10:50 |
pinesol |
News from commits: LP1958573 PMC messages created by action triggers not patron-visible <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=ba2fc41399e9723361245a95e64d78a6c1eb2e59> |
12:15 |
mmorgan |
berick++ gmcharlt++ # fixing :) |
12:32 |
Dyrcona |
DBD::Pg++ |
13:03 |
Dyrcona |
Hrm.. URIs aren't displaying correctly on my test VM, and I just merged my production branch into my test branch.... |
13:04 |
Dyrcona |
And, there are no differences in the OPAC according to git diff.... |
13:13 |
Dyrcona |
Grr... Thought maybe a git clean -xfd and a rebuild would help, but no.... |
13:14 |
Dyrcona |
In production, the link is on subfield z and subfield 3 appears after it. On my test VM it is the opposite in both ways. |
15:15 |
JBoyer |
And +1 to spacing also. Were they forced together for some reason or has experience suggested that they need more space than before? |
15:15 |
terranm |
I think it just happened sort of randomly |
15:16 |
JBoyer |
Dyrcona, perhaps worth checking, is the person you're thinking of here? |
15:17 |
terranm |
I think we get more testing participation if they are spaced out rather than in two months back to back |
15:17 |
Dyrcona |
JBoyer: They |
15:17 |
berick |
i should be able to host a bug squashing VM this time around |
15:17 |
terranm |
berick++ |
15:27 |
berick |
sans menu entry changes, I should say |
15:27 |
gmcharlt_ |
the branch is realistically far too large for detailed patch-level scrutiny |
15:27 |
berick |
yeah, it's a beast |
15:28 |
gmcharlt_ |
but if we merge soon (and I do see reasons for doing that), I think we're going to need to organize community testing on a large scale; possibly larger than previoulsy attempted |
15:28 |
berick |
my thinking is if the UI is not yet exposed, a minimal amount of the new code will actually be used. |
15:29 |
berick |
once we do expose the new UI's, which may come after 3.10, then loads of testing would be needed |
15:29 |
gmcharlt_ |
yeah, but that just kicks the can down the road, since of course the point is eventually to switch to it |
15:29 |
Dyrcona |
gmcharlt_++ |
15:29 |
Dyrcona |
+1, too.... I think if it goes in, turn it on. |
15:30 |
berick |
sort of. if we can expose less ambitious interfaces that use the underlying code, then we get the chance to test parts of it without flipping a giant switch |
15:30 |
gmcharlt_ |
so I think what I'm saying is that organizing broad testing needs to happen on the heals of a merge |
15:30 |
gmcharlt_ |
*heels |
15:31 |
berick |
gmcharlt_: so you're thinking flip the big switch on day one too? |
15:32 |
terranm |
I could work up a list of tasks that need to be tested for each of the new interfaces as a starting point. It might be easier to get people to test if it's broken into bite-sized pieces. |
15:32 |
mmorgan |
Just to clarify, right now, the new uis are under an org unit setting? |
15:32 |
gmcharlt_ |
more like stand up test systems that have the switch flipped |
15:32 |
berick |
mmorgan: sort of. it's kind of a mess what we have. |
15:32 |
berick |
that we were going to test with |
15:32 |
berick |
i'm of the opinion that when we do flip the switch, it has to be fully flipped or it's going to get really confusing real fast |
15:33 |
berick |
hopping between the same UI's depending on how you got there |
15:33 |
gmcharlt_ |
(also noting that soon there will be a dump of the Angular acquisitions sprint 4 stuff, though that's more self-contained than the patron/circ stuff) |
15:34 |
JBoyer |
terranm++ I do think having a reminder of what all needs to be tested helps. I wonder if some of the staff interfaces have the 1-2 things users do *most* and then say "yeah, didn't fall over" by end users. A guide would help them test things they maybe do quarterly or less. |
15:34 |
mmorgan |
terranm++ |
15:34 |
mmorgan |
+1 to broad community testing |
15:35 |
shulabear |
terranm++ |
15:35 |
gmcharlt_ |
also, I think the more that the branch can be cleaned up to have new and changed shared component live in self-contained patches, the better |
