| Time |
Nick |
Message |
| 04:30 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:11 |
|
rjackson_isl joined #evergreen |
| 07:40 |
|
rlefaive joined #evergreen |
| 07:59 |
|
kmlussier joined #evergreen |
| 08:02 |
kmlussier |
Good morning #evergreen! Happy Friday! |
| 08:02 |
kmlussier |
@coffee [someone] |
| 08:02 |
* pinesol_green |
brews and pours a cup of People's Daily Espresso, and sends it sliding down the bar to npoolos |
| 08:02 |
kmlussier |
@tea [someone] |
| 08:02 |
* pinesol_green |
brews and pours a pot of Masala Chai, and sends it sliding down the bar to rashma (http://ratetea.com/tea/rishi/masala-chai/4495/) |
| 08:03 |
kmlussier |
Since it's Friday... |
| 08:03 |
kmlussier |
@dessert [someone] |
| 08:03 |
* pinesol_green |
grabs some banana split for wsmoak |
| 08:03 |
wsmoak |
:) |
| 08:06 |
|
littlet joined #evergreen |
| 08:10 |
|
littlet_ joined #evergreen |
| 08:16 |
|
Dyrcona joined #evergreen |
| 08:33 |
|
_adb joined #evergreen |
| 08:49 |
|
bos20k joined #evergreen |
| 08:50 |
|
agoben joined #evergreen |
| 09:12 |
gmcharlt |
@quote random |
| 09:12 |
pinesol_green |
gmcharlt: Quote #5: "<senator> the armenian regression sounds like a spy novel" (added by bshum at 03:44 PM, February 22, 2011) |
| 09:19 |
|
maryj joined #evergreen |
| 09:23 |
|
rlefaive joined #evergreen |
| 09:29 |
Dyrcona |
@dunno |
| 09:29 |
pinesol_green |
Dyrcona: have you tried local mean solar time for the named city as the reference point? |
| 09:30 |
Dyrcona |
Yay, for __DATA__ sections. I can include the input file in the program and make it look like I was super-productive. :) |
| 09:31 |
|
yboston joined #evergreen |
| 09:37 |
JBoyer |
Data statements bring back memories of BASIC from the 80s. :D |
| 09:38 |
kmlussier |
:) |
| 09:38 |
Dyrcona |
Basic programmers don't die. They just GOSUB WITHOUT RETURN. |
| 09:40 |
Dyrcona |
10 PRINT "Hi!" |
| 09:40 |
Dyrcona |
20 GOTO 10 |
| 09:40 |
Dyrcona |
:) |
| 09:40 |
Dyrcona |
Ok, back to writing bash. |
| 09:43 |
JBoyer |
Dyrcona++ |
| 09:45 |
Dyrcona |
POKE 0x34 0x5b 0x6c ... :) |
| 09:45 |
Dyrcona |
Yeah, those were the days.... |
| 09:50 |
Dyrcona |
POKE 755, 4 |
| 09:50 |
Dyrcona |
Turns text upside down on a 8-bit Atari. :) |
| 09:51 |
* Dyrcona |
wishes he had a print out of his space invaders clone for the Commodore Vic-20. |
| 09:51 |
* Dyrcona |
imagines the cassette tape it is stored on is unusable. |
| 09:56 |
Dyrcona |
And, there we go. Big URL purge scripted to start tonight. |
| 09:57 |
JBoyer |
gmcharlt, when you have a moment could you take a look at lp1693560 ? |
| 09:57 |
JBoyer |
or, bug 1693560 |
| 09:57 |
pinesol_green |
Launchpad bug 1693560 in Evergreen "UPDATE_MARC Permission Error in Search Catalog" [Undecided,New] https://launchpad.net/bugs/1693560 |
| 09:57 |
JBoyer |
that's better. |
| 09:58 |
Dyrcona |
Think it works with a space: lp 1693560 |
| 09:58 |
Dyrcona |
Lp 1693560 |
| 09:58 |
Dyrcona |
Hm... I thought that worked. |
| 09:59 |
JBoyer |
I wondered about that, or a space and a #, seems like bug is the easiest way to trigger it. |
| 09:59 |
JBoyer |
If not the easiest for me to remember, heh. |
| 09:59 |
Dyrcona |
Maybe the plugin changes? |
| 09:59 |
Dyrcona |
changed even. |
| 09:59 |
kmlussier |
JBoyer++ # Sleuthing for bug 1693560 |
| 09:59 |
pinesol_green |
Launchpad bug 1693560 in Evergreen "UPDATE_MARC Permission Error in Search Catalog" [Undecided,New] https://launchpad.net/bugs/1693560 |
| 10:00 |
kmlussier |
JBoyer: You can delete those local storage values by highlighting it, then clicking the X that appears to the left of the Filter box. |
| 10:01 |
JBoyer |
As soon as someone said "it always happens on this testing account" and it didn't happen on my browser it was really easy. |
| 10:01 |
* kmlussier |
can't help with the actual bug, but has had plenty of experience deleting those local storage keys. |
| 10:01 |
JBoyer |
Oh, hey, that's much simpler, heh. |
| 10:01 |
Dyrcona |
Sure, just as I connect to the training the server, the VPN closes on me. |
| 10:02 |
JBoyer |
"Good night, sweet crypt..." |
| 10:02 |
bshum |
Dyrcona: It should work with lp space. The problem was you tried to request a bug within the same timeframe that it had already retrieved it. There's a timer on it so that it doesn't continually spam the channel if people all reference the same bug number in the same couple of minutes |
| 10:02 |
Dyrcona |
OK. |
| 10:03 |
* Dyrcona |
waits for PgAdmin3 to crash as it tries to retrieve information about a db that it can't talk to because I asked it to disconnect. |
| 10:03 |
csharp |
@quote add < Dyrcona> Basic programmers don't die. They just GOSUB WITHOUT RETURN. |
| 10:03 |
pinesol_green |
csharp: The operation succeeded. Quote #166 added. |
| 10:05 |
Dyrcona |
instead of watiing: pkill pgadmin3 |
| 10:05 |
csharp |
Dyrcona: that's a very annoying aspect of pgadmin3 |
| 10:05 |
csharp |
I usually kill it with Alt-F2 / xkill |
| 10:05 |
Dyrcona |
yeah, it's like they think you'll only ever use over a LAN. |
| 10:05 |
Dyrcona |
Whatever you do, do not use its built SSH tunneling. |
| 10:06 |
csharp |
