| Time |
Nick |
Message |
| 03:19 |
|
jaswinder joined #evergreen |
| 06:32 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 07:20 |
|
jaswinder joined #evergreen |
| 07:35 |
|
rlefaive joined #evergreen |
| 07:41 |
|
rjackson_isl joined #evergreen |
| 07:51 |
|
agoben joined #evergreen |
| 08:08 |
|
rlefaive joined #evergreen |
| 08:13 |
|
dickreckard joined #evergreen |
| 08:14 |
dickreckard |
hi all |
| 08:18 |
|
Dyrcona joined #evergreen |
| 08:29 |
dickreckard |
maybe someone can help.. i am working at a new evergreen installation.. we are trying to batch import marc21 records with holdings attached, after failing because some fields of the import profiles where not correct, can't anymore delete queues and get Event: 2001:DATABASE_UPDATE_FAILED -> The attempt to write to the DB failed |
| 08:31 |
Dyrcona |
dickreckard: Look in the PostgreSQL logs around that time for the actual error. |
| 08:31 |
dickreckard |
2018-04-06 14:20:17 CEST [5655-7] evergreen evergreen2 ERROR: record "new" is not assigned yet |
| 08:31 |
dickreckard |
2018-04-06 14:20:17 CEST [5655-8] evergreenevergreen2 DETAIL: The tuple structure of a not-yet-assigned record is indeterminate. |
| 08:31 |
dickreckard |
2018-04-06 14:20:17 CEST [5655-9] evergreenevergreen2 CONTEXT: PL/pgSQL function vandelay_import_item_imported_as_inh_fkey() line 3 at IF |
| 08:31 |
dickreckard |
SQL statement "DELETE FROM vandelay.import_item WHERE record = OLD.id" |
| 08:31 |
dickreckard |
PL/pgSQL function vandelay.cleanup_bib_marc() line 8 at SQL statement |
| 08:31 |
dickreckard |
SQL statement "DELETE FROM ONLY "vandelay"."queued_bib_record" WHERE $1 OPERATOR(pg_catalog.=) "queue"" |
| 08:32 |
dickreckard |
ups sorry will use pastebin next time |
| 08:39 |
Dyrcona |
dickreckard: The error is on a trigger that expects to fire during an import, and you say it is firing when you try to delete? |
| 08:41 |
|
kmlussier joined #evergreen |
| 08:43 |
pastebot |
Someone at 64.57.241.14 pasted "2018-04-06 14:30:57 CEST [6584" (9 lines) at http://paste.evergreen-ils.org/1730 |
| 08:44 |
|
mmorgan joined #evergreen |
| 08:44 |
dickreckard |
it seems to still try to import when trying to delete the queue? |
| 08:46 |
Dyrcona |
That's what it looks like. You may have to delete something else, first, but I'm not the Vandelay expert. |
| 08:47 |
kmlussier |
I haven't done much with vandelay in a while, but, usually, you can just delete a queue once you're ready to get rid of it. |
| 08:48 |
|
ngf42 joined #evergreen |
| 08:48 |
Dyrcona |
kmlussier: You missed the part where there were problems with the import. |
| 08:48 |
dickreckard |
its not possible to delete import items individually |
| 08:48 |
dickreckard |
or i dont see where |
| 08:49 |
Dyrcona |
It's possible in the database, I believe. |
| 08:57 |
|
bos20k joined #evergreen |
| 08:58 |
|
rlefaive joined #evergreen |
| 09:00 |
dickreckard |
evergreen2=# DELETE FROM vandelay.import_item; |
| 09:00 |
dickreckard |
ERROR: record "new" is not assigned yet |
| 09:03 |
Dyrcona |
Yeah, that trigger again. |
| 09:04 |
Dyrcona |
Is there anything in vandelay.import_item ? |
| 09:05 |
Dyrcona |
I really don't know why this is happening. |
| 09:05 |
dickreckard |
yea like 90 failed imports |
| 09:06 |
dickreckard |
but somehow cant delete them. something with that tuple structure of a not-yet-assigned record which is indeterminate ?:) |
| 09:08 |
Dyrcona |
Well, my take is that trigger is meant to run when records are imported only. Since you're deleting, it should not be firing. |
| 09:08 |
Dyrcona |
Because you are deleting there is no NEW record for it to look at. |
| 09:09 |
jeff |