15:36 |
gmcharlt_ |
as individually, those should be much more amenable to piecemeal testing by devs and committers |
15:38 |
berick |
yeah, i can clean up the shared/core changes. that's a small portion of it. |
15:39 |
gmcharlt_ |
and I think the other thing that would help is ensuring that there's one big ol' switch (or maybe a small set of them) to reliably switch between old and new interfaces |
15:39 |
berick |
it gets a little complicated with the patron UI, since it's basically one big application. it kind has to be added as a whole. |
15:58 |
gmcharlt_ |
5. and by mid-November, review the state of the bugs against the Angular interfaces and make a go/no-go decision |
15:58 |
gmcharlt_ |
5a. (which is potentially when the decision to apply the BRS as a commit, if that's what it has to be, gets made) |
15:58 |
gmcharlt_ |
(end of my list) |
15:59 |
gmcharlt_ |
but basically, combine (a) a quick merge so that the code doesn't age or diverge too much with (b) a much more intentional testing stance for a big change like this than may have been the case in the past |
16:00 |
gmcharlt_ |
without kicking the can to 3.11, since the testing needs to happen eventually and it would be good not to block the potentiall for normal enhancements to patrons/circ |
16:00 |
berick |
gmcharlt_: ok, that makes sense. and the BRS will be a commit that just reverses all the link/button/menu changes? |
16:01 |
berick |
or a revert, basically |
16:02 |
gmcharlt_ |
if need be; a global setting / YAOUS would make it easier for ongoing testing if we're not confident enough to release 3.10 with $NewCirc, but if it's not feasible, it's not feasible |
16:03 |
berick |
k. do-able, was just hoping to avoid that, but I understand the desire. |
16:03 |
berick |
a bold plan and I like it |
16:03 |
berick |
gmcharlt_++ |
16:18 |
|
stephengwills left #evergreen |
16:39 |
|
Guest75 joined #evergreen |
16:41 |
Guest75 |
Hello everyone, I had someone ask a question and I do not know the answer... In Evergreen, is there a way to add multiple email addresses for hold notification? |
16:47 |
jeffdavis |
Guest75: I *think* it's possible although I haven't tested. There is some discussion in this bug report: https://bugs.launchpad.net/evergreen/+bug/1755625 |
16:47 |
pinesol |
Launchpad bug 1755625 in Evergreen 3.1 "Web Client: Edit Patron Email Address Box Does Not Allow Multiple Addresses" [Medium,Fix released] |
16:50 |
jeffdavis |
basically you should be able to put multiple comma-separated addresses in the email field in the patron editor, but you may need to modify the "Regex for email field on patron registration" setting to allow it |
16:55 |
Guest75 |
thanks for the info, I tried a comma without space and got an error. I will try comma with space and ; |
18:19 |
csharp_ |
@who wants to delete all other languages? |
18:19 |
pinesol |
Christineb wants to delete all other languages. |
18:20 |
jihpringle |
I mean, that's true :) |
18:20 |
csharp_ |
haha |
18:20 |
csharp_ |
jeffdavis: I *think* that's all you need to do - maybe try on a test server? |
18:24 |
jeffdavis |
yeah I can give it a try, I'm just nervous because locales touch some pretty core stuff in EG |
18:25 |
|
mantis1 joined #evergreen |
18:26 |
jeffdavis |
yet another reason we should all just use Esperanto |
08:34 |
|
Dyrcona joined #evergreen |
08:39 |
|
mmorgan joined #evergreen |
09:09 |
|
Stompro joined #evergreen |
09:28 |
Stompro |
Baker and Taylor support notified me this morning that PASV mode should work again on the normal ftp site. But I haven't tested it yet. |
09:33 |
mmorgan |
Offline circ issue: A library can't record checkouts, likely because the "Working Location" field isn't filled in. But the dropdown for that field is not populated. Any suggestions? |
09:34 |
Stompro |
Seems to be working when I test. Public IP returned when passive commands are sent using netkit-ftp, which doesn't have special handling I believe. |
09:45 |