heh - I haven't tried that |
| 10:06 |
Dyrcona |
Building your own ssh tunnels on the command line and configuring those as servers is more reliable. |
| 10:06 |
csharp |
we have a vpn |
| 10:06 |
* kmlussier |
tests bshum's tip with lp 922128 |
| 10:06 |
pinesol_green |
Launchpad bug 922128 in Evergreen "Scrollbars non sunt multiplicanda praeter necessitatem" [Undecided,Triaged] https://launchpad.net/bugs/922128 - Assigned to Ben Shum (bshum) |
| 10:06 |
Dyrcona |
I have a vpn, but still need to tunnel the production servers. |
| 10:06 |
csharp |
sylvar++ |
| 10:06 |
kmlussier |
bshum: I think that one can be closed, BTW. :) |
| 10:07 |
Dyrcona |
I've not seen that in a long time. |
| 10:07 |
csharp |
like the Latin tongue itself, it should be eternally open |
| 10:07 |
bshum |
kmlussier: Hahahaha, oh wow, yeah we should get rid of that one :) |
| 10:07 |
bshum |
Even though I would miss it, yeah, hah |
| 10:07 |
csharp |
bug report Hall of Fame |
| 10:08 |
bshum |
I think that bug and the frog one from Sitka where they made a frog image to replace the hold transit image |
| 10:08 |
bshum |
Are my favorite bugs of all time |
| 10:08 |
bshum |
You know, cause holds "hop" to another branch |
| 10:12 |
csharp |
heh |
| 10:16 |
bshum |
https://bugs.launchpad.net/evergreen/+bug/763944 |
| 10:16 |
pinesol_green |
Launchpad bug 1229757 in Evergreen "duplicate for #763944 support distinct pop-up images for holds, transits, and hold transits" [Wishlist,Fix released] |
| 10:17 |
bshum |
Aww, it didn't link to the original cause of the duplicate |
| 10:17 |
Dyrcona |
Error closing log file: No space left on device <- That can't be good. |
| 10:17 |
bshum |
The bug title is the best "a frog would improve usability of transit slip" |
| 10:18 |
csharp |
we should totally put that frog on some EG merch |
| 10:18 |
Dyrcona |
hah. |
| 10:19 |
Dyrcona |
The /openils partition is full. :( |
| 10:19 |
csharp |
eww |
| 10:19 |
Dyrcona |
Time to truncate some logs. |
| 10:19 |
Dyrcona |
It's only training. |
| 10:19 |
csharp |
ah |
| 10:19 |
berick |
/closedils |
| 10:19 |
Dyrcona |
I had it log there instead of syslog for reasons. |
| 10:20 |
csharp |
berick++ |
| 10:20 |
Dyrcona |
osrfsys.log is 42GB. |
| 10:20 |
Dyrcona |
nothing a truncate won't fix. :) |
| 10:21 |
dbwells |
Dyrcona: Got you beat, dealt with a 66GB osrfsys.log just yesterday :) |
| 10:21 |
JBoyer |
berick++ |
| 10:22 |
Dyrcona |
dbwells++ |
| 10:22 |
csharp |
debug must be on? or no process to automate truncating/deletion? |
| 10:22 |
Dyrcona |
I left a test vm running over the weekend with everything being logged. Came in on Monday to find the 60GB space for it full, and no one had touched it. |
| 10:23 |
Dyrcona |
/home and /var are pretty close to full, too. |
| 10:24 |
* kmlussier |
just realized that we no longer have holds, transit, or other images on slips in the web client. :( |
| 10:25 |
csharp |
frog time! |
| 10:26 |
dbwells |
csharp: or both? ;) |
| 10:26 |
csharp |
(or maybe Xzibit) |
| 10:26 |
Dyrcona |
Looks I'm the culprit on /home. Wonder what I have in there that takes up 35GB? |
| 10:27 |
Dyrcona |
A database dump. That makes senses. |
| 10:28 |
Dyrcona |
17G/var/log/messages |
| 10:28 |
Dyrcona |
17G/var/log/messages.1 |
| 10:28 |
Dyrcona |
That just looks wrong, but.... |
| 10:30 |
Dyrcona |
I'm starting to think that logging everything to syslog should not be the default. |
| 10:30 |
Dyrcona |
Maybe we should go back? |
| 10:31 |
gmcharlt |
sanbergja++ # https://github.com/sandbergja/faculty_notifier |
| 10:32 |
abneiman |
sandbergja++ # indeed, what a great idea! I bet it's popular with faculty |
| 10:32 |
Dyrcona |
I think I should adjust my postgresql logging settings downwards and stop sending them to syslog on training. NOTE: I did not set this up originally. |
| 10:32 |
JBoyer |
Dyrcona, it's great so long as you're also using the recommended syslog config, but yes, it should not be dumped in messages. |
| 10:34 |
Dyrcona |
JBoyer: The default configs now send everything to syslogs, IIRC. I think we should set the defaults back to /openils/var/log and /var/log/apache2 with notes on setting up a syslog server for bigger sites. |
| 10:34 |
Dyrcona |
The defaults are not ideal for standalone installations and/or test vms. |
| 10:34 |
Dyrcona |
I know... I know... I should LP it. |
| 10:35 |
JBoyer |
Dyrcona, I know, but we also include some rsyslog configs that pull everything out of messages and puts them in /var/log/evergreen/* which is pretty handy. |
| 10:35 |
Dyrcona |
In the stock configs or you do that in Indiana? |
| 10:35 |
bshum |
They're stock examples. But not installed by default. |
| 10:35 |
Dyrcona |
I missed those. |
| 10:36 |
bshum |
Or configured in opensrf_core.xml to point at by default either. |
| 10:36 |
Dyrcona |
But, I have my eg logs going to /openils/var/log on this server. |
| 10:36 |
JBoyer |
OpenILS/examples/evergreen-rsyslog.conf |
| 10:36 |
Dyrcona |
the production machines go to a syslog server, of course. |
| 10:36 |
dbwells |
Just did a Google search for 'dyrcona pingest', got back "Showing results for percona pinterest" :) Thanks, Google. |