dickreckard: within psql, if you run "\d+ vandelay.import_item" are there any triggers listed other than the one named "inherit_import_item_imported_as_fkey", and does it mention "DELETE" or is it just the expected "AFTER INSERT OR UPDATE ON"? |
| 09:09 |
Dyrcona |
It is also only supposed to happen on UPDATE or INSERT. |
| 09:09 |
jeff |
dickreckard: also, what version of Evergreen are you working with? |
| 09:09 |
dickreckard |
3.0.1 |
| 09:10 |
rhamby |
I hadn't seen that with Vandelay but I've seen that with a few triggers in various place that were addressed in various early 3.0.x versions |
| 09:10 |
Dyrcona |
Looking at master, the trigger only fires on update or insert. |
| 09:10 |
dickreckard |
ah! |
| 09:10 |
dickreckard |
inherit_import_item_imported_as_fkey AFTER INSERT OR DELETE OR UPDATE ON vandelay.import_item DEFERRABLE INITIALLY IMMEDIATE FOR EACH ROW EXECUTE PROCEDURE vandelay_import_item_imported_as_inh_fkey() |
| 09:11 |
dickreckard |
delete too here! |
| 09:11 |
jeff |
bug 1724223 |
| 09:11 |
pinesol_green |
Launchpad bug 1724223 in Evergreen "Unable to delete copy notes / remove items from buckets, etc." [Medium,Fix released] https://launchpad.net/bugs/1724223 |
| 09:12 |
jeff |
or similar, wait. |
| 09:12 |
jeff |
bug 1152753 |
| 09:12 |
pinesol_green |
Launchpad bug 1152753 in Evergreen "Serial Units won't go into Copy Buckets" [Medium,Fix released] https://launchpad.net/bugs/1152753 |
| 09:12 |
jeff |
that one. |
| 09:13 |
jeff |
which incorporates a branch from bug 1708291 |
| 09:13 |
pinesol_green |
Launchpad bug 1708291 in Evergreen "webstaff: serials module" [Wishlist,Fix released] https://launchpad.net/bugs/1708291 |
| 09:13 |
jeff |
none of these bug titles seem relevant :-) |
| 09:14 |
jeff |
commit 4d96474 |
| 09:14 |
pinesol_green |
jeff: [evergreen|Galen Charlton] LP#1152753: fix fake FK trigger on vandelay.import_item - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=4d96474> |
| 09:14 |
jeff |
but interestingly, that just changes the trigger function, doesn't make it not run on delete. |
| 09:14 |
jeff |
where did that change come in... |
| 09:14 |
dickreckard |
hehe super |
| 09:15 |
dickreckard |
yes i have to get used to how to describe evergreen issues with the terms that one would find in the bug reports^^ |
| 09:16 |
Dyrcona |
dickreckard: https://pastebin.com/2mKsdmSv |
| 09:16 |
dickreckard |
but does that mean we would have to recreate the database with the patched sql files? |
| 09:17 |
dickreckard |
aah |
| 09:17 |
dickreckard |
Dyrcona: youre too fast |
| 09:17 |
Dyrcona |
I haven't tried that. Just off the top of my head. |
| 09:18 |
dickreckard |
constraint or trigger? |
| 09:20 |
Dyrcona |
Doesn't work. Probably needs to be drop trigger. |
| 09:21 |
rjackson_isl |
Once you change Copy Location Order (requires ADMIN_COPY_LOCATION_ORDER permission) where in Evergreen does this change show? It doesn't impact the list of locations in advanced search nor in Edit Item attributes? |
| 09:22 |
kmlussier |
rjackson_isl: the holds pull list |
| 09:22 |
rjackson_isl |
kmlussier - so that is by intent only to change there? |
| 09:22 |
kmlussier |
If you sort by copy location on the pull list, it should sort in that order. I'm not sure which holds pull list it works on in the xul client, but I'm fairly sure it was working in the web client at one time. |
| 09:22 |
kmlussier |
rjackson_isl: That's my understanding. |
| 09:23 |
rjackson_isl |
kmlussier++ |
| 09:23 |
|
yboston joined #evergreen |
| 09:23 |
Dyrcona |
dickreckard: I edited the paste to something that works, have another look. |
| 09:24 |
Dyrcona |
Well, almost. Look now. :) |
| 09:25 |
dickreckard |
yay |
| 09:25 |
dickreckard |
clap clap |
| 09:25 |
dickreckard |
thanks |
| 09:26 |
dickreckard |
