| Time |
Nick |
Message |
| 07:21 |
|
collum joined #evergreen |
| 08:00 |
|
BDorsey joined #evergreen |
| 08:29 |
|
mantis1 joined #evergreen |
| 08:38 |
|
rfrasur joined #evergreen |
| 08:45 |
|
Dyrcona joined #evergreen |
| 08:57 |
|
kworstell-isl joined #evergreen |
| 08:59 |
|
kworstell_isl joined #evergreen |
| 09:00 |
|
mmorgan joined #evergreen |
| 09:10 |
mantis1 |
Has anyone been getting bouncebacks texts from AT&T numbers? |
| 09:10 |
mantis1 |
for some reason that's been an issue even after changing the gateway |
| 09:11 |
Dyrcona |
You're trying to use the MMS rather than SMS gateway? |
| 09:32 |
|
Christineb joined #evergreen |
| 09:35 |
|
dbriem joined #evergreen |
| 09:56 |
|
Stompro joined #evergreen |
| 10:03 |
jeff |
for those here, is cost or code the primary barrier to using something other than email-to-sms gateways? |
| 10:03 |
jeff |
I'm curious how well-used / accepted a feature to use more reliable options (which have a per-message cost) would be. |
| 10:11 |
mmorgan |
jeff: Another obstacle may be tuits. |
| 10:13 |
|
dbriem joined #evergreen |
| 10:16 |
|
dguarrac joined #evergreen |
| 10:17 |
Dyrcona |
I would be OK with an optional module to use a paid service added so long as the email gateways remained the default. |
| 10:18 |
Stompro |
jeff, we made the switch many years ago, so no barrier here. email-to-sms was way too much work. |
| 10:18 |
Dyrcona |
jeff: Do I remember correctly that you use Twilio? |
| 10:24 |
jeff |
we do, and Stompro uses flowroute, I believe. |
| 10:25 |
jeff |
Stompro: I knew you were in the same/similar situation as us with regard to not using email-to-sms gateways :-) |
| 10:25 |
Dyrcona |
If you want to add the code to mainline as options, I'd be cool with that. |
| 10:26 |
jeff |
I suspect that there are systems with large enough volume that moving to a per-message-cost model would not be a decision made lightly. |
| 10:26 |
jeff |
So I think some of the UI / preference bits would still need to permit that, which is unfortunate but understandable. |
| 10:27 |
Dyrcona |
Like CW MARS for instance... I don't want to force anyone to use something that has a cost per use or any cost basis either if there is a free alternative available. |
| 10:42 |
csharp_ |
@band add Proper Tamale |
| 10:42 |
pinesol |
csharp_: Band 'Proper Tamale' added to list |
| 10:43 |
Dyrcona |
@ana proper tamale |
| 10:43 |
pinesol |
Dyrcona: Temporal rape |
| 10:43 |
csharp_ |
oof |
| 10:43 |
Dyrcona |
Err... That's not good. |
| 10:43 |
csharp_ |
pinesol: bad |
| 10:43 |
pinesol |
csharp_: Sorry, that command is only available to Evergreen Premium™ Subscribers. Please upgrade your subscription ASAP! |
| 10:44 |
* Dyrcona |
apologizes for that slip up. |
| 10:46 |
Stompro |
We are paying $0.004 per sms sent with flowroute. So $4 per 1000. But, our messages are longer than 170, so it looks like we pay 0.014 per message, so we pay $14 per 1000 notices. I'm going to ask them about that. |
| 10:51 |
csharp_ |
just learned that @ana uses 10:42 < csharp_> @band add Proper Tamale |
| 10:51 |
csharp_ |
10:42 < pinesol> csharp_: Band 'Proper Tamale' added to list |
| 10:51 |
csharp_ |
oops |
| 10:51 |
csharp_ |
I just learned that @ana uses a site called anagram-genius.com to generate the anagrams |
| 10:52 |
csharp_ |
was going to see if we could easily have a banned list of words, but we're at their mercy |
| 10:53 |
csharp_ |
re: SMS, PINES would not be subscribing to a service unless the providers shut down the email gateways |
| 10:55 |
csharp_ |
mantis1: several possibilities, including SPF records (or lack thereof) - here's our documentation on SPF: https://pines.georgialibraries.org/dokuwiki/doku.php?id=admin:settings:spf |
| 10:55 |
Stompro |
Ah, per carrier surcharges. |
| 10:56 |
Dyrcona |
A free option should be the default as long as it exists. |
| 10:56 |
csharp_ |
+1 |
| 10:57 |
* Dyrcona |
runs off to pickup lunch early to avoid the noon rush. |
| 11:33 |
dbriem |
scottangel: ng-model="strict_barcode" on that checkbox sets up two-way data binding |