| 10:37 |
JBoyer |
Oh, but there is something we're doing here that is missing there. I have a &~ after each line in that file so the messages are discarded after being written to the evergreen location. |
| 10:37 |
Dyrcona |
This looks like it is the postgresql config. I was aware of it going to /var/log/messages before, but I think I need to stop it from logging all queries. :) |
| 10:37 |
bshum |
JBoyer: Oooh that sounds like it could be nice! |
| 10:37 |
Dyrcona |
heh. Google++ |
| 10:37 |
berick |
Dyrcona Pingest sound like a magical incantation |
| 10:37 |
csharp |
sandbergja++ |
| 10:38 |
* Dyrcona |
is a wizard. :) |
| 10:38 |
berick |
:) |
| 10:38 |
csharp |
@who is a wizard? |
| 10:38 |
pinesol_green |
_bott_ is a wizard. |
| 10:38 |
bshum |
I always suspected... |
| 10:38 |
Dyrcona |
:) |
| 10:39 |
_bott_ |
heh, if you only knew the panic I caused just moments ago by making a PHP typo that shut down a whole Wordpress site :P |
| 10:39 |
berick |
see, powers! |
| 10:39 |
bshum |
_bott_: That just reminds them of your powerful sorcerous ways |
| 10:39 |
JBoyer |
Bippity-Boppity-Blackholed-that-site! |
| 10:40 |
_bott_ |
do not anger the wizard! |
| 10:41 |
berick |
@band Bott Orc Army |
| 10:41 |
pinesol_green |
berick: git diff origin/hamster Fleshing children complete |
| 10:41 |
JBoyer |
Well, that error message goes to some strange places. |
| 10:42 |
berick |
JBoyer: that's odd. i added the 'git diff origin/hamster' -- no idea where the second part came from. |
| 10:42 |
berick |
well, i know where it came from, but not how it ended up there |
| 10:46 |
Dyrcona |
Do not disturb the dragon, for thou art crunchy and taste good with ketchup. |
| 10:46 |
* gmcharlt |
backs away from berick's necromancer hamsters... slowly |
| 10:47 |
csharp |
@dunno search hamster |
| 10:47 |
pinesol_green |
csharp: 2 found: #48: "git diff origin/hamster" and #51: "git diff origin/hamster Fleshing children complete" |
| 10:47 |
csharp |
@dunno search children |
| 10:47 |
pinesol_green |
csharp: 1 found: #51: "git diff origin/hamster Fleshing children complete" |
| 10:48 |
kmlussier |
http://irc.evergreen-ils.org/evergreen/2017-01-18#i_284566 |
| 10:49 |
kmlussier |
pinesol_green didn't know what to do with [DEBG] and added a @dunno there? |
| 10:49 |
pinesol_green |
kmlussier: I am only a bot, please don't think I'm intelligent :) |
| 10:49 |
pinesol_green |
kmlussier: I eat more coconut cream pie before breakfast than most people eat all day |
| 10:50 |
berick |
csharp++ kmlussier++ |
| 10:50 |
berick |
@DEBG |
| 10:50 |
pinesol_green |
berick: I see nothing, I know nothing! |
| 10:50 |
csharp |
kmlussier: that's exactly it |
| 10:50 |
berick |
ah |
| 10:50 |
berick |
yeah |
| 10:51 |
csharp |
@blame csharp |
| 10:51 |
pinesol_green |
csharp: csharp is NOT CONNECTED TO THE NETWORK!!! |
| 10:51 |
csharp |
so true |
| 10:51 |
kmlussier |
heh |
| 10:52 |
csharp |
we should kill off the factoid plugin that no one uses or updates (which is what results in the "I am only a bot..." response) |
| 10:53 |
bshum |
We could easily do so |
| 10:53 |
_bott_ |
mmm coconut cream pie |
| 10:53 |
csharp |
@who wants pie for lunch? |
| 10:53 |
pinesol_green |
cesardv wants pie for lunch. |
| 10:53 |
berick |
csharp: yes please |
| 10:54 |
berick |
re: factoid. well, and pie for lunch |
| 10:55 |
gmcharlt |
csharp is baking us all pies for lunch. got it. |
| 10:55 |
* kmlussier |
looks outside at the cold, rainy weather and contemplates getting chowdah for lunch. |
| 10:57 |
rhamby |
pie++ chsharp: you have the address right? |
| 10:57 |
rhamby |
csharp even ^^^ |
| 10:58 |
* cesardv |
this bot must be psychic psyquic... |
| 11:09 |
csharp |
pinesol_green: your mama |
| 11:09 |
pinesol_green |
csharp: http://cat.evergreen-ils.org.meowbify.com/ |
| 11:09 |
csharp |
hooray! |
| 11:09 |
bshum |
Heh, csharp++ |
| 11:09 |
csharp |
bshum++ # knowing stuff |
| 11:10 |
* bshum |
toys with other bot things |
| 11:10 |
bshum |
@cast movie Sneakers |
| 11:10 |
pinesol_green |
bshum: An error has occurred and has been logged. Check the logs for more information. |
| 11:10 |
bshum |
Nope :) |
| 11:10 |
bshum |
Guess that's no good |
| 11:11 |
bshum |
@band The Evergreeners |
| 11:11 |
bshum |
Hmm |
| 11:12 |
bshum |
@band The Evergreeners |
| 11:12 |
pinesol_green |
bshum: An error has occurred and has been logged. Check the logs for more information. |
| 11:12 |
bshum |
Hehe |
| 11:12 |
csharp |
@band add The Evergreeners |
| 11:12 |
pinesol_green |
csharp: Band 'The Evergreeners' added to list |
| 11:12 |
berick |
yay |
| 11:12 |
csharp |
bshum++ |
| 11:12 |
bshum |
Aha! |
| 11:13 |
csharp |
@band add Fleshing Children |
| 11:13 |
pinesol_green |
csharp: Band 'Fleshing Children' added to list |
| 11:14 |
Dyrcona |
@band help |
| 11:14 |
pinesol_green |
Dyrcona: The Evergreeners |
| 11:14 |
Dyrcona |
@help band |
| 11:14 |
pinesol_green |
Dyrcona: (band [add|remove|search {BAND}]) -- KA-RAAAAY-ZEE band names! Get one, add one, remove one, search! |
| 11:15 |
Dyrcona |
@band add The Armenian Regression |
| 11:15 |
pinesol_green |