so the trigger on delete would actually never be neede there? |
| 09:26 |
Dyrcona |
No, as you see, it blows up if called on delete. |
| 09:26 |
dbwells |
dickreckard: That is right. It is fixed in 3.0.2. |
| 09:29 |
dbwells |
dickreckard: There is actually a few other places this pops up, so upgrading is probably a good idea :) |
| 09:30 |
dickreckard |
hehe thanks for the tip. we are actually moving to 3.0.6 but we were trying migrating on the old server |
| 09:35 |
dbwells |
jeff: For the record (since I was curious), this broke in 1063, fixed in 1081. |
| 09:47 |
|
rlefaive joined #evergreen |
| 09:49 |
|
jvwoolf joined #evergreen |
| 10:04 |
|
mmorgan1 joined #evergreen |
| 10:15 |
dickreckard |
thanks a lot guys, you've been super helpful. will update to 3.0.6 but it's good to get familiar with the db structure, triggers and so on.. to get an idea how the infrastructure works. importing works! :) |
| 10:22 |
kmlussier |
dickreckard: Yay! |
| 10:23 |
kmlussier |
Dyrcona++ jeff++ rhamby++ dbwells++ |
| 10:27 |
dickreckard |
is it an irc feedback bot? :P |
| 10:45 |
|
jaswinder joined #evergreen |
| 10:51 |
Dyrcona |
dickreckard: Yes, we have a bot or two in the channel: pinesol_green and pastebot. |
| 10:52 |
|
Christineb joined #evergreen |
| 10:52 |
Dyrcona |
@karma |
| 10:52 |
pinesol_green |
Dyrcona: Highest karma: "gmcharlt" (282), "berick" (193), "kmlussier" (185), "Dyrcona" (149), and "dbwells" (135). Lowest karma: "comcast" (-32), "systemd" (-12), "^" (-9), "oracle" (-7), and "typos" (-7). You (Dyrcona) are ranked 4 out of 264. |
| 10:52 |
bshum |
Four I think -- egbuilder and serflog |
| 10:52 |
jeff |
dbwells++ thanks. i had just gotten back from being pulled away on a half dozen other things. :-) |
| 10:52 |
Dyrcona |
Oh, yeah. I forget about those two. |
| 10:53 |
bshum |
Though egbuilder, what have you done for me lately?! |
| 10:53 |
Dyrcona |
comcast++ # 'Cause they're sending someone to fix their line and I'm not even their customer. :) |
| 10:53 |
Dyrcona |
systemd-- # just because. :) |
| 10:54 |
|
yboston joined #evergreen |
| 10:55 |
kmlussier |
Dyrcona: I'm sure I'll have plenty of future opportunity to knock comcast back down to -32 again. |
| 10:56 |
Dyrcona |
Well, we will reset the karma after the conference.... |
| 10:56 |
bshum |
systemd-- # ditto |
| 10:56 |
Dyrcona |
oracle-- |
| 10:59 |
dickreckard |
ok gotta go. |
| 10:59 |
dickreckard |
cos |
| 10:59 |
dickreckard |
sun++ |
| 10:59 |
dickreckard |
:P |
| 11:01 |
* kmlussier |
scowls about the snow that's on the way. |
| 11:02 |
Dyrcona |
snow-- |
| 11:02 |
Dyrcona |
dickreckard++ # Just because. :) |
| 11:03 |
Dyrcona |
Now, where was I? Oh, yeah. Getting ready to install the code for my test upgrade branch on a vm. The prereqs should be done by now. |
| 11:22 |
berick |
cold and rainy here tomorrow. I'd rather have snow. :| |
| 11:23 |
kmlussier |
berick: Snow in April is never a good wno. |
| 11:23 |
kmlussier |
Wow! I really mangled that last word, didn't I? |
| 11:24 |
berick |
:) I wno what you mean! |
| 11:25 |
|
jaswinder joined #evergreen |
| 11:25 |
Dyrcona |
heh. |
| 11:26 |
Dyrcona |
It's always nice when you're doing a new installation and OpenSRf just works. :) |
| 11:26 |
Dyrcona |
Particularly, when you do it from memory. |
| 11:26 |
alynn26 |
I stil working on getting hatch to run on a couple of 32 bit Window 7 machines. The hatch Extension console is saying "Access to the specified native messaging host is forbidden". Any clues as to why the host is forbidden? |
| 11:29 |
pastebot |
"alynn26" at 64.57.241.14 pasted "Complete Hatch extension Console message" (10 lines) at http://paste.evergreen-ils.org/1740 |
| 11:37 |
berick |
