
|
ashimema |
ho |
| 08:35 |
|
ashimema |
I'm confused.. somehow I'd got a node_modules dir in my host again |
| 08:36 |
|
Joubu |
database is not empty => you need to down before up |
| 08:36 |
|
ashimema |
owned by root too |
| 08:37 |
|
Joubu |
owned by root because you certainly run yarn install inside the container as root |
| 09:19 |
|
cait |
... pet QA person? *smiley that looks very critical* |
| 09:23 |
|
ashimema |
not critical |
| 09:25 |
|
ashimema |
it's good |
| 09:27 |
|
MichaelaSieberBIB[m] |
❤️ |
| 10:32 |
|
krimsonkharne[m] |
Nikolay Gospodinov: I don't know of any plugin to allow exporting in .doc format. You can find a list of Koha plugins here: |
| 10:32 |
|
krimsonkharne[m] |
https://wiki.koha-community.org/wiki/Koha_plugins |
| 10:33 |
|
oleonard |
kidclamp++ |
| 10:36 |
|
ashimema |
man my xslt is rusty |
| 10:37 |
|
oleonard |
No one should spend so much time in xslt that they're not rusty the next time :D |
| 10:37 |
|
ashimema |
thought bug 34020 might be relatively trivial |
| 10:37 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=34020 enhancement, P5 - low, ---, oleonard, NEW , Sequence of MARC 264 subfields different on XSLT result list and detail page |
| 10:38 |
|
|
magnuse joined #koha |
| 11:14 |
|
|
marcelr joined #koha |
| 11:14 |
|
marcelr |
hi #koha |
| 11:15 |
|
oleonard |
Writing a test plan when it takes longer than writing the patch: "If you want to test a patch, you must first create the universe..." |
| 11:16 |
|
tcohen[m] |
hola #koha o/ |
| 11:16 |
|
marcelr |
hola tcohen[m] |
| 11:17 |
|
marcelr |
tcohen[m]: could you look at 35041 please? |
| 11:18 |
|
tcohen[m] |
sure, I have it on my list. Trying to be as fair as possible |
| 11:20 |
|
marcelr |
its a tiny patch to resolve Jenkins error |
| 11:21 |
|
marcelr |
you want them green, right? |
| 11:23 |
|
ashimema |
meeting today? |
| 11:24 |
|
paulderscheid[m] |
yeah 1400 hours for us I think |
| 11:24 |
|
paulderscheid[m] |
12 UTC |
| 11:41 |
|
|
thd joined #koha |
| 11:43 |
|
magnuse |
so in 18 minutes? |
| 11:43 |
|
magnuse |
s/18/17/ |
| 11:44 |
|
oleonard |
s/17/16 |
| 11:44 |
|
thd |
what happened to the previous time pattern of 14 UTC? |
| 11:46 |
|
cait |
I am here :) |
| 11:48 |
|
thd |
Early for much of the US. |
| 11:49 |
|
oleonard |
Much of the US doesn't usually come to dev meetings |
| 11:51 |
|
thd |
Certainly, that is true. Much of the world does not so this time may not be terrible for the part of the US from which people appear. |
| 11:57 |
|
|
Adarsh joined #koha |
| 11:59 |
|
cait |
it's time... isnt't it? |
| 12:00 |
|
ashimema |
I thought so |
| 12:00 |
|
ashimema |
it's 9am for tcohen |
| 12:00 |
|
cait |
hm can someone else open? |
| 12:01 |
|
ashimema |
8am for kidclamp |
| 12:01 |
|
ashimema |
and 6am for much of the rest of bywater right now |
| 12:01 |
|
ashimema |
so.. yeah.. not really a great to time to catch them |
| 12:01 |
|
ashimema |
I agree.. not sure why it was moved earlier |
| 12:01 |
|
cait |
maybe a confusion about hte timezones with daylight savings? |
| 12:02 |
|
* ashimema |
would be available in an hour if people preffered.. wouldn't mind nipping out for a quick lunchtime walk right now |
| 12:02 |
|
ashimema |
I reckon so |
| 12:02 |
|
thd |
It was moved earlier because there was a conflict with the previously scheduled documentation meeting a few weeks ago. |
| 12:02 |
|
tcohen[m] |
Time for the meeting |
| 12:02 |
|
cait |
I think as there was a reminder on mailing list and all, we shoudl have it now |
| 12:02 |
|
thd |