Dyrcona: Band 'The Armenian Regression' added to list |
| 11:15 |
Dyrcona |
Had to! |
| 11:16 |
csharp |
@band |
| 11:16 |
pinesol_green |
csharp: Fleshing Children |
| 11:17 |
csharp |
@who will come out with me next weekend to see [band]? |
| 11:17 |
pinesol_green |
bos20k will come out with you next weekend to see Fleshing Children. |
| 11:19 |
Dyrcona |
Hm.. I wonder if pg_ctl was removed and pg_ctlcluster takes its place and the man pages are just out of date. |
| 11:19 |
Dyrcona |
I seem to have the latter and not the former. |
| 11:20 |
Dyrcona |
Anyway, I reloaded the config after making some changes. |
| 11:21 |
berick |
Dyrcona: same here on 16.04 / pg 9.6 |
| 11:21 |
|
Christineb joined #evergreen |
| 11:21 |
|
_bott_ joined #evergreen |
| 11:21 |
Dyrcona |
I have Pg 9.5 on Debian, installed from apt.postgresql.org, so I guess the man pages are just out of date. |
| 11:26 |
csharp |
so who is running 9.5+ in production? any issues of note? |
| 11:27 |
Dyrcona |
Not in production, yet, but training and all of my test dbs are on 9.5. |
| 11:28 |
Dyrcona |
If you install 0979 and 0982, then everything seems hunky dory. |
| 11:28 |
csharp |
we're upgrading over Labor Day and current plan is to stay on 9.4 - we're shy of being pioneers atm :-) |
| 11:28 |
Dyrcona |
We plan to upgrade to 9.5 when we get a new db server this summer, probably in July. |
| 11:29 |
Dyrcona |
I still have to actually spec. out the hardware. |
| 11:29 |
Dyrcona |
I've been hitting 9.5 pretty hard, running the tests, cron jobs, all kinds of stuff. |
| 11:29 |
csharp |
cool |
| 11:32 |
berick |
been running 9.6 on test vm's for a bit now. (also with 0979 and 0982). |
| 11:34 |
Dyrcona |
Going well, I take it? |
| 11:34 |
berick |
yeah, so far so good. |
| 11:34 |
bshum |
http://irc.evergreen-ils.org/evergreen/2017-01-26 |
| 11:35 |
bshum |
Looks like Bmagic is on at least 9.5 somewhere |
| 11:35 |
bshum |
csharp too ;) |
| 11:35 |
Bmagic |
yep 9.5 since October |
| 11:35 |
csharp |
bshum: yeah just on concerto-sized vms :-) |
| 11:36 |
bshum |
csharp: Eh, Labor Day is still awhile from now... You've got plenty of time to acquire that adventurous spirit :) |
| 11:37 |
csharp |
true dat |
| 11:38 |
Dyrcona |
I've been testing with copies of production data and with concerto. |
| 11:38 |
Dyrcona |
FWIW. |
| 11:39 |
Dyrcona |
Maybe I'll upgrade my test db server, dumbo, to 9.6. |
| 11:41 |
Bmagic |
anyone struggle to keep enough cstore drones for action triggers in the middle of the night? |
| 11:41 |
bshum |
How many did you set up with? |
| 11:41 |
Bmagic |
I'm at 85 |
| 11:41 |
Bmagic |
started at 45 |
| 11:41 |
Bmagic |
growing pains I guess |
| 11:41 |
bshum |
Didn't we bump it to like 200 back in the day Dyrcona? |
| 11:42 |
bshum |
For hold pull list weirdness |
| 11:42 |
Bmagic |
it can still run out of drones at 85, so, right now, I am at 110. Running the Daily manually now |
| 11:42 |
Dyrcona |
yeah, at MVLC we went to 100 or 200 mainly because Chelmsford's pull list would use something like 75 all by itself. |
| 11:42 |
Bmagic |
the number of "collected" state is over 4k and climbing |
| 11:42 |
bshum |
Bmagic: How much granularity are you using too? |
| 11:42 |
berick |
Bmagic: how many trigger processes are running? |
| 11:43 |
berick |
open-ils.trigger |
| 11:43 |
Dyrcona |
Bmagic: That's on a utility server that just runs drones for itself, right? |
| 11:43 |
Bmagic |
<max_children>15</max_children> |
| 11:43 |
Bmagic |
right, just the utility server is responsible for this |
| 11:44 |
Dyrcona |
You could add it to a "brick" if you have them and have it share drones, so more would be available. |
| 11:44 |
* Dyrcona |
checks how many cstores we run on the utility server. |
| 11:45 |
Bmagic |
maybe I need more trigger children, it's still running 2 hours later |
| 11:45 |
berick |
Bmagic: what does 'ps' show for trigger procs? |
| 11:45 |
Dyrcona |
That could be it. |
| 11:45 |
Bmagic |
no end in sight, collected is increasing |
| 11:45 |
Bmagic |
6 drones and one perl script |
| 11:45 |
berick |
then you have plenty of trigger procs |
| 11:45 |
csharp |
Bmagic: are you using granularity to break them up into batches? |
| 11:46 |
Bmagic |
I have the standard daily/weekly, and a few of my own. So, yes they are broken up.... but it might use a good auditing to see how many in each category. Might be time to break a big one up |
| 11:46 |
csharp |
we found if they we're granulated enough, they would choke and take forever |
| 11:46 |
bshum |
At Biblio, I broke up our preoverdues into four separate batches of granularity (preoverdues1, preoverdues2, etc. and randomly assigned different libs to different groups to have them separated) |
| 11:47 |
bshum |
We found that once or twice, the job went haywire and took out a whole slew of things due to lack of resources, or whatnot |
| 11:47 |
Dyrcona |
Looks like we run 20 trigger and 60 cstore. |
| 11:47 |
berick |
~5 triggers using 85 cstores suggests more cstore leakage, unfortunately. |
| 11:47 |
berick |
they should scale more or less linearly |
| 11:48 |
Dyrcona |
