| Time |
Nick |
Message |
| 07:14 |
|
rjackson_isl joined #evergreen |
| 07:21 |
|
agoben joined #evergreen |
| 07:35 |
JBoyer |
dbs, Funny story: central IT here disables incognito mode. Only in Chrome, mind; FF's Private Browsing and IE's InPrivate modes work just fine (I assume because they can't be disabled, or because our McAffee Internet Secure Surpeme, Enterprise Edition is disabled in Incognito...) |
| 07:36 |
|
JBoyer joined #evergreen |
| 07:38 |
|
kmlussier joined #evergreen |
| 07:59 |
kmlussier |
Good morning #evergreen! Happy Friday! |
| 07:59 |
kmlussier |
@coffee [someone] |
| 07:59 |
* pinesol_green |
brews and pours a cup of Honduras David Mancia, and sends it sliding down the bar to gmcharlt |
| 07:59 |
kmlussier |
@tea [someone] |
| 07:59 |
* pinesol_green |
brews and pours a pot of Bio Pao Chung Pouchong, and sends it sliding down the bar to RBecker (http://ratetea.com/tea/teehaus-bachfischer/bio-pao-chung-pouchong/7609/) |
| 08:37 |
|
Dyrcona joined #evergreen |
| 08:58 |
|
mmorgan joined #evergreen |
| 09:10 |
Dyrcona |
mmorgan: Just the person I wanted to "see" this morning! |
| 09:10 |
Dyrcona |
mmorgan: You run Syrup at NOBLE, yes? |
| 09:10 |
mmorgan |
Dyrcona: Nope, that's actually mdriscoll |
| 09:11 |
Dyrcona |
Well, could you find out from mdriscoll what Linux release it is hosted on and if it is configured with mod_python or mod_wsgi or some other way? |
| 09:11 |
Dyrcona |
It would help me out to know. |
| 09:11 |
Dyrcona |
I suppose I should ask the same question of dbwells. |
| 09:12 |
mmorgan |
Sure. |
| 09:12 |
Dyrcona |
Thanks muchly. |
| 09:14 |
|
mdriscoll joined #evergreen |
| 09:14 |
mmorgan |
Dyrcona: mdriscoll will jump in to answer shortly! |
| 09:14 |
mmorgan |
Or now :) |
| 09:15 |
Dyrcona |
mmorgan++ mdriscoll++ |
| 09:15 |
|
yboston joined #evergreen |
| 09:17 |
* mdriscoll |
is checking NOBLE's syrup install |
| 09:20 |
mdriscoll |
Dyrcona: We are running Syrup on Debian 7 (Wheezy) and opensrf was configured with --enable_python |
| 09:21 |
mdriscoll |
Dyrcona: I also had to apt-get intall python-setuptools and python-psycopg2 |
| 09:22 |
dbwells |
Dyrcona: We're using mod_wsgi on an EOL Ubuntu |
| 09:22 |
Dyrcona |
Yes, that sounds about right. |
| 09:22 |
Dyrcona |
dbwells: Thanks! |
| 09:23 |
Dyrcona |
mdriscoll: I assume you're using mod_ptyhon. You can tell by looking in the <Location/> blocks for a PythonHandler entry. |
| 09:23 |
Dyrcona |
beh. meant to fix that typo before hitting enter. |
| 09:24 |
Dyrcona |
dbwells: Would you be willing to share some of your configuration details for wsgi? |
| 09:24 |
Dyrcona |
I'm trying to get it working on Ubuntu 16.04 and looks like we will have to switch to wsgi. |
| 09:25 |
Dyrcona |
mdriscoll: If you upgrade the Syrup machine to anything more recent than Debian 7, you will need to switch to wsgi, too. |
| 09:25 |
mdriscoll |
Dyrcona: PythonHandler django.core.handlers.modpython |
| 09:25 |
dbwells |
Dyrcona: sure, I'll just post the whole site config file, one second |
| 09:26 |
Dyrcona |
mdriscoll: Thanks. That's definitely mod_python, but I think the presence of PythonHandler is enough. |
| 09:26 |
Dyrcona |
dbwells++ |
| 09:27 |
pastebot |
"dbwells" at 64.57.241.14 pasted "Apache config for syrup site" (23 lines) at http://paste.evergreen-ils.org/511 |
| 09:28 |
dbwells |
This is the first WSGI site we've ever had, so I am certainly no expert. |
| 09:28 |
jeff |
We're using mod_wsgi also. I've also had success running python apps under uwsgi + nginx (haven't tried syrup with that yet). |