The time change was not well remembered but stuck in any case. |
| 12:02 |
|
tcohen[m] |
I need 2 minutes |
| 12:03 |
|
* ashimema |
runs off for the dog walk.. will take phone |
| 12:05 |
|
cait |
it's just thd and me in the jitsi so far |
| 12:05 |
|
cait |
https://wiki.koha-community.or[…]g_18_October_2023 |
| 12:05 |
|
cait |
for link and info |
| 12:07 |
|
cait |
#startmeeting Development Meeting 18 October 2023 |
| 12:07 |
|
huginn |
Meeting started Wed Oct 18 12:07:50 2023 UTC. The chair is cait. Information about MeetBot at http://wiki.debian.org/MeetBot. |
| 12:07 |
|
huginn |
Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. |
| 12:07 |
|
huginn |
The meeting name has been set to 'development_meeting_18_october_2023' |
| 12:08 |
|
cait |
#chair tcohen |
| 12:08 |
|
huginn |
Current chairs: cait tcohen |
| 12:08 |
|
cait |
#info Introductions |
| 12:08 |
|
cait |
#topic Introductions |
| 12:08 |
|
tcohen[m] |
#info Tomas Cohen Arazi |
| 12:08 |
|
thd |
#info Thomas Dukleth, Agogme, New York City |
| 12:09 |
|
cait |
#info Katrin Fischer, BSZ, Germany |
| 12:09 |
|
paulderscheid[m] |
#info Paul Derscheid, LMSCloud GmbH, Germany |
| 12:09 |
|
marcelr |
#info Marcel de Rooy, Rijksmuseum |
| 12:09 |
|
MatthewBlenkinsop[m] |
#info Matt Blenkinsop, PTFS Europe, UK |
| 12:10 |
|
cait |
#topic Announcements |
| 12:11 |
|
cait |
#info User meeting for German speaking Koha users on 5th December, see https://koha-dach.eu |
| 12:11 |
|
cait |
#topic Update from the Release Manager |
| 12:11 |
|
cait |
#info QA has been filling up RM's queue - tcohen is trying to catch up. |
| 12:11 |
|
cait |
#info Please react fast if there are any questions, doesn't apply, follow-ups required etc. |
| 12:12 |
|
cait |
#info deadlines are approaching fast |
| 12:12 |
|
cait |
bug 29523 |
| 12:12 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=29523 major, P5 - low, ---, martin.renvoize, Signed Off , Add a way to prevent embedding objects that should not be allowed |
| 12:15 |
|
cait |
#info Preservation module and ERM usage were also mentioned explicitly |
| 12:18 |
|
cait |
#info discussed possible transcription of meetings, resurrecting the meeting factilitator as a Release Team role |
| 12:19 |
|
cait |
#topic Update from the Release Maintainers |
| 12:20 |
|
cait |
#info 22.11 in string freeze now |
| 12:20 |
|
cait |
#action MatthewBlenkinsop to get in touch with bgkriegel |
| 12:21 |
|
cait |
#action tcohen to publish freeze dates via mailing list |
| 12:24 |
|
cait |
#info there was an error in 21.11 that was fixed by tcohen (update issue) |
| 12:24 |
|
cait |
#topic Update from the QA team |
| 12:24 |
|
ashimema |
#info Martin Renvoize, PTFS Europe |
| 12:28 |
|
cait |
#info QA queue is shrinking, but still too big - hopefully we will be able to shrink it some more and get the older bugs out soon |
| 12:29 |
|
cait |
#topic Status of the roadmap projects |
| 12:29 |
|
cait |
#link https://wiki.koha-community.or[…]Roadmap_for_23.11 Rodmap in the wiki |
| 12:30 |
|
cait |
#info bug 17427 |
| 12:30 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=17427 normal, P5 - low, ---, koha-bugs, NEW , Replace CGI::Session with Data::Session |
| 12:31 |
|
cait |
#info bug 33537 is waiting for QA |
| 12:31 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=33537 enhancement, P5 - low, ---, m.de.rooy, Signed Off , Move domain limits from koha-conf to staff SMTP configuration |
| 12:32 |
|
cait |
#info bug 17427 will need to be postponed to next cycle, but would be great to have |
| 12:33 |
|
