IRC log for #koha, 2022-08-01

All times shown according to UTC.

Time S Nick Message
02:19 wajasu i've got my new debian11 with 22.0503000 in the db syspref and
02:20 wajasu the opac always disconnects the connection at the first search.  the staff interface works.
02:21 wajasu i've also just install that same db with KOHA_IMAGE=22.05 kul , koha-upgrade-schema, koha-rebuild-zebra
02:23 wajasu that docker environment has in the db syspref and
02:26 wajasu both the opac and staff succeed in the search in the docker ktd environment.
02:27 wajasu now i want to see if its the specific version 22.0503000 that is broken, or if its my debian11 install.
02:31 wajasu i guess i can try to change my koha git tag to v22.05.02, then KOHA_IMAGE=22.05 kul
02:34 wajasu then the kul docker environment would be as close to my failing installation.  if it succeeds, then 22.0503000 is broken.
02:35 wajasu if it fails, its something else.  i'll have to debug  maybe use debugger, and step through.
02:36 wajasu or look to see if the database will log a statement for a query ...
02:36 wajasu tmrw morning.
02:37 wajasu at least i know my database in the ktd/kul works at
03:55 alohabot joined #koha
05:29 dpk joined #koha
06:20 dpk joined #koha
06:21 reiveune joined #koha
07:01 cait joined #koha
07:05 cait1 joined #koha
07:07 cait1 hm not a lot of ops again
07:07 paulderscheid[m] Morning #koha
07:11 cait1 it looks like mails going through the mailing list can take a few hours at the moment, in one case an email from koha-bugs took almost 24-hours. We've ruled out bugzilla and I also saw a bigger delay on koha-devel with an email I sent - someone from Biblibre around who could check on the mailing list server?
07:13 koha-jenkins Project Koha_21.11_D12 build #103: ABORTED in 39 sec: https://jenkins.koha-community[…]ha_21.11_D12/103/
07:14 cait1 Frido, jajm ^
07:55 cait joined #koha
08:03 alohabot joined #koha
08:08 alohabot joined #koha
09:44 ib_iel joined #koha
10:10 cait joined #koha
10:17 oleonard joined #koha
10:39 oleonard joined #koha
11:25 tcohen hola #koha
11:31 cait1 hola tcohen
11:36 tcohen bye koha-jenkins
11:44 tcohen hi all, anyone willing to check on the jenkins build notifications, can join the #koha-ci channel
11:44 tcohen as decided on some past dev meeting, koha-jenkins will no longer report here
11:51 tuxayo tcohen++
11:51 oleonard Thanks tcohen!
11:54 tcohen remote: Compressing objects: 100% (1287/1287), done.
11:54 tcohen Recibiendo objetos:  44% (3853/8692), 55.82 MiB | 56.00 KiB/s
11:55 tcohen Work will be slow today it seems
12:14 huginn` News from kohagit: Bug 30965: Add a 'x' to go back to the patron search <[…]6a5a282e2d19627a8>
12:14 huginn` News from kohagit: Bug 30650: Add missing koha_object[s]_class methods <[…]0d62c99a23b9629bc>
12:14 huginn` News from kohagit: Bug 30499: Move the include for in to correc... <[…]3c9593bc5bc7e2193>
12:14 huginn` News from kohagit: Bug 31251: Use jQuery val() to clear_entry <[…]2f1be9f87dac4614d>
12:14 huginn` News from kohagit: Bug 31223: Regression tests <[…]8d80ea02283a7d55b>
12:14 huginn` News from kohagit: Bug 31223: Do not explode if plugins disabled <[…]5945467931a10082e>
12:14 huginn` News from kohagit: Bug 30965: Add patron autocomplete search to curbside pickups <[…]be3cbeab174f8cec3>
12:14 huginn` News from kohagit: Bug 30028: Fix patron message delete confirmation to be translatable <[…]e1a91a75c08fe30c8>
12:14 huginn` News from kohagit: Bug 28707: Translate RECEIPT notice <[…]42bcc0af1868017c5>
12:18 tcohen lukeg++ # fixing stuffs all over the place
12:32 huginn` News from kohagit: Bug 30984: DBRev <[…]2a332f328cad515e7>
12:32 huginn` News from kohagit: Bug 30468: Restore 'set -e' and remove echos <[…]ab880d81b8f51d2e3>