Yeah, we have split things up into multiple granularities, too. |
| 11:49 |
Bmagic |
Daily does have 112 definitions, that might be my problem |
| 11:50 |
Bmagic |
in theory though, why would breaking them up matter? If it's allowed to run as long as it takes.... |
| 11:50 |
csharp |
our crontab lines for A/T stuff: https://pastebin.com/GdVbw1zv |
| 11:50 |
* csharp |
realizes that's kind of messy and blushes a little |
| 11:50 |
bshum |
I feel like one time the process got so huge that it ate up all the memory and exploded our utility server. |
| 11:51 |
bshum |
So we started breaking things up into more manageable pieces at a time |
| 11:51 |
bshum |
And by one time, I mean several times over the years |
| 11:51 |
Bmagic |
it's possible that one trigger is still running when the next cron fires |
| 11:52 |
csharp |
it's fine if they run concurrently as long as the number of events being processed is manageable |
| 11:52 |
Bmagic |
csharp: it's not that bad |
| 11:55 |
Bmagic |
is 6522 events "manageable"? |
| 11:56 |
Bmagic |
when an event moves from "pending" to "collected", I notice that all* of the events need to make that move before it can "complete" - Does the code "collect" this into memory for each event? |
| 11:56 |
csharp |
Bmagic: how many processor cores and how much RAM on the utility server? |
| 11:57 |
Bmagic |
csharp: 30GB memory, 8 cores |
| 11:57 |
csharp |
we have 16 cores and 32G of RAM |
| 11:57 |
csharp |
that should be fine, probably, depending on what else is going on on that server |
| 11:58 |
berick |
Bmagic: yes, it "collects" them all into memory before they 'react' |
| 11:58 |
Bmagic |
CPU loads are low at the moment as it's running, less than 1 core utilization |
| 11:58 |
Bmagic |
ok, so the memory really could be my issue |
| 11:58 |
berick |
IIRC, it has to collect them all first so the grouping will work |
| 11:59 |
Bmagic |
no swap space |
| 11:59 |
csharp |
berick: right |
| 11:59 |
|
jihpringle joined #evergreen |
| 11:59 |
Bmagic |
total used free shared buff/cache available |
| 11:59 |
Bmagic |
Mem: 30882892 4168212 3383892 8686672 23330788 17556884 |
| 11:59 |
csharp |
that looks fine to me |
| 12:00 |
Bmagic |
used > total ? |
| 12:00 |
* csharp |
steps afk for a while |
| 12:00 |
berick |
Bmagic: extra digit :) |
| 12:00 |
Bmagic |
no, sorry, an order of magnatude off |
| 12:00 |
berick |
most of that is cached -- plenty of free memory |
| 12:00 |
berick |
well, usable |
| 12:02 |
Bmagic |
hmm, well, it's looking fine right now as I am running it manually |
| 12:02 |
Bmagic |
we will just have to see how it goes in the night |
| 12:06 |
berick |
grabbing 1042 & 1043 |
| 12:07 |
berick |
well, not the binary and |
| 12:08 |
berick |
oops, 1043 and 1044 |
| 12:10 |
Dyrcona |
hmm. trying to get the id from actor.org_unit_descendants for a bunch of ous at once and not having much luck. I swear that I've done this before, though. |
| 12:10 |
Dyrcona |
After lunch.... |
| 12:15 |
|
maryj joined #evergreen |
| 12:16 |
pinesol_green |
Showing latest 5 of 6 commits to Evergreen... |
| 12:16 |
pinesol_green |
[evergreen|Bill Erickson] LP#1672775 Action/Trigger purge PGTAP tests - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=71817cb> |
| 12:16 |
pinesol_green |
[evergreen|Bill Erickson] LP#1672775 Action/Trigger CRON-able purge script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=ea536c7> |
| 12:16 |
pinesol_green |
[evergreen|Bill Erickson] LP#1672775 Action/Trigger retention interval release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=36567bb> |
| 12:16 |
pinesol_green |
[evergreen|Galen Charlton] LP#1672775: add purge_at_events.srfsh to example crontab - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=6ded4c3> |
| 12:16 |
pinesol_green |
[evergreen|Bill Erickson] LP#1672775 Stamping A/T purge SQL upgrade - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=510d4fe> |
| 12:18 |
gmcharlt |
berick++ |
| 12:22 |
JBoyer |
berick++ |
| 12:23 |
JBoyer |
Oh, and I have a question that berick or gmcharlt may have a quick answer to. I want to base the destination of a link on the tt2 splash page on the workstation's system level ou ancestor. what in broad strokes do you think is the best way to find that? |
| 12:24 |
berick |
JBoyer: web staff? |
| 12:24 |
JBoyer |
yes |
| 12:25 |
JBoyer |
I |
| 12:25 |
JBoyer |
I've done very little with Angular so far, time to catch up. :/ |
| 12:31 |
pastebot |
"berick" at 64.57.241.14 pasted "for JBoyer -- untested" (26 lines) at http://paste.evergreen-ils.org/160 |
| 12:31 |
berick |
JBoyer: totally untested |
| 12:32 |
gmcharlt |
then {{sys_org.shortname()}} or whatever in the template |
| 12:33 |
berick |
ah yeah, that's down at the bottom |
| 12:33 |
JBoyer |
Ah, very glad I asked, I was probably going to go off in a totally different direction. |
| 12:33 |
JBoyer |
Untested or no, this will at least point me toward the docs I need to get it working. |
| 12:33 |
JBoyer |
berick++ |
| 12:33 |
JBoyer |
gmcharlt++ |
| 12:34 |
gmcharlt |
JBoyer: you are, of course, obliged to now create a system-level OU whose shortname is HOWDY |
| 12:34 |
berick |