| 09:28 |
dbwells |
Some of that stuff may have been added for debugging as well. |
| 09:31 |
jeff |
this morning's "how is that even working?" was a vendor making Patron API requests to us without SNI to a host where SNI should Matter. It worked purely because all nginx sites on that port are using the same wildcard certificate, so lack of SNI didn't cause a cert mismatch at the TLS layer, and HTTP Host header handled the rest. |
| 09:32 |
Dyrcona |
jeff: Thanks. I always forget that you use Syrup also. |
| 09:33 |
jeff |
But it will break if we move away from the wildcard certificate or start running any other virtualhosts that use their own cert on that ip+port, unless we make that vendor's target host the default and hope that they will be the only ones to lack SNI. |
| 09:33 |
Dyrcona |
dbwells++ |
| 09:33 |
jeff |
(or move them to their own port+ip) |
| 09:35 |
rjackson_isl |
anyone run into issues with permissions required to create report tempaltes in the web client? Same permissions (Cat2 group) allows creation in legacy staff client but aborts with "[PCRUDContext 0]: open-ils.pcrud.create.rt failed." in web client |
| 09:36 |
Dyrcona |
dbwells: Do you have a git repo with your Syrup modifications? I'm curious about wsgi.py in particular. |
| 09:41 |
dbwells |
Dyrcona: we're only tracking changes locally, but funny enough, wsgi.py isn't being tracked at all. Not sure why that is. It's pretty short, do you want it pasted? |
| 09:41 |
Dyrcona |
dbwells: That would be awesome, thanks! |
| 09:42 |
Dyrcona |
I also just opened the hektech github page. Think I'll star or follow it so I don't have to search next time. |
| 09:43 |
dbwells |
Yeah, we don't use that for too much, mostly for helping with other projects relying on Github. |
| 09:43 |
kmlussier |
While looking at the Print Item Label page, I was thinking that the field descriptions might work better if we had a help icon that displayed the descriptions on hover. Does anyone know if we already have similar help text in the web client? |
| 09:44 |
Dyrcona |
I've seen it before and probably didn't follow because of that reason, but I've followed, now. |
| 09:44 |
pastebot |
"dbwells" at 64.57.241.14 pasted "Syrup wsgi.py" (8 lines) at http://paste.evergreen-ils.org/512 |
| 09:45 |
dbwells |
That file has a vintage of 3 years ago, so I don't remember much details about it. |
| 09:45 |
dbwells |
I am sure we just copy-pasted it mostly from somewhere. |
| 09:46 |
Dyrcona |
Muchas gracias! That's more than enough to get me going. |
| 09:46 |
dbwells |
No problemo. |
| 09:47 |
mmorgan |
rjackson_isl: I'm taking a look at our report template permissions |
| 09:48 |
rjackson_isl |
thanks mmorgan - the localadmin group has no trouble. We compared settings (Jason and I) and on our test server weren't able to figure out what difference is causing the issue |
| 09:59 |
mmorgan |
rjackson_isl: We have a group containing the single permission CREATE_REPORT_TEMPLATE that works in the xul client. But it doesn't allow even allow accessing the folders in the web client. |
| 09:59 |
miker |
rjackson_isl: that error message doesn't mention permission failure ... is there more in the log? (not saying it's not permissions, mind) |
| 10:05 |
mmorgan |
Adding RUN_REPORTS at the system level allows me to access the folders, but not create a template. |
| 10:05 |
mmorgan |
Changing RUN_REPORTS to the consortium level allowed me to create a template. |
| 10:06 |
rjackson_isl |
miker - Jason indicated the way that the logic is coded (a 'c' module?) there isn't anything excaping to the logs |