12:32 huginn` News from kohagit: Bug 31269: Define sDefaultContent for DataTables <[…]d566adc5bac7f293d>
12:32 huginn` News from kohagit: Bug 30984: (QA follow-up) Switch from mediumtext to varchar(255) <[…]609cf05e360551ce1>
12:32 huginn` News from kohagit: Bug 30984: (QA follow-up) Ternary formatting style <[…]2b4cf23763363b45a>
12:32 huginn` News from kohagit: Bug 30984: Action logs should log the cronjob script name that generated the given log <[…]7484d9e989ed3e5f7>
12:32 huginn` News from kohagit: Bug 30658: (bug 29496 follow-up) CheckMandatorySubfields don't work properly with... <[…]e3e8d76bea3108125>
12:32 huginn` News from kohagit: Bug 30468: koha-mysql does not honor Koha's timezone setting <[…]43c0f45077871b398>
12:32 oleonard Anyone know why I'm getting "Cannot connect to the Docker daemon at unix:///var/run/docker.sock" today when I try to start KTD? macOS, Docker Desktop v4.10.1
12:33 Joubu is docker running?
12:33 oleonard Yes,
12:34 Joubu Try a restart maybe: `service docker restart`
12:35 tcohen oleonard: maybe you have Docker Desktop AND some manually installed tools from homebrew?
12:35 tcohen i.e. version mismatch
12:35 tcohen it is working for me
12:35 tcohen same version
12:35 caroline joined #koha
12:35 tcohen was about to upgrade to 4.11 on friday
12:35 tcohen I suggest you restart it
12:36 oleonard Restarting a third time was what I needed I guess :|
12:37 Joubu you need to restart it and feel it will fix the problem. If you just restart it without faith, it cannot work
12:38 tcohen Joubu: the Docker Desktop thing... is a bit fragile.
12:39 tcohen it uses a VM and restarting it fixed things in the past
12:39 tcohen this is not Linux
12:39 tcohen and I agree: you need to believeeeeee
12:47 caroline joined #koha
12:53 cait1 tcohen: did you see my note on your holds bug? I can no longer set holds to waiting :(
12:55 caroline joined #koha
13:02 tcohen cait1: saw it, working on that after the push I'm doing
13:02 tcohen thanks for the reminder!
13:02 cait joined #koha
13:02 tcohen it must have been a bad rebase
13:04 huginn` News from kohagit: Bug 30984: DBIC update <[…]9435436e51b519ff1>
13:04 huginn` News from kohagit: Bug 30612: Rename other occurrences <[…]5e823d9bda87f1efc>
13:04 huginn` News from kohagit: Bug 30612: (QA follow-up) Rename accountlines method to account_lines <[…]910f334871231ab0a>
13:04 huginn` News from kohagit: Bug 27849: (follow-up) Add test <[…]54b2375b994753951>
13:04 huginn` News from kohagit: Bug 30964: (QA follow-up) Add classes to allow for additional styling <[…]5c0afd3bfb42a3de0>
13:04 huginn` News from kohagit: Bug 30964: Add 'restricted' and 'has overdues' on curbside pickup <[…]8cb19bb5ef438cc3b>
13:04 huginn` News from kohagit: Bug 30612: Update schema files <[…]820d9081f10256df8>
13:04 huginn` News from kohagit: Bug 30612: Add accountlines method to Koha::Checkout and Koha::Old::Checkout <[…]2a4339b4b1cf02637>
13:04 huginn` News from kohagit: Bug 28864: Have patron card creator use PatronsPerPage <[…]43da0adb423ad7bd8>
13:04 huginn` News from kohagit: Bug 27849: Koha::Token may access undefined C4::Context->userenv <[…]7df0bb94b4cfbbb12>
13:19 Dyrcona joined #koha
13:36 FlorianB joined #koha
13:47 kidclamp joined #koha
14:02 cait1 zero emails on koha-bugsfor some time now :(
14:20 lukeg joined #koha
14:20 lukeg hi
14:20 wahanui privet, lukeg
14:25 wajasu i believe
14:25 * cait1 waves
14:39 tundunf joined #koha
14:39 tundunf hi everyone
14:44 oleonard Hi tundunf
14:48 reiveune bye
14:48 reiveune left #koha
14:50 lukeg joined #koha
14:55 tundunf I can find instructions for adding item checkin alerts, but does anyone know if it's possible to add item checkout alerts?
