| Time |
Nick |
Message |
| 08:01 |
|
BDorsey joined #evergreen |
| 08:02 |
|
cbrown joined #evergreen |
| 08:21 |
|
redavis joined #evergreen |
| 08:43 |
|
mmorgan joined #evergreen |
| 08:44 |
|
ianskelskey joined #evergreen |
| 09:01 |
|
dguarrac joined #evergreen |
| 09:03 |
|
Dyrcona joined #evergreen |
| 09:04 |
|
mmorgan joined #evergreen |
| 09:38 |
pinesol |
News from commits: LP#2081832 Adding DEV Docker files to .gitignore <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=f489792c7885d29038cdb3e46e8ceaeab3a2580a> |
| 10:17 |
|
abowling joined #evergreen |
| 10:21 |
abowling |
Following an upgrade from 3.7.3 to 3.13.5, getting this error for all pages. I'm guessing something missed on the DB upgrades. Anyone know where I might look for this? https://ctxt.io/2/AAB4FtwjEg |
| 10:21 |
abowling |
/usr/local/share/perl/5.30.0/OpenILS/WWW/EGWeb.pm is generating an Apache error on line 206 |
| 10:22 |
redavis |
abowling - has opensrf been upgraded? Also what version of PG was being used with 3.7.3? |
| 10:23 |
abowling |
redavis: it has; also, 9.6 |
| 10:24 |
redavis |
I don't remember what version, but IIRC, PG needs to be upgraded possibly even to PG13 for 3.13 |
| 10:24 |
* redavis |
goes to look |
| 10:29 |
redavis |
So, the minimum supported for 3.11 is 10...but checking that it might be even higher for 3.13 |
| 10:30 |
redavis |
It's still saying PG10 for 3.13. So, a good place to start at least. But the minimum for 3.14 is PG13 |
| 10:57 |
Dyrcona |
Pg 10 is still the minimum for 3.13. Pg 13 is the minimum for 3.14, though it technically still works on Pg 10. |
| 10:58 |
redavis |
Dyrcona++ |
| 10:59 |
Dyrcona |
And, I should have read redavis++ last comment before adding mine... |
| 10:59 |
redavis |
lol, nuh uh |
| 10:59 |
Dyrcona |
I was just going to mention that I updated the Hack-A-Way 2024 page on the wiki just now: https://wiki.evergreen-ils.org/doku.php?id=hack-a-way:hack-a-way-2024 |
| 11:00 |
redavis |
That list of accomplishments is too legit to quit |
| 11:00 |
Dyrcona |
I added a summary of the accessibility discussion based on my notes, and added a count for the list of bug updates. |
| 11:01 |
Dyrcona |
Yes, I think it was a productive hack-a-way. |
| 11:02 |
redavis |
I wish we could do it semi-annually (not attached to the conference - that's too much cognitive load), but I know that's a pipedream. |
| 11:04 |
csharp_ |
abowling: sounds like either missing updates in the running code or possibly opensrf configs need updating |
| 11:05 |
csharp_ |
the error seems like a perl library is just out of date |
| 11:05 |
csharp_ |
if it's an upgraded server (OS-wise), make sure all traces of the previous perl install are gone (e.g. /usr/local/share/perl/<old Perl release>) |
| 11:11 |
berick |
agreed, hack-a-way was productive |
| 11:11 |
berick |
Bmagic: i def. cribbed from your ansible stuff early on :) |
| 11:15 |
csharp_ |
berick++ |
| 11:15 |
csharp_ |
berick: about to dive back in on that - may make sense to 1) update for newer versions of ansible and 2) do the other stuff I need in separate chunks |
| 11:16 |
csharp_ |
I was getting overwhelmed |
| 11:17 |
berick |
csharp_: i committed my include_tasks: patch, btw. https://github.com/berick/evergreen-ansible-installer/commit/a41a534edc4f17e40f5c852cf9a305a0fceae4e5 |
| 11:17 |
berick |
and your HTTPS patch |
| 11:17 |
csharp_ |
noice |
| 11:17 |
berick |
i /just/ ran the 24.04 version with success |
| 11:17 |
Bmagic |
berick++ |
| 11:17 |
csharp_ |
I should probably start with 24.04, come to think of it |
| 11:17 |
csharp_ |
why live in the past? |
| 11:18 |
csharp_ |
(well, stability, known issues, working code) |
| 11:19 |
csharp_ |
@ana sudo apt -y autoremove |
| 11:19 |
berick |