we haven't made that a stock/required org yet? huh |
| 12:35 |
gmcharlt |
somehow nobody ever gets around to signing off on the pullrequest |
| 12:35 |
gmcharlt |
not sure why |
| 12:35 |
JBoyer |
It has one user, a sqeaky shark with a cowboy hat. |
| 12:36 |
gmcharlt |
JBoyer++ |
| 12:37 |
berick |
heh |
| 12:40 |
|
sandbergja joined #evergreen |
| 12:40 |
Dyrcona |
Guess I'll try a recursive with query... |
| 12:46 |
Dyrcona |
That works! :) |
| 13:01 |
|
Freddy joined #evergreen |
| 13:02 |
Freddy |
Hi there |
| 13:02 |
Freddy |
anyone here? |
| 13:02 |
Guest35285 |
FIrst time using this method of communication |
| 13:02 |
Guest35285 |
ok..anyone here, really need some help |
| 13:03 |
kmlussier |
Hi Guest35285! |
| 13:04 |
kmlussier |
Yes, a lot of us are around today. |
| 13:04 |
Guest35285 |
Just great. Well, here is what happens |
| 13:04 |
Guest35285 |
I´m studyng ILS for my university proyect |
| 13:05 |
Guest35285 |
I want to compare koha and evergreen, since those are the open source ILS most used un North America |
| 13:05 |
Guest35285 |
I was able to install Koha, but wasnt able to do the same with evergreen |
| 13:06 |
Guest35285 |
it is a little more..complex |
| 13:06 |
Dyrcona |
Well, theres your first point of comparison.... :) |
| 13:07 |
jonadab |
Installing Evergreen is actually a lot easier these days than it used to be, IMO. |
| 13:07 |
kmlussier |
Guest35285: Where did you have trouble in the installation? |
| 13:07 |
Guest35285 |
Im stuck in Part 6: Instalation instruccion |
| 13:07 |
Guest35285 |
It says..... cd /openils/var/web/xul |
| 13:08 |
Guest35285 |
But I after doing the ls -l in var... |
| 13:08 |
Guest35285 |
the web directory doesnt exist |
| 13:08 |
Guest35285 |
is there something im missing? |
| 13:08 |
Dyrcona |
Did you do the make install step before that? |
| 13:08 |
Guest35285 |
i was able to install OpenSRF |
| 13:08 |
Dyrcona |
Right. This is for Evergreen. |
| 13:09 |
Guest35285 |
Yes...i been following the instruction...I was doing good until I reach this point |
| 13:10 |
Guest35285 |
Just to clarify... There are 3 compotents that I have to download right? |
| 13:10 |
Dyrcona |
If the web directory is not there, something went wrong in the Evergreen make install step. |
| 13:11 |
Guest35285 |
The OpenSRF, the server and the client right? and then use the tar command....am I right? |
| 13:11 |
jonadab |
Well, there are also dependencies. |
| 13:12 |
Guest35285 |
you mean this?: make -f Open-ILS/src/extras/Makefile.install <osname> |
| 13:12 |
Guest35285 |
Im using ubuntu 14.04...so it is like this: make -f Open-ILS/src/extras/Makefile.install ubuntu-trusty |
| 13:13 |
Guest35285 |
the procedure was...according to the video... First download the server.... use the tar command... and then use the make -f Open-ILS/src/extras/Makefile.install ubuntu-trusty |
| 13:13 |
jonadab |
Right, that installs the dependencies. |
| 13:13 |
jonadab |
Or at least it's supposed to. |
| 13:14 |
jonadab |
You do have to watch the output for errors. |
| 13:14 |
Guest35285 |
uhm.... |
| 13:14 |
Guest35285 |
I didnt receive any errors |
| 13:16 |
Guest35285 |
or maybe I have to donwload the Windows Client before doing the make STAFF_CLIENT_STAMP_ID=rel_2_10_12 install |
| 13:16 |
Guest35285 |
what are your thoughts? |
| 13:16 |
kmlussier |
Guest35285: No, you don't have to download the client until after the installation is done. |
| 13:17 |
jonadab |
I've only attempted to mess with the Windows client once. And going forward, there'll just be Hatch for that anyway. |
| 13:18 |
Dyrcona |
Guest35285: You must have had errors somewhere if the /openils/var/web directory is missing. |
| 13:19 |
jonadab |
^ Probably when doing the make install step. |
| 13:19 |
jonadab |
(NOT the make -f path/to/Makefile.install distro step, which is dependencies.) |
| 13:19 |
Dyrcona |
Do the make install step again with "|& tee ~/install.log" added on the end. |
| 13:20 |
Guest35285 |
Uhm....what could I be missing.... The OpenSRF installation was successful since I cound use the osrf_control --localhost --start-all command... |
| 13:20 |
Dyrcona |
That will put all of the output into a file in the home directory called install.log |
| 13:20 |
Guest35285 |
Ill try |
| 13:20 |
Dyrcona |
You can then look through that to see any errors that might have gone by. |
| 13:21 |
jonadab |
Guest35285: Ok, so, there's a dependencies step, a build step, and an install step, plus some config stuff, for OpenSRF; and then there's all that again for Evergreen itself. |
| 13:25 |
Dyrcona |
Guest35285: I reallly recommend using the README from the archives for the instructions or using the ones linked on the download pages. |
| 13:27 |
JBoyer |
speaking of errors, I just ran into jeffdavis' 5/23 "server: died with error Can't use an undefined value as a symbol reference at .../Server.pm line 307." http://irc.evergreen-ils.org/evergreen/2017-05-23#i_306001 |
| 13:27 |
JBoyer |
Same service too, open-ils.search. |
| 13:27 |
JBoyer |
:( |
| 13:28 |
Guest35285 |
Yeah..the thing is...this is actually my first time attempting to install Evergreen. Im more used to Koha and PMB |