cait |
#info most of the roadmap is on the way for release |
| 12:36 |
|
cait |
#topic KohaCon25 update |
| 12:37 |
|
cait |
#info postponed! |
| 12:37 |
|
cait |
#topic Roles for 24.05 |
| 12:38 |
|
cait |
#link https://wiki.koha-community.or[…]i/Roles_for_24.05 |
| 12:38 |
|
cait |
#info Elections will be held next Dev Meeting |
| 12:39 |
|
cait |
#action tcohen will send an email about the roles to the mailing list |
| 12:39 |
|
cait |
#topic Actions from previous meetings |
| 12:41 |
|
|
emlam joined #koha |
| 12:46 |
|
cait |
#info perldoc still needs updating, need the script to run |
| 12:47 |
|
cait |
#action paulderscheid[m] to share his script for perldocs |
| 12:48 |
|
cait |
#info Move from Gitea to Forgejo moved to beginning of next cycle |
| 12:49 |
|
cait |
#topic General development discussion |
| 12:51 |
|
cait |
#info DMARC: thd got in touch with Rachel for Koha main list, they would be happy for someone to take over hosting |
| 12:51 |
|
cait |
#info Joubu could change settings on the Bblibre hosted lists |
| 12:53 |
|
cait |
#info bug 34927 |
| 12:53 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=34927 enhancement, P5 - low, ---, wizzyrea, NEW , Adding DMARC compatibility to mailing lists |
| 12:55 |
|
Joubu |
done |
| 12:55 |
|
cait |
you already made the changes? |
| 12:56 |
|
Joubu |
yes |
| 12:56 |
|
cait |
all or only koha-devel? |
| 12:56 |
|
Joubu |
koha-devel, see comment on the bug |
| 12:56 |
|
cait |
right now or a while ago? :) |
| 12:56 |
|
Joubu |
now |
| 12:57 |
|
marcelr |
Joubu: Should prefetch-count be activemq.prefetchSize in worker script? See https://metacpan.org/pod/Net::Stomp versus https://www.rabbitmq.com/stomp.html |
| 12:58 |
|
cait |
#info koha-devel is in testing now with the DMARC improved configuration |
| 12:58 |
|
cait |
#action thd to inform mailing list |
| 12:59 |
|
cait |
#info moving master to main to be done next cycle |
| 12:59 |
|
cait |
#info Draft style guide is still pending, modals info should be added |
| 13:00 |
|
cait |
#link https://wiki.koha-community.or[…]migration_weblate |
| 13:01 |
|
cait |
# Please visit the wiki page, there is bugs we need to have pushed/worked on and also open question on Weblate |
| 13:01 |
|
cait |
#topic Review of Coding guidelines |
| 13:01 |
|
Joubu |
for info koha-manual will move from master to main in the next couple of weeks |
| 13:01 |
|
cait |
#info no topics on agenda |
| 13:01 |
|
Joubu |
I need urgently bug 35079; bug 34959; bug 35043; bug 35024 to test the migration to weblate |
| 13:01 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=35079 enhancement, P5 - low, ---, julian.maurice, Signed Off , Add option to gulp tasks po:update and po:create to control if POT should be built |
| 13:02 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=34959 major, P5 - low, ---, jonathan.druart+koha, Passed QA , Translator tool generates too many changes |
| 13:02 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=35043 enhancement, P5 - low, ---, jonathan.druart+koha, Signed Off , Handling of \t in PO files is confusing |
| 13:02 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=35024 normal, P5 - low, ---, jonathan.druart+koha, Needs Signoff , Do no wrap PO files |
| 13:02 |
|
cait |
#info koha-manual will move from master to main in a couple of weeks |
| 13:02 |
|
cait |
#Set time of next meeting |
| 13:02 |
|
marcelr |
Joubu will look Friday |
| 13:05 |
|
cait |
#info Next meeting: 1 November 2023, 13 UTC |
| 13:13 |
|
cait |
#endmeeting |
| 13:13 |
|
huginn |