* mmorgan |
goes offline to test offline circ for real |
09:49 |
Dyrcona |
mmorgan: You have to be completely offline. If it can see the server at all, offline doesn't work. |
09:50 |
Dyrcona |
:) |
09:50 |
Dyrcona |
I figured she knew, but..."Won't someone please think of the logs!?" |
11:02 |
Dyrcona |
There are also 3 autovacuum worker processes running in the database. I've been meaning to look into our autovacuum settings. |
11:16 |
Dyrcona |
INFO: scanned index "symspell_dictionary_updates_tid_idx" to remove 126549937 row versions |
11:22 |
Dyrcona |
miker: https://pastebin.com/E4iNzSzw |
11:23 |
miker |
yeah, the delayed reification isn't actually being tested... wheeee |
11:23 |
miker |
you're just testing the deadlock code (again) |
11:23 |
Dyrcona |
Did I miss a flag to turn that on? |
11:24 |
miker |
the pingest param is supposed to casue a function to be called that turns that delay feature on, and then at the end it calls a separate "clean it all up" function |
11:24 |
miker |
you know pingest better than me, so I'd appreciate eyes on that logic change |
11:26 |
Dyrcona |
OK. So, I missed using that option. |
11:27 |
Dyrcona |
Can I ask why that is optional? Shouldn't it just work that way, all the time? |
11:28 |
Dyrcona |
I simply forgot about the option when I started the test run. This is what happens when you a) get older, b) work on many things, and c) have to put one thing down for a week or more before coming back to it. |
11:33 |
miker |
because for non-batch ingest it just creates headaches. human bib editing doesn't interact often enough to matter, but because it does cause serialization (INSERT ... ON CONFLICT does that to prevent deadlocks) with intentionally parallel updates the delay makes it overall faster. same thing as with the browse part of ingest |
11:34 |
miker |
now, I could certainly see making it the default for pingest! but I didn't want to change the default behavior on my own |
11:34 |
miker |
default behavior of pingest, I mean |
08:55 |
|
jvwoolf joined #evergreen |
10:36 |
|
Stompro joined #evergreen |
10:58 |
|
RFrasur joined #evergreen |
11:00 |
Bmagic |
I'm testing an upgrade to PG 10. Performing pg_dump, then pg_restore on pg10. I'm pre-creating the database on PG10 following the Evergreen create database extension file "create_database_extensions.sql" - That setup does not call for tsearch2, which was removed back in EG2.4. When I restore my dumped database, I get errors about missing tsearch2 |
11:00 |
Bmagic |
ERROR: could not open extension control file "/usr/share/postgresql/10/extension/tsearch2.control": No such file or directory |
11:01 |
Bmagic |
Command was: CREATE EXTENSION IF NOT EXISTS tsearch2 WITH SCHEMA public; |
11:02 |
Bmagic |
I found this patch bug 1175287 which was part of db upgrade 0743. which was* ran on this db according to config.upgrade_log. But maybe I'm barking up the wrong tree |
16:12 |
Dyrcona |
Trying to find patrons with over 80 circulations and this happened: SSL SYSCALL error: Connection timed out |
16:15 |
|
jvwoolf left #evergreen |
16:18 |
Bmagic |
That's "fun" |
16:24 |
Dyrcona |
Well, I reconnected and ran a query with a lower having and it returned pretty quickly, so I bumped it back up to 80 and got some accounts to test with. |
17:17 |
|
mmorgan left #evergreen |
17:24 |
|
jihpringle joined #evergreen |
17:30 |
|
Christineb joined #evergreen |
11:14 |
Bmagic |
B&T had a "meeting" yesterday. Still no worky today |
11:29 |
Dyrcona |
B&T seems to be working with PASV today, but it is taking two minutes per connection. I have logs that look we're getting things via the fetcher. I've not checked the pusher, yet. |
11:30 |
Dyrcona |
We're looking at configuring our firewall for active FTP. |
11:31 |
Dyrcona |
miker: Cool! I was thinking more eyes would be good, but I'm not sure who else would get to testing it nor when they could make time. |
11:31 |
Dyrcona |
miker: I'm testing with production data upgraded to rel_3_9 latest today. |
11:56 |