https://www.youtube.com/watch?v=HqjhHVUzl8o |
| 11:19 |
pinesol |
csharp_: Mouse up-to-date ovary |
| 11:19 |
csharp_ |
ick |
| 11:19 |
csharp_ |
pinesol: be better |
| 11:19 |
pinesol |
csharp_: Have you tried taking it apart and putting it back together again? |
| 11:19 |
Dyrcona |
berick: I'm signing off on your commits for Lp 2083856 and looking into the test failure. I swear it did not do that last Wednesday. Also, I swear I added all of the prerequisites, but maybe I didn't or switched branches and lost 'em? |
| 11:19 |
pinesol |
Launchpad bug 2083856 in Evergreen "Add Support for PostgreSQL 17" [Wishlist,Confirmed] https://launchpad.net/bugs/2083856 - Assigned to Jason Stephenson (jstephenson) |
| 11:20 |
csharp_ |
berick++ # don't blame me, I voted for Kodos |
| 11:20 |
berick |
Dyrcona: cool, yeah, i didn't get a chance to follow up on the failures. |
| 11:21 |
|
Christineb joined #evergreen |
| 11:28 |
Dyrcona |
berick: It looks like the same output with the 100 and 600 tags in opposite order. |
| 11:29 |
Dyrcona |
I'm going to reverse them in one of the copies and see what diff says after. |
| 11:31 |
Dyrcona |
Yeah, that's it. If I modify a copy of what we're looking for from the test to have the 100 before the 600, and the 905 (to put 110 before 600), the output is the same. |
| 11:32 |
Dyrcona |
I wonder if we have been depending on undefined behavior again? |
| 11:32 |
Dyrcona |
Bleh that 110 should be 100... Anyway.... |
| 11:33 |
Dyrcona |
I suppose I can make the test check the Pg version and do something different if it 17+. It wouldn't be the first time we've done that. |
| 11:35 |
Dyrcona |
I wonder though if the auth overlay function needs fixing for more predictable output? |
| 11:38 |
pinesol |
News from commits: LP#1721026: Default owner for copy tags and types <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=79980aff138c39b6056f8c0795fa308a8293fa79> |
| 11:42 |
csharp_ |
oh, I didn't realize the next Ubuntu release is "noble" - that kind of synergy doesn't happen often, does it NOBLE? |
| 11:44 |
Dyrcona |
I name my test vms for the release codenames: bullseye, bookworm, jammy, noble, etc. So, I have a test vm named noble.cwmars.org. :) |
| 11:45 |
csharp_ |
SO CONFUSING |
| 11:45 |
Dyrcona |
I suppose I could have used the animal name: numbat.cwmars.org.... |
| 11:47 |
* mmorgan |
wants to say it's cool, but agrees with csharp_ SO CONFUSING! |
| 11:49 |
Dyrcona |
Should the fields in MARC output be more or less in order? |
| 11:55 |
Dyrcona |
Hm.. I can't find anything that says the blocks have to be in order. I know that fields don't have to be in order within the numbered blocks, just that all the 6XX fields have to be together, etc. |
| 12:09 |
|
jihpringle joined #evergreen |
| 12:15 |
csharp_ |
ejabberd-- |
| 12:19 |
Dyrcona |
I think an `order by tag` in authority.generate_overlay_template might fix it. |
| 12:28 |
Dyrcona |
huh. That didn't help. |
| 12:30 |
Dyrcona |
Well, let's see. maybe pg17 produces the old output now? |
| 12:31 |
Dyrcona |
Nope. The order by isn't making a difference... |
| 12:31 |
Dyrcona |
OK. order by on the inner loop, then. |
| 12:33 |
Dyrcona |
That seems to work. |
| 12:39 |
Dyrcona |
I think the issue with the authority.generate_overlay_template function should be a separate bug. |
| 12:45 |
Bmagic |
now that redis is merged onto OpenSRF main, I'm drooling over LP2032835. Did someone already have a branch? |
| 12:46 |
Bmagic |
it seems that pinesol isn't coded to detect "LPXXXXX" hows this: bug 2032835 |
| 12:46 |
pinesol |
Launchpad bug 2032835 in OpenSRF "Discussion: Merge OpenSRF Into Evergreen?" [Wishlist,New] https://launchpad.net/bugs/2032835 |
| 12:46 |
berick |
