| Time |
Nick |
Message |
| 01:32 |
|
smayo joined #evergreen |
| 06:48 |
|
collum joined #evergreen |
| 08:30 |
|
mmorgan joined #evergreen |
| 08:48 |
|
smayo joined #evergreen |
| 09:01 |
|
Dyrcona joined #evergreen |
| 09:21 |
Dyrcona |
OK. I think I've figured out what's up with 1 of our failed background imports. The import type in the table is "authority" but the program looks for "auth." I'm going to update that row and see what happens. |
| 09:22 |
Dyrcona |
ERROR: new row for relation "background_import" violates check constraint "background_import_import_type_check" |
| 09:22 |
Dyrcona |
OK, so that's another bug.... |
| 09:26 |
Dyrcona |
So, it looks like everywhere that backgroup_import_mgr.pl checks the import_type, it's looking for "auth" but it should look for "authority." |
| 09:26 |
Dyrcona |
backgroup.... background. I've been making that typo a lot yesterday and today. |
| 09:34 |
Dyrcona |
Now, this: Method [open-ils.vandelay.auth_queue.create] not found for OpenILS::Application::Vandelay. Guess that's another bug... |
| 09:35 |
csharp_ |
ah - we're only using the background importer for acq bib imports, so we wouldn't have found those |
| 09:39 |
Dyrcona |
Yeahp. Understandable. |
| 09:39 |
Dyrcona |
I'll fix line 174 and see what happens after that. |
| 09:40 |
Dyrcona |
That's where the new queue gets created. The call should be open-ils.vandelay.authority_queue.create. |
| 09:45 |
Dyrcona |
Yay! The background import for the authorities says its "complete." |
| 09:53 |
eeevil |
turns out, the password field virtualization patch I merged last week makes (at least, though likely more than) the quipu development unhappy because it wasn't complete enough for its liking. I'm going to force the completion through asap. so this is just a head's up... |
| 09:56 |
Dyrcona |
Now, I've got a bib import to figure out why it's not completing. |
| 09:56 |
Dyrcona |
eeevil++ |
| 09:57 |
Dyrcona |
Hrm... Looks like the background import is found, but after that, nothing. |
| 10:00 |
eeevil |
to that end, grabbing 1487 |
| 10:09 |
csharp_ |
eeevil++ |
| 10:10 |
Dyrcona |
OK. The queue is not getting created. |
| 10:15 |
Dyrcona |
So, it's getting to resolve_queue() but then nothing.... |
| 10:17 |
Dyrcona |
queue and existing_queue are null.... |
| 10:19 |
pinesol |
News from commits: LP#1729889: Complete password field virtualization <http://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=659a3753c2ca3c9c7d485c358d19aaffa7fd9f41> |
| 10:20 |
Dyrcona |
holdings_profile isn't set in the params. That's what I'll look into next. |
| 10:22 |
Dyrcona |
Huh. That parameter isn't used apparently. |
| 10:23 |
Dyrcona |
Oh. It's import_attr_def.. Can be NULL. |
| 10:27 |
* Dyrcona |
is stumped.... No errors in the log, it just fails. |
| 10:28 |
Dyrcona |
No errors in the postgresql log for this queue. |
| 10:30 |
Dyrcona |
Ah ha! I think I've got this one.... The params have "queue_name" but resolve_queue() is looking for "new_queue_name." |
| 10:40 |
csharp_ |
didn't realize how busted this thing is! |
| 10:40 |
csharp_ |
we've been running it fine for our use case |
| 10:41 |
Dyrcona |
I don't know if the params problem is a bug or user error. |
| 10:42 |
Dyrcona |
An acq import worked just fine. |
| 10:42 |
Dyrcona |
Oh. I think I see why email isn't working, too. |
| 10:44 |
Dyrcona |
The event's not active. |
| 10:49 |
|
Christineb joined #evergreen |
| 10:51 |
|
sandbergja joined #evergreen |
| 11:00 |
|
sandbergja joined #evergreen |
| 12:00 |
Dyrcona |
Lp 2126669 |
| 12:01 |
pinesol |
Launchpad bug 2126669 in Evergreen 3.15 "Background vandelay import does not work for authorities" [Undecided,New] https://launchpad.net/bugs/2126669 |
| 12:43 |
Dyrcona |
Ugh.. I hate when I see a typo in one of my commit messages. |
| 13:04 |
* Dyrcona |
resists the urge to amend the commit. |
| 13:10 |
|
mantis joined #evergreen |
| 13:10 |
mantis |
sorry for the lateness, DIG meeting info for today has been sent and updated on the Wiki |
| 15:08 |
|
mantis left #evergreen |
| 15:11 |
csharp_ |
"by the time mantis announced that, DIG had already become DUG"</dadjokefromhell> |
| 15:45 |
berick |
*golf claps* |
| 16:00 |
* csharp_ |
bows |
| 16:09 |
Bmagic |
has anyone tackled the issue with emailed reports and the generic link to the report? Meaning: staff gets an email, they click the link and it prompts them to login, even though they are already logged in. The reason: the link doesn't use their specific subdomain |
| 16:10 |
Bmagic |
In addition: the login mechanism is once removed from the new Angular login procedure. |
| 16:45 |
eeevil |
Bmagic: no, nobody has. this is one reason for staff to just log into a single hostname. their workstation identifies their location, that's its purpose. |
| 17:14 |
mmorgan |
We did have some libraries login to the client using a subdomain. Doing so set search scope and preferred library in the tpac, avoiding the need to set those as workstation settings. |
| 17:18 |
|
stompro joined #evergreen |
| 17:19 |
|
mmorgan left #evergreen |
| 17:24 |
|
jonadab joined #evergreen |
| 19:32 |
|
sandbergja joined #evergreen |
| 20:02 |
|
stompro joined #evergreen |