stompro__ |
berick, we added an ftp proxy helper to our firewall and switched to Active mode for B&T ftp last week and that has been working. We noticed that on the 14th, they broke their own invoice uploading to the ftp server so invoices that should have been received that day are missing. |
12:03 |
stompro__ |
berick, I just tested passive mode and it does seem to be working now, at least more than it was. The initial file list works now. |
12:04 |
berick |
thanks stompro__ |
12:05 |
stompro__ |
I'm not seeing the 2 minute delay that Dyrcona mentioned... but maybe that is for actual uploads? |
12:06 |
stompro__ |
I tried downloading a response and it seemed to work fine... but I'm just testing with filezilla on my workstation. |
12:12 |
Dyrcona |
The 2-minute delay seems to happen when we get 0 files. |
12:12 |
|
jihpringle joined #evergreen |
12:20 |
miker |
Dyrcona++ |
06:02 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
07:05 |
|
kmlussier joined #evergreen |
07:05 |
kmlussier |
bshum++ |
07:05 |
kmlussier |
parts++ |
09:13 |
mmorgan |
Good morning! |
09:13 |
|
Stompro joined #evergreen |
09:22 |
|
rjackson_isl_hom joined #evergreen |
09:40 |
eby |
miker: / berick : going to test more next week but curious with your mention of setting workstation prefs as org unit settings. Does it act as a fall back default or as a forced preference? |
09:49 |
phasefx |
berick: do you want to talk SIP2M sometime today? I don't have an agenda, but we could determine/sanity-check next steps |
09:51 |
berick |
eby: it acts as fall back default. ignored in cases where a workstation setting value is already saved. |
09:52 |
berick |
best way to create is to save the value you want at a workstation, then copy the JSON from the value into the newly created org unit setting. |
09:59 |
pinesol |
Dyrcona: go with local git |
10:07 |
* Dyrcona |
wonders if we should add some instrumentation to pingest.pl (and possibly drop the .pl extension). |
10:16 |
Dyrcona |
Hmm. I guess rebasing from master to rel_3_7 is not a good idea. |
10:18 |
miker |
Dyrcona: do I feel my ears burning? ;) (if you're looking at what I think you're looking at, if you can test closer to master I think you'll have more success) |
10:18 |
Dyrcona |
Remote syntax in git can be annoying: orign/rel_3_7 for one command and origin[space]rel_3_7 for another. |
10:19 |
Dyrcona |
miker: I am. I plan to start testing with rel_3_7 today and an upgrade to rel_3_9 next week if that's OK. |
10:20 |
|
mdriscoll joined #evergreen |
10:20 |
Dyrcona |
I suppose the different syntax is what they call commitish (in the / case) and remote[space]branch in the latter. |
10:22 |
Dyrcona |
I should copy the database that I want to use so it won't get obliterated this weekend. |
10:23 |
* Dyrcona |
is tempted to test on Pg 14 since there was an update this morning... ;) |
10:43 |
miker |
Dyrcona: as far as order and timing goes, I'm just glad someone's looking at it, but just to note that deadlock+symspell branch was developed after 3.7 (recall, nocase browse) so starting with vanilla 3.8 or 3.9 (/not/ patched with this and then upgraded) may be less of a headache. I get that "with your data" testing isn't particularly feasible at scale for versions past what you're actually on, though. |
10:46 |
Dyrcona |
miker: Right. I previously made a 3.7 backport and only had had to change the db upgrade script. Looks the same still, but I'll see what happens. If it just blows up, I'll move to upgrading to 3.9 over the weekend. |
10:47 |
Dyrcona |
I'm going to compare the implementation of metabib.reingest_metabib_field_entries in the two db upgrades. I think we can delete the one from the WWW upgrade. |
10:51 |
Dyrcona |
Yeah, only difference is the check for reification. |
13:16 |
JBoyer |
I'm also basically out of pocket / on the bench / choose your own favorite game / sport related euphemism for unavailable for any real hacking at the fest. Additionally, if anyone wanted to take the reins / read the agenda for the dev meeting this afternoon I would be happy to just listen in and pop in should I have something worth saying. |