Bmagic: no branch yet, but I'm def. interested |
| 12:46 |
|
abowling joined #evergreen |
| 12:47 |
Bmagic |
copy/paste the folder, update the install instructions, ship it. LOL |
| 12:47 |
Bmagic |
I know it's going to have a tangled web of crap we're going to need to cut through |
| 12:48 |
Bmagic |
the first question is where* does OpenSRF go... obviously a new root folder in the Evergreen repo for it? |
| 12:48 |
berick |
i would expect so, yes |
| 12:50 |
Bmagic |
The first step isn't to integrate all of the (perl) stuff. Just get the files over, and update the install instructions. And maybe update the make install to install the pre-reqs for both in one. It wouldn't make sense to run two make installs |
| 12:51 |
berick |
agreed, one set of docs, one set of prereq installers |
| 12:52 |
Bmagic |
it gets a little hairy in there. I actually might have time to take a stab at it next week or so. But if you've already got a clear path in your head, I think I should defer. |
| 12:54 |
berick |
k, either way. if/when I get a chunk of time i'll be sure to grab the bug. |
| 12:54 |
Dyrcona |
Oops! I inserted the wrong output for the "bad" output. |
| 12:54 |
Bmagic |
berick++ |
| 12:55 |
Dyrcona |
Oh wait. No, that's correct. |
| 12:55 |
Dyrcona |
berick: Lp 2088529 |
| 12:55 |
pinesol |
Launchpad bug 2088529 in Evergreen "authority.generate_overlay_template and undefined behavior" [Undecided,New] https://launchpad.net/bugs/2088529 - Assigned to Jason Stephenson (jstephenson) |
| 12:58 |
berick |
Dyrcona++ |
| 12:59 |
abowling |
Still getting the error. System ERROR: Exception: OpenSRF::DomainObject::oilsMethodException 2024-11-17T20:45:07 OpenSRF::AppRequest /usr/local/share/perl/5.30.0/OpenSRF/AppSession.pm:1171 <404> Method [open-ils.search.fetch_context_library_groups.atomic] not found for OpenILS::Application::Search\n\n |
| 12:59 |
abowling |
The files are all there |
| 12:59 |
|
BDorsey_ joined #evergreen |
| 13:02 |
Dyrcona |
abowling: Try this: `find /usr -name Search.pm -exec perl -c {} \;` |
| 13:02 |
Dyrcona |
That should find any errors with syntax in the Search modules. |
| 13:03 |
abowling |
Dyrcona++ |
| 13:03 |
abowling |
BEGIN failed--compilation aborted at /usr/local/share/perl/5.30.0/OpenILS/Application/Search/Biblio.pm line 13. |
| 13:03 |
abowling |
Compilation failed in require at /usr/local/share/perl/5.30.0/OpenILS/Application/Search.pm line 13. |
| 13:03 |
Dyrcona |
OK, so see what line 13 in Biblio.pm is doing. It's probably a missing Perl module or something like that. |
| 13:05 |
abowling |
Dyrcona: Yep |
| 13:05 |
abowling |
use OpenILS::Application::Search::Biblio; |
| 13:06 |
Bmagic |
run through the installation of OpenSRF again |
| 13:17 |
Dyrcona |
Well, it's also complaining about Biblio.pm at line 13, so that's probably the line with the module that's missing. |
| 13:24 |
abowling |
Dyrcona: The module is there. Any ideas? |
| 13:24 |
Dyrcona |
You could run perl -c /usr/local/share/perl/5.30.0/OpenILS/Application/Search/Biblio.pm to see if anything more pops up. |
| 13:25 |
Dyrcona |
It looks like something is missing somewhere. If the module on line 13 is there, try running perl -c on its file. |
| 13:28 |
Dyrcona |
abowling: As Bmagic suggested earlier, you could do the prerequisite installation for OpenSRF and Evergreen again for the new versions. Then you might have to rebuild and reinstall OpenSRF and Evergreen. |
| 13:29 |
abowling |
Dyrcona: It's what csharp_ referenced earlier; my 5.26.1 folder has the dependencies, but not the 5.30.0 folder |
| 13:30 |
Dyrcona |
Well, 5.26.1 should still be in the Perl path... You could try csharp_'s suggestion of deleting the 5.26.1 stuff and then reinstalling the prerequisites. |
| 13:31 |
|
