| Time |
Nick |
Message |
| 05:02 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 06:54 |
|
agoben joined #evergreen |
| 07:15 |
|
rjackson_isl joined #evergreen |
| 07:28 |
|
Dyrcona joined #evergreen |
| 08:00 |
Dyrcona |
So, I have a record where metabib.reingest_metabib_field_entries is adding a space in the title, "Scrapbooks" becomes "Scra pbooks." I have verified that the 245 does not have some invisible Unicode space in that position. |
| 08:03 |
Dyrcona |
I should add that the space appears in the metabib.title_field_entry.value column. |
| 08:18 |
|
bos20k joined #evergreen |
| 08:30 |
Dyrcona |
Interesting... mods32 contains this: <titleInfo><nonSort>Scra</nonSort><title>pbooks :</title> |
| 08:31 |
Dyrcona |
Bad indicators. |
| 08:31 |
Dyrcona |
Why am I figuring this out and not the actual catalogers? |
| 08:37 |
JBoyer |
How it was reported may have something to do with it. If it's "we can't find this by searching for 'Scrapbooks'" I guess I can see someone not immediately assuming indicators, but it would be nice if someone at least glanced at them before it got to you. |
| 08:37 |
rhamby |
Dyrcona: I'm not allowed to actually answer that because I'm bound by the terms of release the last time the Cataloger Gestapo got ahold of me |
| 08:38 |
Dyrcona |
JBoyer | rhamby: No, the catalogers reported it to me because it display as "Scra pbooks" in the catalog. |
| 08:39 |
JBoyer |
Hmm. :/ |
| 08:39 |
|
mmorgan joined #evergreen |
| 08:39 |
rhamby |
I'm still on my first cup so my intelligent well composed response is: ug |
| 08:43 |
Dyrcona |
So, I'm explaining the problem and giving the ticket back to the cat center. |
| 08:44 |
* Dyrcona |
catalogs by editing the marcxml in a text file and then updating the database via psql. :) |
| 08:45 |
Dyrcona |
Heh. I should write a MARC mode for emacs. :) |
| 08:45 |
* Dyrcona |
checks if one already exists. |
| 08:49 |
Dyrcona |
Heh. Bmagic might like this: marcopolo - Emacs client for Docker API |
| 09:08 |
|
jvwoolf joined #evergreen |
| 09:21 |
|
kenstir joined #evergreen |
| 09:21 |
Dyrcona |
This also leads me to think that there maybe a bug of adding an extra space for non-filing characters... |
| 09:50 |
Dyrcona |
Ah. I think I see why the space gets added. There's a newline and other spaces between </nonSort> and <title> in the <titleInfo> element when returned from the PostgreSQL XPATH function. |
| 10:25 |
|
khuckins joined #evergreen |
| 10:32 |
|
Christineb joined #evergreen |
| 11:01 |
bos20k |
Hello. Does anyone have any idea what could cause us to get the error below in the logs when searching in the OPAC and no results are returned? |
| 11:01 |
bos20k |
Can't call method "opac_visible" on an undefined value at /usr/local/share/perl/5.22.1/OpenILS/Application/Storage/Driver/Pg/QueryParser.pm line 1191. |
| 11:08 |
JBoyer |
bos20k, a problem with the org tree can do that. If you specify an invalid org in a query param it will usually dump an ISE on you, but I think some changes without a catch-up autogen.sh can cause this |
| 11:09 |
JBoyer |
(or potentially an out of date org tree in memcache) |
| 11:10 |
bos20k |
JBoyer: Thanks. I was thinking it was probably something like that. I didn't see any issues but I'll go over it again. |
| 11:22 |
bos20k |
Looks like it was memcached on the test database server. Thanks again! |
| 11:22 |
bos20k |
JBoyer++ |
| 11:23 |
JBoyer |
yay! |
| 11:27 |
bos20k |
I think I need to add it to our documentation to restart memcached on the test database server after refreshing one of the databases from production. I think that is probably what set it off. |
| 11:30 |
JBoyer |
After a full db refresh that might be the way to go, yeah. If you just want to wipe out the org cache real quick you can just use memcrm to clear the orgtree. , orgtree.en-US , and orgtree.en-us keys and you should be in business. |
| 11:32 |
bos20k |
I'm wondering if this would work instead of a restart: echo 'flush_all' | nc localhost 11211 |
| 11:32 |
jeff |
it could have similar effects, assuming memcached is bound to that interface. |
| 11:33 |
jeff |
there are things related to slab allocation that will not be reset, but the goal of the restart in the above context is to quickly flush all keys, so... |
| 11:34 |
bos20k |
jeff++ |
| 11:46 |
Dyrcona |
JBoyer++ # For being more than a "rubber duck." :) |
| 12:08 |
mmorgan |
Anyone else who's looked at autorenewals noticed that the new row created by the autorenewal in action.circulation gets auto_renewal = TRUE and desk_renewal = TRUE? |
| 12:08 |
mmorgan |
Seems to me that only auto_renewal should be TRUE |
| 12:09 |
|
jihpringle joined #evergreen |
| 12:12 |
JBoyer |
mmorgan, so I had not noticed this, but you're right. it should be auto only. |