Meeting ended Wed Oct 18 13:13:45 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) |
| 13:13 |
|
huginn |
Minutes: https://meetings.koha-communit[…]-10-18-12.07.html |
| 13:13 |
|
huginn |
Minutes (text): https://meetings.koha-communit[…]3-10-18-12.07.txt |
| 13:13 |
|
huginn |
Log: https://meetings.koha-communit[…]18-12.07.log.html |
| 13:16 |
|
cait |
Joubu: could you run meeting script or shoudl I ask someone else? |
| 13:16 |
|
cait |
format is hopefully so it just works |
| 13:18 |
|
thd |
#info May have missed the meeting notes pauldersheid reported that website prototype source is at https://github.com/pders01/koha-community.org . |
| 13:19 |
|
Karter[m] |
Does anyone use the nederlands language pack with their koha version 22.11 . We have a translation problem and I would like to see if it is a common issue as the previous translation file was meaning to translate 'Search Catalog' but it did not. And the newer file at this link : https://gitlab.com/koha-community/koha-l10n - also has the 'search Catalog' translated but on different lines ( which as stated earlier this is not exact |
| 13:19 |
|
Karter[m] |
identification ) but for both translations files to not be translating certain strings is concerning for myself. I have followed the guide on trying to update the file but nothing actually changes on my server. --- Question: 1: Does anyone else with nederlands language pack have the same issues? 2: Regarding Manual updating of the translation files, Is it just the nl-NL-staff-prog.po file we update or is it every file related to nl-NL we need |
| 13:19 |
|
Karter[m] |
to replace, both directory, .mo file AND .po files in the /po directory? |
| 13:19 |
|
* Karter[m] |
uploaded an image: (66KiB) < https://matrix.org/_matrix/med[…]guagepack-bug.png > |
| 13:21 |
|
thd |
s/(paulders)(heid)/$1c$2/ |
| 13:23 |
|
caroline |
Karter, does your translation file have "fuzzy" before Search catalog? |
| 13:23 |
|
caroline |
https://gitlab.com/koha-commun[…]_type=heads#L4326 |
| 13:23 |
|
caroline |
You need to remove that, otherwise it will not appear as translated |
| 13:24 |
|
Karter[m] |
caroline: Thankyou , yes it does, I found somewhere fuzzy was a different identifier woopsie |
| 13:25 |
|
caroline |
I think fuzzy is when the string is "automatically" translated, but needs a human to look at it |
| 13:26 |
|
Joubu |
cait: the script does not work, everything is messy |
| 13:33 |
|
Karter[m] |
<caroline> "I think fuzzy is when the string..." <- hero this was the problem ! I understood fuzzy as being computer generated for there might be conflicts of language and could be wrong translations and not meaning human check needed. Removing Fuzzy worked straight away |
| 13:34 |
|
thd |
Joubu: Do you have both of these? dmarc_is_list set to "Munge From". |
| 13:34 |
|
thd |
dmarc_moderation_action set to "Munge From". |
| 13:34 |
|
Joubu |
There is only one |
| 13:35 |
|
thd |
Joubu: They may be in different sections |
| 13:36 |
|
Joubu |
ok found, done |
| 13:38 |
|
thd |
The second is only an issue if the moderation queue is triggered by too many recipients listed or something, but is needed for moderator approved messages. |
| 13:55 |
|
cait |
Joubu: argh. but I looked up the format and all |
| 13:55 |
|
cait |
Joubu: how do we fix? |
| 13:56 |
|
Joubu |
I don't know, manually |
| 13:57 |
|
Joubu |
I had to redirect the wiki page, because the title was not correct (was missing "IRC") |
| 13:57 |
|
Joubu |
but then the script is failing, I guess it does not handle redirect |