| 10:06 |
rjackson_isl |
s/excaping/escaping |
| 10:07 |
rjackson_isl |
mmorgan - I have that one on my list of things to test/ tested - let me look closer |
| 10:07 |
miker |
mmorgan: that's because the permission is marked with global_required="true" in the IDL |
| 10:08 |
miker |
re consortium level perms |
| 10:09 |
miker |
rjackson_isl: sounds like you also need RUN_REPORTS at the top of the org tree |
| 10:10 |
miker |
see lines around 9586 in fm_IDL.xml |
| 10:12 |
rjackson_isl |
miker - I changed the perms on the failing Cat2 account to RUN_REPORTS Consortium but it still won't save - not sure I am following the top of the tree? (Sorry) |
| 10:13 |
rjackson_isl |
oh wait - the legacy staff client is filling it back - need to go thru the tables :( |
| 10:13 |
rjackson_isl |
s/filling/flipping |
| 10:14 |
miker |
rjackson_isl: top of the org tree |
| 10:14 |
miker |
the consortium level |
| 10:14 |
rjackson_isl |
miker++ - think I was missing the non-working legacy staff client updates to the account - gives the success msg then it reverts |
| 10:23 |
rjackson_isl |
miker++ mmorgan++ - problem solved |
| 10:23 |
mmorgan |
One down ;-) |
| 10:23 |
rjackson_isl |
:) |
| 10:23 |
rjackson_isl |
and perhaps one happy staff member! |
| 10:32 |
|
ejk joined #evergreen |
| 10:50 |
|
Christineb joined #evergreen |
| 11:09 |
|
mmorgan1 joined #evergreen |
| 11:23 |
|
mmorgan joined #evergreen |
| 11:25 |
gsams |
So I'm just now attempting to look over some old reports for adaptation in the new reporter and they won't load when I go to clone them. |
| 11:25 |
gsams |
I'm running 3.0.3 and all I get is a grey bar that says Loading... |
| 11:26 |
gsams |
The new report creator kicks pulls up just fine, it just looks like the xul creator doesn't initiate. |
| 11:27 |
gsams |
kicks in/ pulls up... one of those. |
| 11:28 |
|
yboston joined #evergreen |
| 11:37 |
|
rlefaive joined #evergreen |
| 11:40 |
|
jihpringle joined #evergreen |
| 11:48 |
mmorgan |
gsams: I'm seeing the same thing on our 3.0.3 test server, but not on our 2.12.4 production server. |
| 11:48 |
mmorgan |
I found lp 1642344, but it looks like the breakage is more severe in 3.0.3. |
| 11:49 |
pinesol_green |
Launchpad bug 1642344 in Evergreen "Web Staff Client - Problems cloning existing reports" [Medium,Confirmed] https://launchpad.net/bugs/1642344 |
| 11:56 |
gsams |
mmorgan: I went digging around in logs while trying to pull up the page and I'm seeing some file not found errors for a reports.js file. |
| 11:58 |
* kmlussier |
gets nervous as the lights start flickering. :( |
| 11:59 |
* mmorgan |
gets nervous as the wind picks up :-( |
| 12:02 |
gsams |
oof, that looks like a bit of a nasty storm. |
| 12:03 |
gsams |
hope things calm down, maybe I should check in on my sister. |
| 12:05 |
kmlussier |
gsams: Where is your sister? |
| 12:06 |
gsams |
Salem |
| 12:06 |
kmlussier |
Oh! Not far from mmorgan! |
| 12:06 |
gsams |
Just checked in on her, she's doing fine. |
| 12:06 |
gsams |
Everything is tied down! |
| 12:09 |
mmorgan |
Good! there are some roads there I'm planning to avoid later today, but I'm sure she's quite familiar with where those are :) |
| 12:20 |
gsams |
hmm, looks like I'll need to use the xul client to start this process. |
| 12:38 |
|
yboston joined #evergreen |
| 13:10 |
dbwells |
miker: I was toying with those same thoughts on bug #1738488 last night, and just posted a potential branch along the lines you were thinking: |
| 13:10 |
pinesol_green |