13:19 |
miker |
Dyrcona: huzzah! with that option it /should/ be as fast as before (insert-only into the unlogged table) and then at the end there'll be a pause while it shoves the changes in, in one go |
13:22 |
Dyrcona |
miker: We'll see. |
13:22 |
Dyrcona |
I also just tested a 3.7.3 to 3.9 db upgrade that I prepared for next week, and I got this error: psql:3.7.3-3.9-upgrade-db.sql:3509: ERROR: cannot drop type search.search_result because other objects depend on it |
13:22 |
Dyrcona |
DETAIL: function search.query_parser_fts(integer,integer,text,integer[],integer[],integer,integer,integer,boolean,boolean,integer) depends on type search.search_result |
13:22 |
Dyrcona |
I wonder if we have something old or custom hanging around? |
13:23 |
Dyrcona |
The suggestion in the error report was to use drop cascade. |
13:24 |
Dyrcona |
Looks like we won't be needing it. |
13:27 |
jeffdavis |
We have two versions of search.query_parser_fts in our environment and I've had to manually drop the second one for the upgrade scripts to work in testing. |
13:27 |
jeffdavis |
So yeah, probably an old version of that function sticking around on ancient EG installs. |
13:30 |
Dyrcona |
jeffdavis: Thanks. That's what this looks like. |
13:30 |
Dyrcona |
I'm using cascade on the type drop, and that appears to work. |
13:32 |
Dyrcona |
It's fun running upgrades over multiple version. :) |
02:33 |
|
akilsdonk joined #evergreen |
06:00 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live//archive/2022-06/2022-06-10_04:00:02/test.49.html> |
07:26 |
|
rfrasur joined #evergreen |
07:32 |
|
rjackson_isl_hom joined #evergreen |
07:38 |
|
collum joined #evergreen |
09:02 |
csharp_ |
screw it, I'll rewrite it in perl |
09:04 |
csharp_ |
do we use warnings and use strict, Barry? Yes, other Barry, we do. |
09:07 |
JBoyer |
-Wall -Werror |
09:08 |
JBoyer |
jeffdavis++ for looking into the http tests. |
09:09 |
JBoyer |
Not sure the nginx issue is an issue for most testing though, I don't know what benefit there would be to standing that proxy up just for the tests. |
09:19 |
|
mantis1 joined #evergreen |
09:29 |
* Dyrcona |
missed something. To the logs! |
09:31 |
Dyrcona |
JBoyer: Is it a nginx issue or is it the directories missing again? |
09:32 |
Dyrcona |
csharp_: You can also install Modern::Perl and 'use Modern::Perl;" to get use strict; use warnings; and some other useful things. |
09:32 |
Dyrcona |
The deb is libmodern-perl-perl |
09:33 |
|
mantis2 joined #evergreen |
09:33 |
Dyrcona |
JBoyer: IIRC, none of the other live tests require even Apache to be running, just OSRF services. |
09:34 |
* Dyrcona |
likes it when a program that I haven't used/touched in 3 years or so still works. |
09:36 |
JBoyer |
Dyrcona, I was referring to what jeffdavis said yesterday about one of the tests giving a false positive if nginx is running on the test machine. You might have to hit the logs if you're not running a bouncer. |
09:36 |
JBoyer |
And there are a couple http tests, but not many, no. |
09:52 |
Dyrcona |
JBoyer: OK. I was out yesterday. Also, I notice that image uploader test failed this morning. |
10:15 |
csharp_ |
Dyrcona: thanks - I'll check it out |
10:16 |
* csharp_ |
tries to rid his brain of Bowie's Modern Love |
10:17 |
berick |
hey csharp_, roam if you want to |
14:22 |
|
rfrasur joined #evergreen |
17:08 |
|
mmorgan left #evergreen |
17:31 |
|
jvwoolf left #evergreen |
18:00 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live//archive/2022-06/2022-06-10_16:00:03/test.49.html> |
18:56 |
pinesol |
News from commits: LP1950345-Format the Current Hold Groups table in bootstrap opac <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=1b466467109a9013dd3799d69be6e80c968ce433> |
19:56 |
pinesol |
News from commits: Hold Management page update <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=2678ee1dfcc448968ba8e8ae7fd0da1bc69eeecf> |