14:57 Joubu like sending a notification by email?
14:58 tundunf More like an alert for staff issuing the item to borrowers
14:58 tundunf For example, we have a restricted loan collection that doesn't auto-renew, and we wanted to see if we could have an alert pop-up when it's being issued to remind circulation staff to tell borrowers that the item is non-renewable
15:01 cait1 tundunf: have a look at $3 materials
15:01 cait1 that creates some warning on checkout
15:01 cait1 and checkin
15:01 tundunf ah thank you
15:02 tundunf I'm assuming there's no way to apply this to an entire item type?
15:02 Joubu bug 25261
15:02 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=25261 enhancement, P5 - low, ---, martin.renvoize, CLOSED FIXED, Multiple parts handling - confirmation alert
15:04 cait1 Joubu: that bug makes it a confirm yep - showed before that too
15:05 cait1 tundunf: maybe it would be a good idea to file a bug to have something like the checkin note also for checkouts (by itemtype)
15:05 cait1 it could make a nice enh
15:06 tundunf will do!
15:06 tundunf thanks for your help all ~
15:06 FlorianB I'm having some issues with the kohadevbox, can I ask for help here as well?
15:17 cait1 FlorianB: you can
15:18 cait1 FlorianB: but a lot of people are using koha-testing-docker these days, so not sure when ti comes to fixes etc
16:16 cait1 left #koha
16:18 cait joined #koha
16:23 m23 joined #koha
16:36 buckleshoe joined #koha
16:50 m23 joined #koha
17:15 huginn` News from kohagit: Bug 30077: DBRev <[…]5349261a3c2b54ac8>
17:15 huginn` News from kohagit: Bug 30077: Compiled CSS <[…]185cf57cb92d9cf49>
17:15 huginn` News from kohagit: Bug 23919: DBRev <[…]13ee492ef68e27a47>
17:15 huginn` News from kohagit: Bug 23919: (QA follow-up) Tidy query build code <[…]11a0bf788900c369d>
17:15 huginn` News from kohagit: Bug 23919: Add tests for Items search by ISBN and ISSN with variations <[…]aa29e1d24463c5bd7>
17:15 huginn` News from kohagit: Bug 23919: Add tests for biblioItems search by ISBN and ISSN with <[…]40604d299db445c8a>
17:15 huginn` News from kohagit: Bug 23919: Items search by ISBN variations and ISSN variations <[…]f941670a49ec06925>
17:15 huginn` News from kohagit: Bug 23919: Zebra search by ISSN variations <[…]b5caa9c99bb30331e>
17:15 huginn` News from kohagit: Bug 23919: Add new system preference SearchWithISSNVariations <[…]12b5bcca0246777f1>
17:15 huginn` News from kohagit: Bug 30077: (follow-up) Set max width for library drapdown <[…]9d907ebcb44bfa06a>
17:15 huginn` News from kohagit: Bug 30077: (follow-up) retain branch selection <[…]d511a126cf659bc49>
17:15 huginn` News from kohagit: Bug 30077: Add system preference IntranetAddMastheadLibraryPulldown <[…]fd85a3e1fdf0cfdef>
17:39 huginn` News from kohagit: Bug 30392: DBRev <[…]b6a3ea7cd720584c1>
17:39 huginn` News from kohagit: Bug 30392: DBIC update <[…]77848bb889d5c97b7>
17:39 huginn` News from kohagit: Bug 30392: (QA follow-up) Correct COMMENT <[…]a32e33c7a4ce24fc1>
17:39 huginn` News from kohagit: Bug 29579: (QA follow-up) Fix up t/db_dependent/Koha/Reports.t <[…]658e3cd8bb48930bf>
17:39 huginn` News from kohagit: Bug 30392: Set deleted_on when an item is (marked as) deleted <[…]e0495a6720d2bbfcc>
17:39 huginn` News from kohagit: Bug 30392: DB changes <[…]b57698acae5d0827d>
17:39 huginn` News from kohagit: Bug 29579: Show saved SQL report ID in database query <[…]f901e2d63a6c97192>
17:49 huginn` News from kohagit: Bug 30077: (QA follow-up) Avoid warning <[…]7b232d2ceb0aa61e7>
17:53 wajasu i need some help. i need a hint as where to look for my problem.