alynn26: there appear to be some oustanding issues with the installer. either files are not getting installed or they don't have the right permissions. there are some debugging steps in bug #1756084 that might help narrow it down. |
| 11:37 |
pinesol_green |
Launchpad bug 1756084 in Evergreen "Printer List Empty In Hatch Printer Settings" [Undecided,New] https://launchpad.net/bugs/1756084 |
| 11:42 |
alynn26 |
Thanks, I been running through several steps and everything looks ok, just not connecting. I see what else I can find out, as I am troubleshooting for another library entirely. |
| 11:46 |
alynn26 |
had them run hatch.bat test, and it showed all his printers and hatch was set up correctly there. So, on to trying something else. |
| 11:47 |
|
beanjammin joined #evergreen |
| 11:51 |
jaswinder |
Hey Guys, I am running into an issue where my web client stopped working and I am getting internal Server Error. Nothing in logs. I have restarted everything yet no change. Any other areas where I can enable/see logs? |
| 11:53 |
berick |
jaswinder: internal server errors should produce apache error logs. e.g. /var/log/apache2/error.log |
| 11:55 |
jaswinder |
There is no error log file in that directory |
| 11:56 |
jaswinder |
LogLevel info |
| 11:56 |
jaswinder |
# - log locally |
| 11:56 |
jaswinder |
# CustomLog /var/log/apache2/access.log combined |
| 11:56 |
jaswinder |
# ErrorLog /var/log/apache2/error.log |
| 11:56 |
jaswinder |
# - log to syslog |
| 11:56 |
jaswinder |
CustomLog "|/usr/bin/logger -p local7.info" common |
| 11:56 |
jaswinder |
ErrorLog syslog:local7 |
| 11:56 |
jaswinder |
This is what apache settings for ErrorLog: |
| 11:58 |
jeff |
your apache error logs are being handed off to syslog, which may or may not be logging to disk. i'd try /var/log/syslog to start. |
| 11:58 |
jeff |
also, be aware that not all errors may be logged under that setup. |
| 11:59 |
jeff |
your messages may also (either exclusively or additionally) be in /var/log/messages, assuming a semi-typical debian/ubuntu setup. |
| 12:00 |
jaswinder |
Yeah, I noticed |
| 12:00 |
jaswinder |
I got the error logs. the default evergreen setup writes to syslogs. |
| 12:00 |
jaswinder |
thanks for looking |
| 12:01 |
jeff |
hopefully the error log entries are useful in pointing you to the problem. :-) |
| 12:08 |
|
kdt8 joined #evergreen |
| 12:15 |
|
mmorgan joined #evergreen |
| 12:25 |
Dyrcona |
See, I thought that patch was a bad idea. |
| 12:26 |
Dyrcona |
Just because we normally log to syslog for production, doesn't make it a good idea to always log to syslog. |
| 12:28 |
|
abowling joined #evergreen |
| 12:45 |
|
jihpringle joined #evergreen |
| 12:58 |
|
mmorgan joined #evergreen |
| 13:01 |
jaswinder |
Dyrcona: Agreed. |
| 13:15 |
|
yboston joined #evergreen |
| 13:18 |
|
rlefaive joined #evergreen |
| 13:27 |
|
hbrennan joined #evergreen |
| 13:32 |
|
abowling1 joined #evergreen |
| 13:49 |
|
yboston joined #evergreen |
| 13:49 |
hbrennan |
I'm failing to find the location of patron registration settings.. I need to add a field. Help? Thanks! |
| 13:51 |
kmlussier |
hbrennan: I don't think you can add a field that way. The settings only allow you to specify if it should be required or suggested. |
| 13:51 |
kmlussier |
hbrennan: Is it something you can add through a stat cat? |
| 13:51 |
hbrennan |
Oh sorry, not Library Settings |
| 13:51 |
hbrennan |
Just, where is the file |
| 13:52 |
hbrennan |
We had Equinox set it up for us when we migrated and we've never made changes so I'm not sure where that info is located |
| 13:52 |
kmlussier |
hbrennan: Web client or xul client? |
| 13:52 |
hbrennan |
Xul |
| 13:53 |
* kmlussier |
doesn't know the answer for the xul client. |
| 13:53 |
hbrennan |