| 11:33 |
dbriem |
if checkbox is checked/unchecked, $scope.strict_barcode on the controller is set to true/false |
| 11:33 |
dbriem |
no manual flip needed with automagical two-way data binding (for better or worse) |
| 11:33 |
dbriem |
ng-change directive triggers onStrictBarcodeChange() just to update the wks pref |
| 11:45 |
Stompro |
Dyrcona, NCIPServer related, can I run two instances of the NCIP server on one server? I'm trying to have a unique oils_ncip.xml for each system. |
| 11:46 |
Stompro |
Here is my apache config https://gitlab.com/LARL/ncipserver/-/snippets/2515355 |
| 11:47 |
Dyrcona |
Stompro: I'd have to look at the code to double check, but probably not without changes. |
| 11:48 |
Stompro |
Maybe if I have a completely separate /home/opensrf/NCIPServer setup? Adjusting the paths as needed to allow that to work? |
| 11:49 |
Dyrcona |
IDK. It looks like what you are doing should work. I've not tried to run more than 1 at a time, though I did have that in mind at the beginning, because there's the possibility of having multiple drivers. |
| 11:50 |
Dyrcona |
I'd really like to get away from Dancer. That wasn't my idea, and I was trying to be compatible with another implementation. That lasted all of 0 time. |
| 11:51 |
Stompro |
Thanks, I'll keep poking at it as it is and see if I can figure out what is going wrong. Requests to the second endpoint are either timing out or erroring out. |
| 11:51 |
Dyrcona |
Dancer and Plack may share some environment that's breaking things, but I don't know for certain. |
| 11:53 |
|
kworstell-isl joined #evergreen |
| 11:55 |
Stompro |
Hmm, maybe I just have a config setting wrong. |
| 11:55 |
|
jvwoolf joined #evergreen |
| 12:01 |
|
jihpringle joined #evergreen |
| 12:05 |
Stompro |
Dyrcona, just had the wrong password set, oops. Seems to work fine. |
| 12:05 |
Dyrcona |
Heh. I'm glad it was simple. |
| 12:22 |
|
dguarrac joined #evergreen |
| 12:38 |
|
dbriem joined #evergreen |
| 12:45 |
|
dbriem joined #evergreen |
| 12:50 |
|
dbriem joined #evergreen |
| 12:52 |
|
collum joined #evergreen |
| 13:01 |
|
dbriem joined #evergreen |
| 13:04 |
|
mantis1 joined #evergreen |
| 13:09 |
jeff |
csharp_: yeah, I figured that PINES wouldn't be too keen on changing SMS methods. |
| 13:10 |
jeff |
it will be interesting to see how long it takes carriers to shut down the gateways completely, vs the current ever-degrading status quo of the gateways mostly-existing, but frequently failing to function. |
| 13:11 |
jeff |
(that assessment is based on second-hand observation and hearsay, since we thankfully don't have to deal with it) |
| 13:11 |
* jeff |
ducks |
| 13:13 |
Dyrcona |
Uh-huh. |
| 13:17 |
|
jvwoolf joined #evergreen |
| 13:34 |
|
tsadok joined #evergreen |
| 13:54 |
|
Bmagic joined #evergreen |
| 13:54 |
Bmagic |
Did everyone get kicked out of IRC or just me? |
| 13:55 |
Dyrcona |
Bmagic: Just you. |
| 13:55 |
Bmagic |
hmm, well, it seems to be working now. Did anyone see/respond to my question about the error? |
| 13:56 |
jeffdavis |
I don't think your question came through. |
| 13:56 |
Bmagic |
looking at the IRC logs, I don't see my message |
| 13:56 |
Bmagic |
Anyone got an idea on this: MarkItemLost trigger looks standard, but when it runs, I look at the logs and OpenSRF throws an error when trying to retrieve the copy: ERR :3391991:oils_sql.c:2801:] open-ils.cstore: Expected string or number within IN list; found JSON_NULL |
| 13:56 |
Bmagic |
Did that come through ^^^ |
| 13:57 |
berick |
Bmagic: yep |
| 13:57 |
Bmagic |
sweet |
| 13:57 |
Bmagic |
and it's in our IRC logs! Double surely there. And don't call me Surely |
| 14:00 |
berick |
Bmagic: if you search back in the logs you should see a cstore query with an empty array in there "where" part of the query |
| 14:01 |
Bmagic |
berick: I think this is the line that you're asking about? https://pastebin.com/iQVBmNkS |
| 14:06 |
berick |
hm, no, i would expect an "IN" query.. a search or json query |
| 14:07 |