17:53 wajasu fresh koha debian 11 install, but with my legacy database brought in.
17:54 tcohen go on
17:54 tcohen we are all reading
17:54 tcohen :-D
17:55 wajasu the database has 22.0502000 version in sys prefs
17:55 wajasu my problem is when i do a search in OPAC, i get connection timeout in browser
17:55 wajasu staff interface works.
17:56 tcohen is the request reaching Apache?
17:56 wajasu i did load that same database into ktd kul docker env at version 22.0503000 and it works
17:56 tcohen (i.e. in the logs)
17:56 wajasu letme check
17:56 tcohen it feels like a deployment issue
17:56 tcohen maybe NS, etc
17:59 wajasu i just did a bunch of apt update upgrade stuff.
18:00 wajasu it works now.  let me reboot the server
18:00 wajasu when i reboot, does zebra always rebuild?
18:03 wajasu i reboot.  now the opac comes up, i search and i get dropped connection. when i search
18:05 wajasu its workingnow.
18:07 wajasu my staff interface always works with search.  let me reboot and see if some backend services weren't up yet.
18:07 tcohen zebra doesn't rebuild on its own
18:08 wajasu but i was usually waiting for an hour...
18:08 tcohen unless you're talking about the real-time updating that takes place on changes
18:08 tcohen which is handled by koha-indexer
18:08 wajasu no updateing.  just plain search out the box.
18:08 wajasu not this is on a 4GB linode shared VPS.
18:09 wajasu not sure if other users can influence performance on my kvm
18:10 wajasu rebooting again
18:17 wajasu well if i change my search criteria, i saw the result pae come up, but maybe cover resource lookup are delayed.
18:22 wajasu ok.  if i search for author.  i get connection timed out.  then in the browser i resubmit, it works.
18:23 tcohen how many workers are you running?
18:24 wajasu 2
18:24 wajasu plack
18:24 wajasu i'm the only user
18:24 wajasu its standard out of the box deployment.
18:25 wajasu just IP config for apache
18:25 wajasu localhost
18:25 wahanui somebody said localhost was literally "localhost" or I should sub that for something else?
18:26 tcohen so
18:26 tcohen if you restart things does it work?
18:26 tcohen koha-zebra --stop $(koha-list --enabled) ; koha-zebra --start $(koha-list --enabled)
18:27 wajasu when is searcj, i see tail -f /var/log/apache/other_vhosts_access.log shows  access lines
18:27 wajasu ok. i'll try
18:27 tcohen tail -f /var/logs/koha/<instance>/*.log
18:31 pastebot "wajasu" at pasted "tail -f /var/logs/koha/<instance>/*.log" (7 lines) at
18:32 wajasu remember its my old 3.10 koha brought to 22.05
18:33 wajasu maybe i need branch_limit configured because it may be dooing some work based on that and timing out
18:34 wajasu line 491 mentions building cache_key
18:34 tcohen you did a fresh install, right?
18:34 tcohen and just carried your /var/lib/koha/<instance> stuff
18:34 tcohen and hte DB
18:34 tcohen right?
18:35 wajasu i did a few this weekend. with same results. things work with same database in ktd.
18:36 wajasu yes. just fresh install.  takes 20 min with my placker script, and then copied the db backup sql to the box. then shutdown koha-common and apache. then dropped db. then streamed db backup into new db.