| 13:57 |
|
tcohen[m] |
Joubu: would you agree xt/author/valid-templates.t failing is a false positive |
| 13:58 |
|
tcohen[m] |
it feels silly to add an include on an include if the only template using the include, contains the include |
| 13:58 |
|
tcohen[m] |
heh |
| 13:58 |
|
Joubu |
why that? |
| 13:59 |
|
Joubu |
if the include needs another include, then include it from there, not from the "parent" |
| 14:00 |
|
Joubu |
if you want to reuse the include from somewhere else you will have to know that you need to include another include..? |
| 14:00 |
|
Joubu |
if you want to keep like that you will have to add an exception to the test + we will lose the validation on this include |
| 14:00 |
|
Joubu |
well, just add the include :D |
| 14:02 |
|
marcelr |
kidclamp: care to have a look at 32305 ? |
| 14:02 |
|
kidclamp |
will do |
| 14:03 |
|
marcelr |
top |
| 14:03 |
|
ashimema |
bug 32305 |
| 14:03 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=32305 major, P5 - low, ---, m.de.rooy, Needs Signoff , Background worker doesn't check job status when received from rabbitmq |
| 14:04 |
|
Joubu |
marcelr: I think the correct place was in the Koha::BackgroundJob->process method |
| 14:04 |
|
marcelr |
I think we should solve the problem now :) |
| 14:05 |
|
Joubu |
"We should do it too for es_indexer_daemon.pl": and this is showing I am right |
| 14:05 |
|
marcelr |
that script should have been merged with regular one |
| 14:05 |
|
marcelr |
this is redundant code |
| 14:05 |
|
Joubu |
now, yes, but the original patch is 1y old ;) What was the problem exactly? |
| 14:06 |
|
ashimema |
lol |
| 14:06 |
|
Joubu |
no, read the history, it was not trivial to have a single script |
| 14:06 |
|
ashimema |
merged... |
| 14:06 |
|
ashimema |
wasn't it split out |
| 14:06 |
|
marcelr |
there are several issues with background jobs and you have no time ? |
| 14:06 |
|
Joubu |
I... was working on it 1 year ago... |
| 14:06 |
|
kidclamp |
The es daemon can combine jobs, and we have a lot of them, whoever split that out was a smart guy ;-) |
| 14:06 |
|
marcelr |
the hybrid approach with mq and db created a lot of confusion |
| 14:07 |
|
Joubu |
now you wake up and I don't have time right now, but I will be happy to help later if we coordinate and focus on it for real |
| 14:07 |
|
marcelr |
it needs a decision on db or mq and then refactoring |
| 14:07 |
|
Joubu |
fix the existing bug first, before thinking refactoring |
| 14:07 |
|
ashimema |
we still turn off rabbit here.. it's just never been reliable enough |
| 14:08 |
|
marcelr |
what we do now, we could as well remove rabbitmq |
| 14:08 |
|
ashimema |
maybe it's better now.. but the wound is still fresh |
| 14:08 |
|
Joubu |
bug 32395 for instance |
| 14:08 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=32395 enhancement, P5 - low, ---, koha-bugs, In Discussion , Why do we send jobs params to the broker |
| 14:08 |
|
marcelr |
yes thats the underlying issue perhaps |
| 14:08 |
|
* oleonard |
buys a round for all the smart guys |
| 14:08 |
|
kidclamp |
we keep testing rabbit after upgrades, but we haven't yet been successful at running it without issues |
| 14:09 |
|
Joubu |
which issues? |
| 14:10 |
|
kidclamp |
Liz would have more details - but not keeping up, then jobs being lost - I think 34990 may help |
| 14:10 |
|
caroline |
we don't use rabbit here either, I thought we were outliers, but it seems not |
| 14:10 |
|
marcelr |
34990 and 32305 are small and fix some issues |
| 14:10 |
|
ashimema |
School run, bbiab |