Launchpad bug 1738488 in Evergreen 3.0 "Web client: patron billing history results in long running query" [High,Confirmed] https://launchpad.net/bugs/1738488 |
| 13:10 |
dbwells |
http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/collab/dbwells/lp1738488_optimize_flattener_joins |
| 13:16 |
miker |
dbwells: ooo, fun. did you see a good reduction in join clauses, and better plans? |
| 13:17 |
dbwells |
Yes, much better. No more plans with 44 septillion rows! |
| 13:18 |
miker |
:) |
| 13:20 |
Dyrcona |
dbwells: client 192.168.100.100:40274] ImportError: No module named conifer.settings |
| 13:20 |
Dyrcona |
Do you have an example, that I could crib from? |
| 13:22 |
Dyrcona |
Oh wait. That's settings.py.... |
| 13:23 |
Dyrcona |
Hm.... |
| 13:27 |
Dyrcona |
Methinks I need to set the python path. |
| 13:27 |
* miker |
wants to define a new reltype of "must_have" that forces an inner join because there's an fkey on the field and it's not nullable... |
| 13:28 |
dbwells |
Dyrcona: I can say that 'conifer' is a folder in my "python-path" in my "WSGIDaemonProcess" config line. Is that true for you as well? |
| 13:28 |
Dyrcona |
It is, yeah. |
| 13:28 |
dbwells |
But yeah, seems like some kind of path issue. |
| 13:28 |
Dyrcona |
It's should be trying to load .../conifer/settings.py |
| 13:28 |
Dyrcona |
But ... isn't in the path. |
| 13:30 |
Dyrcona |
Oh, bummer.... |
| 13:30 |
Dyrcona |
WSGIPythonPath .... context server config. |
| 13:31 |
Dyrcona |
So, I need a separate apache installation for each library that uses Syrup.... |
| 13:31 |
Dyrcona |
Well, let me get it working for one of them. |
| 13:31 |
Dyrcona |
I was afraid of this. Psgi works in a similar way. |
| 13:32 |
Dyrcona |
And, no, I'm not gonna use Docker. |
| 13:32 |
Dyrcona |
:) |
| 13:33 |
Dyrcona |
But, hmm... Maybe I can get away with SetEnv PYTHONPATH on the directory? |
| 13:35 |
dbwells |
Dyrcona: I think you might not need WSGIPythonPath set at all if you have the path set on the WSGIDaemonProcess. At least that's how ours is working :) |
| 13:36 |
Dyrcona |
dbwells: I'm not using the daemon process, I left restrictembedded off because I'm trying to run more than one per server, but it looks like I will eventually fail at that. |
| 13:37 |
dbwells |
I do remember having a heckuva fun time when we tried to get a dev and live instance running together. |
| 13:38 |
Dyrcona |
Yeah. And for the logs, SetEnv doesn't work, at least not in the Directory stanza. |
| 13:42 |
Bmagic |
csharp: I know you do a bunch with logs. Do you have a magic grep that you use to tease out issues from osrfsys.log ? |
| 13:43 |
Dyrcona |
Well, this is getting somewhere, I suppose: There is no South database module 'south.db.postgresql_psycopg2' for your database. |
| 13:44 |
Dyrcona |
Think I'll try again with just 1 site. |
| 14:01 |
|
rlefaive_ joined #evergreen |
| 14:05 |
|
rlefaive joined #evergreen |
| 14:25 |
* Dyrcona |
paused his "Eurythmics Greatest Hits" play list. |
| 14:25 |
* Dyrcona |
starts over with "Here Comes the Rain." |
| 14:25 |
Dyrcona |
@weather |
| 14:25 |
pinesol_green |
Dyrcona: Methuen, MA :: Rain :: 43F/6C | Wind Chill: 32F/0C | Friday: Rain and wind. High 43F. Winds NNE at 25 to 35 mph. Chance of rain 100%. Rainfall may reach one inch. Winds could occasionally gust over 50 mph. Friday Night: Cloudy. Periods of rain and windy at times early. Snow may mix in. Low 36F. Winds N at 20 to 30 mph. Chance of rain 70%. |
| 14:26 |
pinesol_green |
[evergreen|Dan Wells] Translation updates - newpot - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=2c9489f> |