18:37 wajasu then did koha-upgrade-schema <instance>
18:37 wajasu then koha-rebuild -f -v <instance>
18:37 wajasu then startup koha-common  and apache
18:38 wajasu staf interface works, but opac had the problem.
18:41 tcohen is PLack enabled for both OPAC and staff?
18:41 tcohen thinking of deployment issues
18:42 tcohen koha-plack --enable <instance>
18:42 tcohen if you tweaked the files manually you might have broken koha-plack --enable
18:42 * tcohen has been there
18:44 wajasu they both show enabled, and i disabled and reeneabled. plus i check the apache <instance>.conf to see if uncommented
18:44 tcohen great
18:45 wajasu restarted koha-common and apache still has problem.
18:47 wajasu do you think that branch-limit undef warning in the paste above could be an issue?  staff search doesn't show that
18:52 wajasu i guess i can try to turn on db logging to see the a query that might be slow.  maybe i don't have an index.
18:53 tcohen do the zebra logs show anything?
18:53 tcohen did you change searchengine?
18:53 tcohen did you happen to restart memcached as well? (it caches koha-conf.xml and some stuffs like MARC frameworks)
18:54 wajasu nothing logged in zebra-error.log or zerba-output.log for that instance
18:55 wajasu restarting memcached didn't help
18:55 wajasu i left searach engine alone.  remember my 3.10 db wasnt using memcached
18:56 wajasu sysprefs shows session storage in mysql
18:56 fridolin joined #koha
18:57 fridolin yellow
18:57 wajasu maybe changing it the memcached will help
18:57 tcohen no no
18:57 wajasu ok
18:57 tcohen memcached is not used for sessions only
18:57 tcohen it is configured by default in koha-conf.xml
18:58 tcohen and used by Koha::Cache
18:58 tcohen which is used all over the place to speed up things
18:58 wajasu remember i installed this same db in place of koha_kohadev on a parallel running KOHA_IMAGE=22.05 kul
18:58 wajasu it works there.
18:59 lukeg joined #koha
19:00 wajasu ok/2Gps in network)
19:00 wajasu i'm just trying on this linode VPS.  which was a 1M 1CPU VPS, and now I'm on a 2M ram, 1 CPU (5GBdisk, 40Gbps out
19:01 wajasu top shows no cpu or much load, swap 512M but not being used
19:11 tcohen I'm clueless right now
19:11 tcohen what I'd do is adding some warn ""
19:11 tcohen in the relevant places
19:12 wajasu ok. heading there now.  just turned on mysqld query log to see if anything hanging.
19:12 tuxayo wajasu: if you need a backtrace: https://wiki.koha-community.or[…]a_stack_backtrace
19:13 wajasu i might see about how linode does NS and see if thats an issue.
19:13 wajasu thx for the debug info.
19:14 wajasu at least you gave me some ideas to follow
19:15 wajasu i wonder if plack has a rquest timeout setting.  and how long it is. then i can bump it up to see.  then search for the slow work being done.
19:16 wajasu at least i have a working ktd to compare settings to.
19:19 wajasu i was trying to get the apt update to pickup to make things more like mt kul environment.  but do i need staging in apt sources.list.d/koha.list to pick that up?
19:28 wajasu thx for the effort tcohen. i'll investigate some more.
20:39 m23 joined #koha
21:01 tuxayo tcohen: are you the right the person to report to that the next meeting is missing on the calendar?
21:03 tcohen I never managed to make it work :-(
21:03 cait i think it was not added automatically
21:03 cait i had to add the minutes manually to the last meeting too
21:05 cait maybe something to look into or someone else maybe could run? not sure if ashimema can
21:05 tcohen I ran the koha-meetings script
21:05 tcohen which generated the next one
21:05 cait yeah, but it didn't edit the old one
21:05 tcohen But I am missing the google auth link
21:05 cait to add link to next and minutes
21:05 tcohen and I have no idea about the minutes
21:05 cait and didn't update the calendar, so something is wrong
21:06 cait it does all that usually
21:06 tuxayo davidnind++ thanks for adding in the calendar
21:06 cait possibly it gets the last meeting date from the calendar and failed therefore?