| 13:28 |
JBoyer |
:q |
| 13:28 |
Guest35285 |
But PMB is out of the question, and Peruvian libraries could really be benefited from this ILS |
| 13:29 |
Guest35285 |
I really appreciate your help guys. Let see how it goes |
| 13:30 |
Dyrcona |
JBoyer: Is that the same line that caused NCIPServer to crash? |
| 13:30 |
JBoyer |
No, that one was related specifically to Timeszones. This one is in write_child() |
| 13:31 |
Dyrcona |
JBoyer: I'm also tempted to set all fix released bugs for NCIPServer to fix committedd. |
| 13:31 |
JBoyer |
syswrite($child->{pipe_to_child}, $write_size . $xml); |
| 13:31 |
JBoyer |
specifically |
| 13:31 |
Dyrcona |
OK, no need to so specific about it. :) |
| 13:31 |
Dyrcona |
be, even.... |
| 13:31 |
Dyrcona |
It's Friday, yeah? |
| 13:31 |
bshum |
It had better be... |
| 13:31 |
JBoyer |
I'm (sometimes) quite thorough. ;) |
| 13:31 |
Dyrcona |
:) |
| 13:32 |
JBoyer |
And as for NCIPServer, since it's never really "released" that does make sense, yeah. |
| 13:32 |
berick |
JBoyer: is this during service startup or after it's been running a while? |
| 13:32 |
Dyrcona |
OK. I'll change them. There are only a handful. |
| 13:32 |
JBoyer |
Running for most of the day. It just hit 45 mins ago. |
| 13:33 |
JBoyer |
I'm looking through osrfsys.12.log to see what was going on but a whole lot can happen in one second. D: |
| 13:34 |
berick |
JBoyer: also check /openils/var/log/open-ils.search_stderr.log |
| 13:34 |
Dyrcona |
Guest35285: Everyone has trouble the first time with installing Evergreen. Keep trying and we're here if you need any more help. |
| 13:36 |
berick |
JBoyer: also grep for SIGPIPE in osrfsys |
| 13:37 |
Dyrcona |
SIGPIPE....I know you well... |
| 13:37 |
JBoyer |
berick, that appears to be almost entirely the usual ISBN searches throwing up: DBD::Pg::st execute failed: ERROR: syntax error in tsquery: "(9781441585196:*&1441585192:*:*)" |
| 13:37 |
JBoyer |
(the stderr log, that is) |
| 13:37 |
Dyrcona |
Some leave one end of a pipe open by mistake? |
| 13:38 |
Dyrcona |
Yeah, that doesn't look right.... |
| 13:38 |
JBoyer |
And no mention of SIGPIPE anywhere in the whole file. |
| 13:38 |
csharp |
@quote search SIGPIPE |
| 13:38 |
pinesol_green |
csharp: 1 found: #17: "* csharp hasn't smoked a good SIGPIPE since his..." |
| 13:38 |
csharp |
@quote get 17 |
| 13:38 |
pinesol_green |
csharp: Quote #17: "* csharp hasn't smoked a good SIGPIPE since his college days" (added by gmcharlt at 12:33 PM, October 28, 2011) |
| 13:39 |
JBoyer |
Dyrcona, that tsquery error message? As far as I can remember that's been around for years. |
| 13:39 |
berick |
seems like the child process is dying and getting reaped (closing the pipe) via out-of-band signal handler while the parent is in the middle of trying to write to the child. |
| 13:39 |
Dyrcona |
OK. I've not seen it before. |
| 13:39 |
JBoyer |
Not saying I don't want it to go away forever, mind. ;) |
| 13:39 |
berick |
after the signal handler completes, the write continues, and failes on a closed pipe. (just a theory, of course) |
| 13:40 |
berick |
might need a pipe check just before the syswrite -- or an evail on the syswrite |
| 13:40 |
berick |
er, eval |
| 13:42 |
JBoyer |
Something interesting I noticed in that file, Dyrcona, it looks like the ISBN searches that fall over are coming from Z39.50 searches, and the ISBN is sent as 12324567890123* and the * is offending someone's sensibilities. |
| 13:43 |
|
maryj joined #evergreen |
| 13:43 |
Dyrcona |
JBoyer: OK. I don't think tsquery like the *. |
| 13:44 |
Dyrcona |
berick: I've had similar fun with something written in C. |
| 13:45 |
|
Juan_ joined #evergreen |
| 13:45 |
Dyrcona |
JBoyer: This is OpenSRF 2.5, right? |
| 13:46 |
JBoyer |
Ah, that's one old mystery solved. search use the numeric search in the catalog for 9781250057495* get an instant "no hits" search for 9781250057495 and get bagpipes. Something to look into later (since all ISBN searches from some endpoints are failing, that's nice..) |
| 13:46 |
JBoyer |
Dyrcona, yes. |
| 13:46 |
JBoyer |
and 2.12 |
| 13:46 |
JBoyer |
(2.12.2-ish) |
| 13:46 |
berick |
JBoyer: and just to be sure, no memory exhaustion? (cuz then all bets are off) |
| 13:47 |
JBoyer |
Nope, Icinga makes a lot of noise about that long before the OOM killer makes a move. |
| 13:47 |
* berick |
nods |
| 13:48 |
berick |
so many products I don't know how to pronounce :) |
| 13:48 |
JBoyer |
I pronounce it like the cake topping, plus "guh" |
| 13:49 |
berick |
oh good, that's there I ended up |
| 13:49 |
berick |
Icinga the body electric |
| 13:50 |
Dyrcona |
Oh, socketpair.... |
| 13:50 |
Dyrcona |
berick++ |
| 13:56 |
Dyrcona |
Oh, Perl...Why dost thou always mess with the argument order of syscalls that you emulate? |
| 13:56 |
JBoyer |
There's more than one way to do it. |
| 13:56 |
JBoyer |
:p |
| 13:56 |
JBoyer |
(Sometimes, "it" is frustrate programmers) |
| 13:57 |
berick |
heh |
| 13:58 |
Dyrcona |
I notice we're not doing shutdown on the sockets, but close. I wonder if there is any real difference? (Probably not, but I'll look.) |