| 14:33 |
|
ohiojoe joined #evergreen |
| 14:51 |
|
jvwoolf joined #evergreen |
| 15:00 |
* Dyrcona |
fear kmlussier lost power. |
| 15:00 |
Dyrcona |
fears, even.... |
| 15:01 |
berick |
high wind alert here too, but calm so far today |
| 15:03 |
* Dyrcona |
is not having fun with django... |
| 15:07 |
|
jvwoolf1 joined #evergreen |
| 15:10 |
pinesol_green |
[evergreen|Dan Wells] Translation updates - po files - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c0c229a> |
| 15:18 |
pinesol_green |
[evergreen|Remington Steed] Docs: Fix/improve web client keyboard shortcuts - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=65c555a> |
| 15:18 |
pinesol_green |
[evergreen|Remington Steed] Docs: Add missing header underline - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=7a4ca90> |
| 15:21 |
dbs |
Dyrcona: you may be interested in virtualenv for separating python environments? Not sure that will help the apache situation, but I can't live without virtualenv |
| 15:24 |
Dyrcona |
dbs: Yeah, I've used virtualenv before with something I did with flask. |
| 15:24 |
Dyrcona |
This is my first time trying to integrate something with django and apache, and most of my issues seem to be changes in django since Syrup was written. |
| 15:25 |
Dyrcona |
I've not ever tried to integer python code in Apache before, either. |
| 15:25 |
Dyrcona |
I ran the flask thing standalone and it was more of a proof of concept anyway. |
| 15:32 |
|
abowling joined #evergreen |
| 15:37 |
Dyrcona |
I thought I was told this didn't need Evergreen installed with --enable-python? |
| 15:37 |
Dyrcona |
It's looking for Evergreen python modules. |
| 15:41 |
|
kmlussier joined #evergreen |
| 15:46 |
dbwells |
Dyrcona: I probably misremembered something there. We do not have --enable-python on our Evergreen server, but Syrup runs elsewhere, so I do not have a good memory of what OpenSRF pieces we did over there. Sorry! |
| 15:47 |
Dyrcona |
S'OK. I simply forgot it needed the libs, really. |
| 15:47 |
Dyrcona |
Just more moles to whack. :) |
| 15:49 |
Dyrcona |
I did get past the simplejson issue. I'll have patches later. |
| 15:50 |
kmlussier |
@love tethering |
| 15:50 |
pinesol_green |
kmlussier: The operation succeeded. kmlussier loves tethering. |
| 15:52 |
|
rlefaive joined #evergreen |
| 15:52 |
Dyrcona |
kmlussier: Welcome back! |
| 15:52 |
kmlussier |
Dyrcona: I have about 36% left on my battery, so I don't plan to be here long. |
| 16:10 |
Dyrcona |
Maybe this is obvious by now, but Syrup needs work to modernize it. |
| 16:14 |
|
jvwoolf joined #evergreen |
| 16:27 |
pinesol_green |
[evergreen|Jeff Davis] LP#1738064: Optionally suppress display of badges in OPAC - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=5e2e8d2> |
| 16:27 |
|
yboston joined #evergreen |
| 16:28 |
|
rlefaive joined #evergreen |
| 16:32 |
Dyrcona |
Well, that's as far as it goes today: ImproperlyConfigured: Creating a ModelForm without either the 'fields' attribute or the 'exclude' attribute is prohibited; form MembershipAdminForm needs updating. |
| 16:43 |
pinesol_green |
[evergreen|Remington Steed] Compile release notes for 3.1 beta - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=8e7e84f> |
| 16:58 |
|
mmorgan left #evergreen |
| 17:10 |
|
rlefaive joined #evergreen |
| 17:10 |
gmcharlt |
dbwells: I'm going to have a last-minute signoff for 1751520 in a few minutes; if you're not already in the throes of rolling the beta tarball, I recommend waiting for it |