Bmagic |
I suppose that line might be related to some other activity |
| 14:08 |
Bmagic |
grepping for 1294376 (which is the PID I believe) |
| 14:08 |
Dyrcona |
An empty array would something like this: [] |
| 14:10 |
Bmagic |
alright, new paste: https://pastebin.com/9CYAtxAS |
| 14:11 |
Bmagic |
I figured there'd be an error in the postgres logs, but I'm not finding one |
| 14:15 |
Bmagic |
makes me think that the query didn't make it to the database |
| 14:17 |
Dyrcona |
Does grepping the log for "3391991" turn up anything else? |
| 14:18 |
Dyrcona |
You might want to actually search for ':3391991' because you might find a lot of unrelated stuff with just the number. |
| 14:18 |
Bmagic |
checking |
| 14:18 |
Bmagic |
nada |
| 14:18 |
Dyrcona |
OK. Thought it was worth having a look. |
| 14:18 |
Bmagic |
not sure where you got that number, it's not in my paste |
| 14:19 |
berick |
from the original error |
| 14:19 |
Dyrcona |
Yeah... |
| 14:19 |
berick |
:3391991:oils_sql.c:2801: |
| 14:19 |
Bmagic |
oh, I see it |
| 14:20 |
Bmagic |
well, in that case, yes, I have lots of logs with that PID (on the other machine where I pulled that error from) |
| 14:20 |
Bmagic |
I'm starting to think that the MarkItemLost trigger needs environment vars or params... |
| 14:20 |
berick |
Bmagic: wild guess, is 'target_copy' in the A/T environment for MarkItemLost? I wonder if the reactor is not anticipating that. |
| 14:21 |
berick |
jinx |
| 14:24 |
Bmagic |
this might be wrong: environment: target_copy.call_number |
| 14:24 |
Bmagic |
almost none of the other markitemlost triggers have environment |
| 14:26 |
* Dyrcona |
checks our trigger's environment. |
| 14:29 |
Dyrcona |
Our 28 day mark item lost trigger has no rows in action_trigger.environment. |
| 14:30 |
Bmagic |
well, it seems it's required for this definition, because now the error is: open-ils.trigger [WARN:3704270:CStoreEditor.pm:155:] editor[1|0] retrieve actor.user called with no ID... |
| 14:32 |
Bmagic |
I think I found it |
| 14:32 |
Bmagic |
there's a space in front of "editor" on the event_params |
| 14:32 |
Bmagic |
" editor" |
| 14:40 |
Dyrcona |
That might do it. |
| 14:41 |
berick |
Bmagic: do you have values for context_bib_path or context_item_path in the A/T? |
| 14:41 |
berick |
i'm not 100% on how those work, but they look kind of like they might expect environment data |
| 14:45 |
berick |
hm, i need to see what those do.. looks like we're about to have those values as well |
| 14:45 |
Dyrcona |
I was about to say that I don't see those. |
| 14:46 |
berick |
https://bugs.launchpad.net/evergreen/+bug/1207533 |
| 14:46 |
pinesol |
Launchpad bug 1207533 in Evergreen "Triggered event log times out for large-data sites" [Medium,Fix released] |
| 14:46 |
Dyrcona |
berick++ # I'll have a closer look, too. |
| 14:54 |
|
mmorgan1 joined #evergreen |
| 14:55 |
|
jihpringle joined #evergreen |
| 15:02 |
Bmagic |
berick: Dyrcona: The issue ended up being two fold: the editor value was preceeded by a space AND the environment was "target_copy.call_number" - which doesn't match up with MarkItemLost. Fixed the preceeding space AND* deleted that environment. It's working now |
| 15:03 |
berick |
awesome |
| 15:03 |
Bmagic |
thanks for being my sounding wall |
| 15:03 |
Bmagic |
berick++ Dyrcona++ |
| 15:38 |
Dyrcona |
Bmagic++ |
| 16:06 |
|
abowling joined #evergreen |
| 16:27 |
Dyrcona |
Speaking of action_triggers.... We've got one that apparently hasn't run since October, but it shares a granularity with another that has been running. |
| 16:41 |
Dyrcona |
"Use of uninitialized value in subtraction (-) at /usr/local/share/perl/5.26.1/OpenSRF/AppSession.pm line 952." My open-ils.trigger_stderr.log is full of that, which looks _start is never being set in the AppSession, so something is dying before it gets there. |
| 16:42 |
Dyrcona |
Oh well. Guess that's a problem for tomorrow. |
| 16:46 |
|
jvwoolf left #evergreen |
| 17:04 |
|
mmorgan left #evergreen |
| 19:34 |
|
jihpringle joined #evergreen |