21:06 cait davidnind++
21:06 cait time to sleep here... it just started raining :)
21:06 tuxayo > which generated the next one
21:06 tuxayo tcohen thanks, it already existed (the wiki page) so I don't know what that did
21:06 cait bye
21:08 tuxayo cait++ thanks for adding the minutes to the last one
21:09 tuxayo tcohen: so in the future for the calendar I can bother David N and for creating the next's wikipage it's you, is it?
21:13 cait joined #koha
21:16 wajasu tcohen: i think i found the problem.
21:19 wajasu i changed the number of plack workers to 1.  because this is a VPS with 1 cpu.  now the opac-search doesn't crash.
21:20 fridolin wajasu: we experianced one can set 2 workers per CPU, should be OK
21:20 fridolin 1 will be a big bottle-neck
21:21 fridolin note that each worker uses RAM thats another limit
21:21 wajasu i moved to a 2GB VPS so ram is not an issue.  not swapping now.
21:22 wajasu but my 1CPU VPS crashes with 2 workers.
21:22 wajasu maybe the second worker freezes and it times out or whatever.
21:23 wajasu now my staff interface works with the 2 workers.  but not opac.
21:30 wajasu i'm wondering if there is a plack/psgi/starman config i can use to increase whatever timeout for requests
21:30 wajasu just to see if the increased queries in the opac bump it over.
21:31 wajasu i logged mysql queries and see they run for up to 10sec.
21:31 wajasu both on the opac and staff
21:33 wajasu i have this deployed on home box with ktd, and its behaving fine.
21:34 wajasu maybe i should try a shared linode with 2CPU.  just to see if that CPU cap is the issue.
21:35 fridolin database performance seems to be the bootle-neck
21:36 fridolin did you change innodb_buffer_pool_size ?
21:36 fridolin see
21:36 wajasu yes. i see all those queries. :)
21:36 wajasu maybe som n-selects problems.  with joins ...   session update.
21:37 fridolin oups wrong link
21:37 fridolin what is the hard drive ?
21:37 fridolin SSD ?
21:37 wajasu spinning
21:38 fridolin ich, database server dont like that
21:38 wajasu i move this 3.10version of koha that had been running a a box inthe corner since 2010/2012ish
21:38 fridolin do you have innodb_buffer_pool_size >= database size ?
21:39 wajasu after 10 years the disk died. at least i did backup to a second drive :)
21:39 fridolin great life ;)
21:39 wajasu let me check
21:39 fridolin that whould have been said by mysqltuner
21:42 lari_ joined #koha
21:42 kivilahtio_ joined #koha
21:44 claw23 joined #koha
21:44 irc joined #koha
21:47 wajasu tried innodb_buffer_pool_size=900M  still crashes.  let me look into tuning.
21:48 paxed joined #koha
21:50 fridolin1 joined #koha
22:01 wajasu Joins performed without indexes: 81
22:02 wajasu maybe i am missing some indexes.  i'll try to compare this db to a fresh 22.05 install
22:05 bag joined #koha
22:08 dpk joined #koha
22:13 fridolin1 wajasu: i bet some queries without indexes  cant be avoided
22:20 wajasu yes. but i could be missing an isbn index or something added since 2012.
22:22 wajasu  Deploying with starman  suggests one worker per cpu.
22:22 wajasu i might configure my ktd or local kvm on my box with 1 cpu and see if i'm in the same boat
22:23 wajasu time for a break. thx for ideas.
22:23 wajasu i assume koha these days can't run without plack.
22:24 wajasu time for a break
22:55 davidnind[m] joined #koha
23:41 tuxayo wajasu: I think it can run without plack
23:42 tuxayo Wasn't that tcohen that wasn't using plack in production at some point?
23:45 tcohen I submitted koha-plack in 2015 and have been using it ever since :-D
23:45 tcohen bug 13791
23:45 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=13791 new feature, P5 - low, ---, tomascohen, CLOSED FIXED, Plack - Out of the box support on packages
23:47 tcohen I think ashimema has a dev environment they use for identifying patches that introduce slowliness
23:49 tcohen 'crashes'
23:49 tcohen needs more details

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