| 12:13 |
mmorgan |
Ok, thanks, I'll open a bug. |
| 13:04 |
|
jvwoolf joined #evergreen |
| 13:26 |
JBoyer |
mmorgan, good news, I think I'm already on the trail of that one and should have a branch to test very soon. |
| 13:26 |
mmorgan |
JBoyer++ |
| 13:31 |
Dyrcona |
JBoyer++ |
| 13:32 |
|
stephengwills joined #evergreen |
| 13:34 |
remingtron |
dluch: still need help creating a wiki page? |
| 13:38 |
remingtron |
Just go to a non-existing wiki page and look for the "Create this page" icon on the right (pencil with plus sign) |
| 13:38 |
remingtron |
Feel free to ping me if you get stuck. |
| 13:56 |
|
collum joined #evergreen |
| 14:18 |
|
jvwoolf joined #evergreen |
| 14:27 |
|
khuckins joined #evergreen |
| 14:34 |
|
JBoyer joined #evergreen |
| 14:37 |
JBoyer |
I know printers have a varied (and valid!) history as villains in the tales of IT, but no hardware's betrayal is quite so swift and brutal as the keyboard's. |
| 14:38 |
JBoyer |
What are the odds that when I reach for a (NEW IN BOX) keyboard to replace a flaky wireless model, that the replacement will itself confound me? |
| 14:38 |
JBoyer |
THE ODDS ARE GOOD, FRIENDS. |
| 14:38 |
* JBoyer |
stops ranting.. |
| 14:38 |
|
tlittle joined #evergreen |
| 14:39 |
mmorgan |
@hate wireless keyboards |
| 14:39 |
pinesol |
mmorgan: The operation succeeded. mmorgan hates wireless keyboards. |
| 14:39 |
dluch |
remingtron: thanks! i'll give it a try |
| 14:44 |
Dyrcona |
The best keyboard that I ever used was on an Olivetti portable, manual typewriter. |
| 14:48 |
JBoyer |
I highly doubt my wireless model was actually flaky (Had the same model at home for a longer time) but when run through a docking station (requires a reciever, not real Bt) and the laptop runs ENTERPRISE SOFTWARE, perfectly good things can immediately lose their good-ness. |
| 14:48 |
JBoyer |
I do respect the typewriters, but don't miss them |
| 14:50 |
mmorgan |
I miss wite-out sometimes - but not very often |
| 14:52 |
abneiman |
mmorgan: I miss liquid wite-out, the ribbon stuff is just terrible |
| 14:53 |
JBoyer |
Correction tapes! |
| 14:53 |
JBoyer |
for when your typewriter cost as much as a car. ;) |
| 14:57 |
|
nfburton joined #evergreen |
| 15:02 |
Dyrcona |
So, here's a fine note to end my day: The JS console tells me a method doesn't exist, when it does exist: error calling method open-ils.circ.open_non_cataloged_circulation.user : 404 : Method [open-ils.circ.open_non_cataloged_circulation.user] not found for OpenILS::Application::Circ |
| 15:03 |
gmcharlt |
Dyrcona: I sometimes see that sort of thing when a specific Perl module for a service didn't compile cleanly for some reason |
| 15:04 |
Dyrcona |
gmcharlt: I'm going to look into that, but this is a copy of our production branch... |
| 15:04 |
Dyrcona |
I suspect circ has some perl syntax error, but that usually stops the service from starting, and it appears to be running. |
| 15:06 |
Dyrcona |
Doesn't make me feel any better that this is a checkout of the branch we run in production, and the error doesn't occur there.... |
| 15:07 |
Dyrcona |
Oh wait.... I bet it's a Perl version thing.... This is Ubuntu 18.04 and production is Ubuntu 16.04. |
| 15:07 |
JBoyer |
That's not terribly reassuring either. |
| 15:16 |
Dyrcona |
No, but that's the likely culprit. I'll have to look into it later, since it's time for me to end my day. |
| 15:19 |
|
jvwoolf joined #evergreen |
| 15:41 |
|
stephengwills joined #evergreen |
| 17:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 17:12 |
|
mmorgan left #evergreen |
| 17:23 |
|
jvwoolf left #evergreen |
| 17:40 |
|
kenstir joined #evergreen |
| 17:45 |
|
Dyrcona joined #evergreen |
| 17:46 |
Dyrcona |
RE: The errors with open-ils.circ.open_non_cataloged.user I reported earlier. That doesn't happen with master on Ubuntu 18.04. So far I've only seen it on Evegreen 3.2.4 running on Ubuntu 18.04, and we don't officially support that. |
| 17:46 |
Dyrcona |
Just wanted to put that here for the logs/posterity. |
| 18:14 |
|
jihpringle joined #evergreen |
| 18:18 |
Dyrcona |
Odd thing, too. I can't find the problem with 3.2.4 on Ubuntu 18. All of the circulation Perl files checkout with perl -c. There are a lot "subroutine redefined" warnings because of circular dependencies, but those wouldn't cause this sort of failure. |
| 18:19 |
Dyrcona |
I'll check it again. Maybe I missed something. |
| 18:19 |
Dyrcona |
But another time.... |
| 19:43 |
kenstir |
Does anyone know where to find the OSRF method documentation online? I used to use https://webby.evergreencatalog.com/opac/extras/docgen.xsl and it was much easier than grepping the sources. |
| 22:01 |
|
sandbergja joined #evergreen |