| 14:11 |
|
marcelr |
but removing all orhpaned jobs requires decision and refactoring |
| 14:11 |
|
Joubu |
either we remove or we fix, but keeping it with nobody using it is silly |
| 14:11 |
|
marcelr |
yes, the current code begs for removing actually ; it makes no sense anymore |
| 14:11 |
|
kidclamp |
we want to use it, we have also been working on other things, so haven't had the time to test and report issues |
| 14:11 |
|
marcelr |
we use MQ but we dont trust it |
| 14:12 |
|
Joubu |
RabbitMQ does not lose jobs, it's impossible. We are barely using it and it can handle thousands of requests |
| 14:13 |
|
Joubu |
if something wrong is happening it's on Koha side, or prod env |
| 14:13 |
|
marcelr |
bug 34997 |
| 14:13 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=34997 normal, P5 - low, ---, koha-bugs, NEW , Background jobs: Orphaned jobs should not be run after too much time |
| 14:13 |
|
marcelr |
Joubu you do not persist so mq loses jobs |
| 14:13 |
|
marcelr |
at restart |
| 14:13 |
|
marcelr |
and you should be very careful in restarting rabbit before worker |
| 14:14 |
|
Joubu |
marcelr: count me in if you have plans and need help |
| 14:15 |
|
marcelr |
we just need now 34990 and 32305 and then decide on rabbit for 34997 and friends |
| 14:17 |
|
Joubu |
there are more from last year |
| 14:18 |
|
marcelr |
we need an omnibus |
| 14:18 |
|
marcelr |
we might have dups |
| 14:19 |
|
kidclamp |
bug 35086 - i could use opinions on where/how to configure and commit the batches |
| 14:19 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=35086 normal, P5 - low, ---, koha-bugs, NEW , Koha::SearchEngine::Elasticsearch::Indexer->update_index needs to commit in batches |
| 14:22 |
|
marcelr |
is the array growing too large? |
| 14:22 |
|
marcelr |
$biblio_indexer->update_index( \@bib_records ); |
| 14:22 |
|
Joubu |
shouldn't it send several requests, depending on the max size? |
| 14:22 |
|
|
blou joined #koha |
| 14:23 |
|
marcelr |
es_indexer does something strange |
| 14:23 |
|
marcelr |
@jobs = Koha::BackgroundJobs->search followed by Koha::BackgroundJobs->search( { id => [ map { $_->id } @jobs ] }); |
| 14:23 |
|
huginn |
marcelr: downloading the Perl source |
| 14:23 |
|
marcelr |
we are refetching in commit() |
| 14:24 |
|
Joubu |
that's why we do not have a single script ;) |
| 14:24 |
|
Joubu |
it's doing things its own way |
| 14:25 |
|
marcelr |
it should be moved to a BackgroundJob subclass ? |
| 14:25 |
|
Joubu |
See the 2 ->update calls |
| 14:25 |
|
marcelr |
and be a bit smarter |
| 14:25 |
|
Joubu |
we needed a quick fix at the time, and it was good enough to be pushed and tested |
| 14:26 |
|
marcelr |
we can tidy later lol |
| 14:26 |
|
Joubu |
"we can make it reusable later if we need" |
| 14:26 |
|
Joubu |
do we need now? not yet ;) |
| 14:27 |
|
Joubu |
But you are not supposed to have basic MQ connect logic there, what your patches do... |
| 14:27 |
|
marcelr |
we did not create a MQ module ? |
| 14:28 |
|
marcelr |
we should have done |
| 14:28 |
|
Joubu |
done what? |
| 14:28 |
|
marcelr |
we put something inside in BackgroundJob where it does not belong ? |
| 14:28 |
|
marcelr |
BackgroundJob->connext |
| 14:28 |
|
marcelr |
connect |
| 14:28 |
|
kidclamp |
indexer daemon aside, should the background job split the batches into several requests? Actually, I think it should be done in the indexer - decide a max batch size for the requests - then we can move the reindex commit option into a config |
| 14:29 |
|
marcelr |