jihpringle joined #evergreen |
| 13:36 |
Dyrcona |
heh.. The test still fails...... |
| 13:38 |
Bmagic |
tests++ |
| 13:48 |
Dyrcona |
extraneous space from where I inserted the new output. |
| 13:50 |
Dyrcona |
The db test passes on pg 17. I'll test on an earlier pg release before committing the changes. |
| 13:50 |
Dyrcona |
I mean committing them locally and pushing to a working branch. |
| 14:20 |
|
BDorsey joined #evergreen |
| 14:34 |
|
collum joined #evergreen |
| 14:39 |
pinesol |
News from commits: LP#2083959 <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=1d68106a6feb866f6ad6bc1e27b2cfa45a60a0c2> |
| 14:39 |
pinesol |
News from commits: LP#2086737: Angular Pull List Needs for Shelving Locations Selector <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=f9159c0e0d0063fb3c5089d9e41ba63882a6fab7> |
| 14:52 |
sleary |
... dangit, I squashed in the commit that had a better commit title there :( |
| 15:37 |
eeevil |
Bmagic: re earlier merge opensrf into eg repo, part of the point of doing the 4.0-is-redis thing is specifically so that newer (future) EG can still use opensrf 3.3 ... there WILL be churn. let's get through that first? |
| 15:38 |
|
mantis joined #evergreen |
| 15:38 |
Bmagic |
eeevil: so waiting until supported Evergreen versions have aged out non-redis OpenSRF? |
| 15:39 |
pinesol |
News from commits: LP2007284: Don't fail when assigning a new part to multiple items <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=9ddf407807183e3335bc453ee5fd710f5cf66722> |
| 15:39 |
pinesol |
News from commits: LP2007284: Don't fail when assigning a new part to multiple items <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=8207ed42d16aca0bd2b51dc47d567f68ac82124d> |
| 15:39 |
pinesol |
News from commits: LP#2087609: Fix incorrect lit_form parameter in search query <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=7365698692e77f49b20df7dff416670d546eec2f> |
| 15:39 |
pinesol |
News from commits: LP2077343: Stripe - patron can be charged w/o credit in Evergreen <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=3253487b9ee281fc6fbc2c055829f72c2a17a6dc> |
| 15:39 |
redavis |
Bmagic++ |
| 15:40 |
eeevil |
abowling: if you upgraded the OS, and that included a perl version bump, perl ignores (some) old @INC dirs from the previous version. you can either move the modules from the old perl to the new one, or reinstall EG |
| 15:40 |
redavis |
ianskelskey++ smayo++ mmorgan++ |
| 15:42 |
eeevil |
Bmagic: waiting until opensrf 4+ settles down all the way before we start that conversation. I doubt there will be an EG any time soon that can't run on xmpp-opensrf, and if there is, then I fear we're changing deep infrastructure inside EG itself too quickly. |
| 15:43 |
Bmagic |
oh, I see what you're saying. The redis OpenSRF (now on main) will likely not be "final" until more and more installation shake out the bugs |
| 15:57 |
|
jihpringle joined #evergreen |
| 16:16 |
abneiman |
went back to LP and pulled the rest of the fabulous work that was accomplished last week: https://wiki.evergreen-ils.org/doku.php?id=hack-a-way:hack-a-way-2024 |
| 16:16 |
abneiman |
all_yall++ |
| 16:16 |
Bmagic |
yeah, it's impressive |
| 16:36 |
sleary |
abneiman++ |
| 16:53 |
csharp_ |
all_of_yall++ |
| 16:54 |
Bmagic |
all_of_yall++ # lol |
| 16:54 |
Bmagic |
pinesol should be programmed to see that and +1 to all Nick's in channel |
| 16:54 |
pinesol |
Bmagic: Ba ba ba dook Dook DOOK! |
| 16:55 |
csharp_ |
pinesol: terrifying |
| 16:55 |
pinesol |
csharp_: Thank you csharp_! But our princess is in another castle! |
| 17:00 |
|
mantis left #evergreen |
| 17:00 |
|
jihpringle joined #evergreen |
| 17:11 |
|
mmorgan left #evergreen |
| 22:04 |
|
phasefx joined #evergreen |