kmlussier: But I bet you have all of webby mapped in your brain! :) |
| 13:54 |
berick |
templates are in Open-ILS/src/templates/actor/user/ |
| 13:54 |
berick |
register_table.tt2 is the main list of fields |
| 13:54 |
berick |
(note xul uses a dojo app) |
| 13:55 |
berick |
JS is @ Open-ILS/web/js/ui/default/actor/user/register.js |
| 13:56 |
hbrennan |
berick: Thanks! |
| 13:58 |
hbrennan |
hmm I don't have a src folder under openils... but I found the js |
| 14:00 |
hbrennan |
and found register_table.tt2 |
| 14:00 |
berick |
hbrennan: .. |
| 14:00 |
hbrennan |
thanks |
| 14:00 |
berick |
oh good |
| 14:00 |
berick |
sorry bout that, those were the repository paths, not install paths |
| 14:01 |
hbrennan |
no worries |
| 14:01 |
hbrennan |
and we have custom template folders that don't always line up |
| 14:16 |
|
jaswinde_ joined #evergreen |
| 14:18 |
|
rlefaive joined #evergreen |
| 14:18 |
jaswinde_ |
Hey guys, is there a way I can debug the code interactive using debugger? |
| 14:27 |
jaswinde_ |
Also, how do I get the user email using the patron barcode? |
| 14:35 |
|
jaswinder joined #evergreen |
| 14:35 |
|
jaswinder joined #evergreen |
| 14:35 |
jaswinder |
Hey guys, how do I get the user email using the patron barcode? |
| 14:41 |
berick |
jaswinder: depends on where the code is running. what are you trying to do? |
| 14:41 |
jaswinder |
I am trying to extend the EBookAPI. so, the code location /Application/EbookAPI/OneClickDigital.pm |
| 14:45 |
berick |
jaswinder: ok, so from within a server application. there are examples in other applications of finding patron data via barcode. |
| 14:45 |
berick |
for example, Actor.pm has an API call open-ils.actor.user.fleshed.retrieve_by_barcode |
| 14:46 |
berick |
it finds more than just the email, of course, but that should give you some ideas. you may also just want to call the API and let it do the work for you. |
| 14:50 |
jaswinder |
I think you are right. I am looking at the Actor.pm to see if I can get the email |
| 15:02 |
jaswinder |
How about the authenticated token from the same file ? |
| 15:02 |
jaswinder |
how I can get the authtoken? |
| 15:14 |
|
yboston joined #evergreen |
| 15:20 |
jaswinder |
Guys, the do_patron_auth inside the EBookAPI vendor files receives the barcode. Is it possible to send user email instead? |
| 15:35 |
dbwells |
jaswinder: If you want to just call the API berick mentioned, one way would be using 'simplereq()' in OpenILS::Application::AppUtils. |
| 15:36 |
dbwells |
Since you are familiar with using srfsh, it is very similar in how the arguments are managed. |
| 15:37 |
jaswinder |
dbwells: I can try the API call which is helpful for other data as well. However, Is there an easy way that I can pass an email to my vendor class? |
| 15:38 |
kmlussier |
@marc 040 |
| 15:38 |
pinesol_green |
kmlussier: The MARC code for or the name of the organization(s) that created the original bibliographic record, assigned MARC content designation and transcribed the record into machine-readable form, or modified (except for the addition of holdings symbols) an existing MARC record. These data and the code in 008/39 (Cataloging source) specify the parties responsible for the bibliographic record. (1 more message) |
| 15:39 |
dbwells |
jaswinder: a quick look and it seems like it wasn't designed with that in mind. Anything can be extended with enough thought, though. |
| 15:39 |
|
Dyrcona joined #evergreen |
| 15:39 |
|
Stompro joined #evergreen |
| 15:39 |
jeff |
hopefully libraries are notifying their patrons before passing data like library card numbers or patron e-mail addresses to vendors... |
| 15:40 |
jeff |
(and not making it a non-optional side-effect of browsing / searching the library catalog while logged in) |