a MQ module should have a connect and a fetch jobs |
| 14:29 |
|
Joubu |
marcelr: yes, agreed |
| 14:29 |
|
marcelr |
kidclamp: seems you just define a limit to split upon |
| 14:30 |
|
marcelr |
and improve all the stuff along the way :) |
| 14:30 |
|
Joubu |
kidclamp: yes it should be ES side, it's not specific to the background jobs. We should share the ES config with the indexer module |
| 14:30 |
|
kidclamp |
indeed, I just wasn't sure where, but I have an opinion that will at least last through today for now |
| 14:30 |
|
marcelr |
lol |
| 14:33 |
|
|
blou joined #koha |
| 14:33 |
|
marcelr |
created omnibus bug 35092 now for background job issues |
| 14:33 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=35092 enhancement, P5 - low, ---, koha-bugs, NEW , [OMNIBUS] Remaining background job/worker issues |
| 14:36 |
|
Joubu |
kidclamp: are we talking about http.max_content_length ? |
| 14:36 |
|
Joubu |
"Defaults to 100mb", do you really reach that? |
| 14:37 |
|
kidclamp |
Yes, on sites with large bibs (i.e. lots of items) |
| 14:37 |
|
kidclamp |
They did an update of like 40k bibs |
| 14:38 |
|
kidclamp |
and the es_indexer combines jobs, and we had a situation where the worker died - when restarted we ended up combining a bunch of big jobs into a single giant index |
| 14:38 |
|
Joubu |
kidclamp: could we fetch ES config automatically on "Koha startup" and cache it? |
| 14:38 |
|
kidclamp |
but also below that we have had timeout issues on large requests |
| 14:39 |
|
Joubu |
timeout from ES then? |
| 14:40 |
|
kidclamp |
yes to timeouts and yes to caching the config |
| 14:40 |
|
kidclamp |
don't we, since it is in the koha-conf? |
| 14:40 |
|
Joubu |
kidclamp: I am suggesting to retrieve ES config from ES REST API |
| 14:43 |
|
Joubu |
marcelr: I still don't understand why we don't skip the job in ->start if status ne 'new' |
| 14:44 |
|
marcelr |
->start checks the status iirc |
| 14:44 |
|
marcelr |
but we should not get there in the first place |
| 14:44 |
|
marcelr |
the initial selection should be correct |
| 14:44 |
|
Joubu |
then you need a method in Koha::BackgroundJobs |
| 14:45 |
|
marcelr |
note that the worker now sets jobs to failed while it should not touch them |
| 14:45 |
|
Joubu |
a filter_by method |
| 14:45 |
|
marcelr |
yes i agree that a filter would be good |
| 14:45 |
|
marcelr |
but we are on bugfix level now |
| 14:45 |
|
Joubu |
lol :D |
| 14:45 |
|
marcelr |
the same argument in another package ;) |
| 14:46 |
|
marcelr |
the number of bugs in the omnibus is getting interesting |
| 14:52 |
|
PedroAmorim[m] |
jajm: around? |
| 14:55 |
|
jajm |
PedroAmorim[m]: yes |
| 14:57 |
|
marcelr |
ok i'm going to the background now; bye |
| 14:59 |
|
cait |
Joubu: so I should create the page manually |
| 14:59 |
|
cait |
and need to make sure it includes IRC? |
| 15:03 |
|
|
alohabot` joined #koha |
| 15:05 |
|
Joubu |
yes |
| 15:06 |
|
Joubu |
cait: yes |
| 15:09 |
|
cait |
I'll try my best |
| 15:09 |
|
reiveune |
bye |
| 15:09 |
|
|
reiveune left #koha |
| 15:13 |
|
cait |
I think the IRC was removed because we use Jitsi now - not knowing it woudl affect the script |
| 15:14 |
|
cait |
shoudl we adapt the script? |
| 15:15 |
|
cait |
Joubu: ^ |
| 16:06 |
|
cait |
Joubu: I'll wait for your reply and creat the wiki page tomorrow |
| 16:08 |
|
cait |
bye #koha |
| 16:08 |
|
|
cait left #koha |
| 16:38 |
|
|
cait joined #koha |
| 16:59 |
|
oleonard |
kidclamp still around? |
| 16:59 |
|
* kidclamp |
waves |
| 17:00 |
|
oleonard |
Hey... With Bug 34014, should I be able to view record 369? |