06:01 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live//archive/2022-06/2022-06-09_04:00:02/test.49.html> |
07:28 |
|
collum joined #evergreen |
07:33 |
|
rjackson_isl_hom joined #evergreen |
07:48 |
|
rjackson_isl_hom joined #evergreen |
09:25 |
|
mantis1 joined #evergreen |
09:30 |
|
jvwoolf left #evergreen |
09:31 |
|
jvwoolf joined #evergreen |
09:58 |
mantis1 |
I upgraded one of our test servers from 3.6.5 to 3.9. Has anyone had weird spacing problems in the Boopac after an upgrade? I have no idea why it's spacing out so much like it is. There is a width=device-width element added in base.tt2, but it doesn't seem to make a difference if it's deleted or not. |
10:37 |
|
mantis2 joined #evergreen |
10:38 |
csharp_ |
mantis1: is it stock EG or did you apply customizations on top of the upgrade? |
10:39 |
csharp_ |
(or conversely, applied stock EG stuff on top of your customized stuff?) |
17:02 |
csharp_ |
I'm on at&t, which while not awesome either, at least it's not comcast :-) |
17:21 |
mmorgan |
not_comcast++ |
17:28 |
|
mmorgan left #evergreen |
18:01 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live//archive/2022-06/2022-06-09_16:00:02/test.49.html> |
20:14 |
jeffdavis |
the 34-lp1787968-cover-uploader.t live test is failing with a 404 on http://127.0.0.1/jacket-upload |
20:16 |
jeffdavis |
29-lp1817645-remoteauth-patron-api.t also does HTTP requests, but that test is actually being skipped right now because the test environment's version of LWP::Protocol::https is not >=6.0.7 |
20:19 |
jeffdavis |
and 24-offline-all-assets.t (which does a wget on https://localhost/eg/staff/offline-interface) technically isn't failing but I'm not sure the test is robust - in an environment with an nginx proxy, the test will pass if apache isn't running because the response is a 502 error and the test only checks for 404 |
20:20 |
jeffdavis |
I'm also not really sure that the wget output is being parsed properly |
20:22 |
jeffdavis |
in other words, none of the 3 live tests that do direct HTTP requests against the test server are really "working" |
20:23 |
jeffdavis |
something to fix at the hackfest maybe? |
02:50 |
|
berick joined #evergreen |
06:00 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live//archive/2022-06/2022-06-08_04:00:02/test.49.html> |
07:25 |
|
rjackson_isl_hom joined #evergreen |
07:49 |
|
collum joined #evergreen |
08:05 |
|
RFrasur joined #evergreen |
15:19 |
Dyrcona |
People do get uptight about the queue order, but it's just a guess at best. |
15:19 |
mmorgan |
We're not using a soft regarget interval. Maybe that would help. |
15:21 |
Dyrcona |
We've experimented with every half hour, every 15 minutes, etc. Then, it started running into itself and it ended up being 1/hour effectively. We skip the hours between midnight and 6:00 am to avoid the overnight jobs that run during those hours. |
15:22 |
Dyrcona |
The soft retarget interval might help. John Amundson could probably tell you better why we use it because it was his testing that made us decide on that value. I forget exactly why we ended up with that value. |
15:23 |
* mmorgan |
would thing you could end up with a LOT more holds being retargeted using a soft interval. |
15:25 |
Dyrcona |
Our "normal" retarget interval is 48 hours to keep things on pull lists a little bit longer. |
15:25 |
Dyrcona |
We'd typically see the running into itself happen for a day or so after an upgrade, then it just kept up (probably because of the soft interval), so we changed the schedule. |
17:09 |
|
mmorgan left #evergreen |
17:14 |
jeffdavis |
hm, that doesn't work for me unfortunately |
17:17 |
jeffdavis |
we're using skins, I wonder if that is causing trouble |
18:00 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live//archive/2022-06/2022-06-08_16:00:03/test.49.html> |
22:45 |
jeff |
jeffdavis: do you see the text-white class in the output html and it isn't working, or does that class not make it to the html seen by the browser? |