| 15:40 |
jaswinder |
I see. this leaves with two options: Call an API from vendor class or pass email from EbookAPI.pm to the vendor class. |
| 15:40 |
jaswinder |
I have to think about which one will be easier in my case and less code |
| 15:42 |
kmlussier |
recordinformation seems like a good candidate to remove from our keyword blob. Is it there so that we can index the 001? |
| 15:42 |
dbwells |
jaswinder: The API call would be something like: my $usr = OpenILS::Application::AppUtils->simplereq('open-ils.actor', 'open-ils.actor.user.fleshed.retrieve_by_barcode', $auth, $barcode); |
| 15:43 |
jaswinder |
Thanks dbwells! I was just about to ask that. The issue is how do I get the $auth variable ? |
| 15:44 |
dbwells |
jaswinder: You would pass it along from the original authenticated request. Is the user already authenticated, and the one making the original requests? If so, you probably already have the user object. |
| 15:45 |
jaswinder |
The user is authenticated by EG I believe. I am writing the statement inside the initialize function of vendor class |
| 15:47 |
jaswinder |
dwells: Is that auth automatically assigned by some internal call? In this case, I need to pass in. how do I do that? |
| 15:48 |
dbwells |
jaswinder: What is the outermost API call from the client? |
| 15:48 |
jaswinder |
do_patron_auth |
| 15:51 |
dbwells |
AFAICT, the do_patron_auth method is just used internally. |
| 15:52 |
jaswinder |
Yes, I want to call external api by sending user email when the do_patron_auth is invoked on the vendor class |
| 15:53 |
dbwells |
Things have to be kicked off with something defined via 'register_method'. Maybe 'start_session'? |
| 15:56 |
dbwells |
I am trying to understand your actual workflow to see where we might best capture the authtoken, do the lookup, etc. |
| 15:57 |
jaswinder |
dbweels: I have the register_method setup. I can do two things here. Call this newly registered method that retreives unique information of user by email when the user logs in OR override the do_patron_auth method inside a vendor class |
| 15:58 |
jaswinder |
sorry I meant *dbwells |
| 16:04 |
dbwells |
jaswinder: No problem. Overall, though, I think you are right in your suspicions. The 'handlers' don't seem to have any access to the authtoken or at least an editor. I think that is something which needs to be rethought a bit. |
| 16:05 |
jaswinder |
dbwells: Agreed. So, how would I either get the auth value to pass to simplereq? |
| 16:11 |
dbwells |
Because you are a service you could call the internal APIs directly which need no authtoken. Basically the internals of user_retrieve_by_barcode minus the '$auth' and 'checkauth' bits. But that is really a workaround, and probably not the best long term. |
| 16:12 |
dbwells |
We wouldn't want to keep reinventing stuff just because we didn't bother to pass an authtoken along :) |
| 16:12 |
jaswinder |
okay, I can try without the $auth token |
| 16:15 |
dbwells |
jaswinder: I gotta go. There will be more folks around next week, of course. I, too, will probably be thinking more clearly on a Monday or Tuesday than on a late Friday afternoon. Take care. |
| 16:23 |
jaswinder |
dbwells: thanks |
| 16:42 |
|
abowling joined #evergreen |
| 16:53 |
|
abowling1 joined #evergreen |
| 17:04 |
|
mmorgan left #evergreen |
| 17:49 |
|
jvwoolf left #evergreen |
| 18:30 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 18:36 |
jeffdavis |
The rel_3_1 branch doesn't have a 3.0.x-3.1.0-upgrade-db.sql script it seems |
| 18:45 |
jeffdavis |
ah, it's not in master |
| 18:53 |
|
beanjammin joined #evergreen |
| 21:45 |
|
jaswinder joined #evergreen |
| 23:17 |
|
jaswinder joined #evergreen |