| 17:00 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=34014 normal, P5 - low, ---, nick, Signed Off , There is no way to fix records with broken MARCXML |
| 17:02 |
|
kidclamp |
in staff client, yes, with a note that it is degraded |
| 17:02 |
|
kidclamp |
that should be true before the patch actually |
| 17:07 |
|
oleonard |
Hm.... I don't find that to be so. I wonder why |
| 17:09 |
|
kidclamp |
do you get an error in the logs? |
| 17:10 |
|
kidclamp |
or a 500 in the interface? |
| 17:10 |
|
oleonard |
"Error trace Invalid data, cannot decode metadata object (biblio_metadata.id=368, biblionumber=369, format=marcxml..." etc. in the interface |
| 17:14 |
|
kidclamp |
simply apply patch, restart all, and then go to the url? I'll try again |
| 17:16 |
|
kidclamp |
oh, now I see that, looking, thanks oleonard |
| 17:26 |
|
kidclamp |
ashimema broke it :-) bug 26314 |
| 17:26 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=26314 enhancement, P5 - low, ---, martin.renvoize, Pushed to master , "Volumes: show volumes" showing regardless of whether there are volumes linked to the record |
| 17:27 |
|
ashimema |
What did I break now? |
| 17:28 |
|
oleonard |
Don't listen to him ashimema, what does he know |
| 17:28 |
|
kidclamp |
record 369 won't load in Koha, because get_volumes_query gets the biblio->metadata->record and doesn't catch the exception |
| 17:29 |
|
ashimema |
Damn |
| 17:30 |
|
ashimema |
Sorry |
| 17:30 |
|
ashimema |
I missed that in the rebases |
| 17:30 |
|
kidclamp |
I mean, i don't blame you, it's the tension of "we shouldn't allow bad records" vs "but we have bad records" |
| 17:30 |
|
ashimema |
We should write a test for that record |
| 17:30 |
|
kidclamp |
+1 |
| 17:30 |
|
ashimema |
Indeed |
| 17:31 |
|
kidclamp |
I thought Joubu said we did have a test for loading degraded records |
| 17:34 |
|
kidclamp |
bug 35099 filed and assigned to you ashimema |
| 17:34 |
|
huginn |
Bug https://bugs.koha-community.or[…]_bug.cgi?id=35099 normal, P5 - low, ---, martin.renvoize, NEW , Cannot load records with invalid marcxml |
| 17:34 |
|
ashimema |
Ta |
| 17:35 |
|
ashimema |
Will take a look in the morning |
| 18:46 |
|
tcohen[m] |
oleonard: around? |
| 18:46 |
|
tcohen[m] |
please rebase 34392 |
| 18:46 |
|
tcohen[m] |
like, now |
| 18:47 |
|
oleonard |
Ugh XD |
| 18:47 |
|
tcohen[m] |
I forgot about it on my last batch... sorry |
| 18:52 |
|
oleonard |
tcohen[m]: Done |
| 19:22 |
|
* cait |
waves |
| 19:26 |
|
davidnind |
Apologies for screwing up the meeting script by changing the development meeting name 8-; Happy to do manually in the meantime... |
| 19:50 |
|
cait |
davidnind: did Joubu reply? i was wondering if we should adjust wiki or script |
| 19:57 |
|
davidnind |
I didn't see something, will have a look through the logs again - although not sure the name should matter as it is used for other meetings |
| 20:15 |
|
cait |
i think the script parses it somehow to create the links or so? |
| 20:15 |
|
cait |
maybe the last meeting section |
| 20:35 |
|
|
fridolin joined #koha |
| 20:38 |
|
davidnind |
thnaks cait - I'll have a look shortly, will be back in about an hour or so... |
| 23:06 |
|
|
AmmarSPV joined #koha |
| 23:06 |
|
AmmarSPV |
Hi |
| 23:06 |
|
AmmarSPV |
I am looking for Arabic publisher book library targets to import MARC records. Any advise? |
| 23:26 |
|
|
alohabot joined #koha |
| 23:37 |
|
davidnind |
AmmarSPVL: I don't have any ideas. Maybe ask on the General Mailing List, as that has a wider audience. |