| 17:11 |
dbwells |
gmcharlt: It is already built and testing, but I can run it again. |
| 17:13 |
gmcharlt |
the webstaff reporter interface won't load without it |
| 17:13 |
dbwells |
yeah, it's worth it, no problem |
| 17:18 |
dbwells |
gmcharlt: So, doing some basic testing, and getting this error in catalog search in web staff: Error: [$injector:unpr] Unknown provider: egAddCopyAlertDialogProvider <- egAddCopyAlertDialog <- egCirc Any ideas? |
| 17:18 |
gmcharlt |
dbwells: ok, I've finished my testing and signoff re bug 1751520 |
| 17:18 |
pinesol_green |
Launchpad bug 1751520 in Evergreen "Reporter doesn't load in web client" [High,Confirmed] https://launchpad.net/bugs/1751520 - Assigned to Galen Charlton (gmc) |
| 17:19 |
gmcharlt |
dbwells: looking at your 17:18:35 now |
| 17:23 |
gmcharlt |
dbwells: not immediately seeing it |
| 17:24 |
dbwells |
gmcharlt: ack, nevermind, I think it was a false alarm. I must have had some older JS cached, a force reload got by it. Sorry. |
| 17:24 |
gmcharlt |
no worries |
| 17:25 |
gmcharlt |
on the other hand, I did just see something else of note |
| 17:25 |
gmcharlt |
one moment |
| 17:29 |
dbwells |
okay |
| 17:30 |
gmcharlt |
dbwells: ok, real issue, but not worth holding up beta further; I'll LP it |
| 17:32 |
dbwells |
what does it relate to? |
| 17:37 |
pinesol_green |
[evergreen|Jason Boyer] LP1751520: Add contextMenu Module to Vendor Bundle - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=5cc3ce7> |
| 17:37 |
pinesol_green |
[evergreen|Galen Charlton] LP#1751520: add specify path to content-menu.js explicitly - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=1935f39> |
| 17:38 |
gmcharlt |
dbwells: its bug 1753005 |
| 17:38 |
pinesol_green |
Launchpad bug 1753005 in Evergreen "webstaff holdings view: call numbers without copy can trigger exception" [Medium,New] https://launchpad.net/bugs/1753005 |
| 17:41 |
dbwells |
gmcharlt: okay, thanks. I think you are right in suggesting we not hold up the beta. |
| 18:07 |
dbwells |
gmcharlt: It looks like the reporter interface still doesn't load quite right for me. Getting a 404 for "iframeResizer.contentWindow.min.js", then "iFrameResize is not defined", and the iframe stays small. I suppose I may be doing something wrong still. |
| 18:13 |
dbwells |
I think it is an unrelated, but similar problem. |
| 18:21 |
cesardv |
dbwells: I've also noticed that iframeresizer issue, something's def not wired right... I'm poking at it... will let you know if I figure it out |
| 19:10 |
dbwells |
Well, I give up. I got the iframeResizer.contentWindow.min.js to load by placing it in JS_FILES in webpack.config.js, but that only solved the first problem. |
| 19:41 |
dbwells |
Okay, I didn't give up, but I will now. |
| 19:42 |
dbwells |
Here are some more clues. |
| 19:43 |
dbwells |
Webpack messes with the iframeResizer.js file, and this causes the iFrameResize function to not be created on window. |
| 19:44 |
dbwells |
Maybe that is what webpack is supposed to do, but we somehow aren't using the whole stack quite right. |
| 19:44 |
dbwells |
I see these lines near the end of that js: |
| 19:44 |
dbwells |
if (typeof define === 'function' && define.amd) { |
| 19:44 |
dbwells |
define([],factory); |
| 19:45 |
dbwells |
become: |
| 19:45 |
dbwells |
if (true) { |
| 19:45 |
dbwells |
!(__WEBPACK_AMD_DEFINE_ARRAY__ = [], __WEBPACK_AMD_DEFINE_FACTORY__ = (factory),... |
| 19:45 |
dbwells |
way out of my league... |
| 20:09 |
|
jvwoolf joined #evergreen |
| 22:36 |
|
yar joined #evergreen |