| 13:59 |
Dyrcona |
shutdown in a signal handler might help.... |
| 14:05 |
Dyrcona |
Well, I don't see any SIGPIPE errors on my 2.12 development/test vm. |
| 14:06 |
Dyrcona |
I just truncated the logs on training or I'd look there, too. |
| 14:06 |
Dyrcona |
I haven't updated to 2.12.2, yet. It is still about 2.12.1. |
| 14:07 |
|
Jillianne joined #evergreen |
| 14:11 |
Bmagic |
Which permission do staff need in order to override an item with holdable=false ? |
| 14:11 |
JBoyer |
I assume it's caused by a race condition that can't be reliably replicated in any case. Wrapping the syscall in an eval and dealing with the fallout is likely the best that can be done. |
| 14:11 |
JBoyer |
PLACE_UNFILLABLE_HOLD ? |
| 14:12 |
Dyrcona |
Bmagic: What JBoyer said. |
| 14:13 |
Dyrcona |
Race conditions don't look favorable today with a sloppy track from last night's rain. |
| 14:13 |
Bmagic |
for some reason, that doesn't do it... But let me run some simulations |
| 14:13 |
Dyrcona |
Bmagic: Could be the depth it's granted at. |
| 14:14 |
Bmagic |
yeah, tried that too, no worries, you gave me what I wanted, just making sure I was barking up the right tree |
| 14:16 |
Dyrcona |
Ah, no. |
| 14:16 |
JBoyer |
Bmagic, well, there are some situations where *nothing* will allow you to place an "impossible" hold. Very frustrating to run into that recently. |
| 14:16 |
Dyrcona |
That permission lets patrons/staff place holds that won't fill right now because of age hold protection. |
| 14:17 |
Bmagic |
it's the Everything permission that is working, but we dont want to grant that to circ staff |
| 14:21 |
Dyrcona |
You know if you place that hold, the target won't target it, right? It would have to be a copy hold, and even then I'm not sure it would fill. |
| 14:22 |
csharp |
could be related to the workstation too - the call is "$place_unfillable = 1 if $e->allowed('PLACE_UNFILLABLE_HOLD', $e->requestor->ws_ou);" |
| 14:23 |
Dyrcona |
Interestingly, I don't see any events thrown in Holds.pm for copy->hodable = false |
| 14:24 |
JBoyer |
copy->hodor? |
| 14:24 |
berick |
haha |
| 14:28 |
Dyrcona |
:) |
| 14:28 |
Dyrcona |
A force hold should work. Not sure why the everything permission works, but holds code is complicated. |
| 14:30 |
JBoyer |
Dyrcona, well, Everything would include HOLD_FORCE or whatever it's called, which I didn't think was commonly given to staff. |
| 14:30 |
Dyrcona |
Yes. |
| 14:31 |
Dyrcona |
It looks like if you attempt to place a title hold with no holdable copies you get HIGH_LEVEL_HOLD_HAS_NO_COPIES. |
| 14:31 |
Dyrcona |
You could try adding an override permission for that... |
| 14:31 |
Dyrcona |
Not sure it will work. |
| 14:35 |
Dyrcona |
But FORCE_HOLD permission is ony checked if you do a force hold. |
| 14:37 |
Dyrcona |
Now, I'm getting a headache.... |
| 14:44 |
JBoyer |
Now I want spaghetti for dinner. :/ |
| 14:45 |
jeff |
bwah! |
| 15:14 |
kmlussier |
I know Friday afternoons are probably a bad time to ask for feedback, but I would love feedback on bug 1670457. |
| 15:14 |
pinesol_green |
Launchpad bug 1670457 in Evergreen "Web client: Action menu requires grid selection when some actions are not performed on selections" [Medium,New] https://launchpad.net/bugs/1670457 |
| 15:14 |
kmlussier |
I also probably could have named that bug better. |
| 15:23 |
csharp |
okay - fieldmapper question... there are three virtual fields on acqpo ("Line Item Count", "Amount Encumbered", "Amount Spent", and "Amount Estimated") that when added to a reports template result in no such column errors |
| 15:23 |
csharp |
I can't find where they are "hooked up" to anything, but I'm not sure how virtual fields work - could someone enlighten me? |
| 15:24 |
berick |
csharp: they are not required to be hooked up anywhere |
| 15:24 |
csharp |
I'm ready to file a bug, but I'm looking for details |
| 15:24 |
miker |
csharp: virtual fields are just places for biz logic to stick data |
| 15:24 |
jeff |
sometimes a virtual field is only filled in by the middle layer code based on some calculation / logic. |
| 15:24 |
miker |
they're calculated in the perl or JS as needed |
| 15:25 |
csharp |
okay - I did some searches for "amount_spent", etc. and came up empty |
| 15:25 |
berick |
csharp: in this particular case, see Financials.pm |
| 15:25 |
berick |
around 1060 |
| 15:26 |
csharp |
okay - I see those now (not sure why I didn't before) |
| 15:26 |
miker |
(specifically, virtual fields that are not a link for has_many or might_have references) |
| 15:26 |
csharp |
well, if the report errors out, that means something's off - is there something more than just adding them to a report to get them to work? |
| 15:26 |
berick |
miker: good point |
| 15:27 |
csharp |
(they're being added to displayed fields) |
| 15:28 |
csharp |
generated SQL: https://pastebin.com/L6iBjjGH |
| 15:28 |
csharp |
is it the SUM() that's throwing them off? |
| 15:29 |
jeff |
the fieldmapper fields in question do not exist as columns in the database, and there is no logic at the reporting level to populate the data in those fields. |
| 15:29 |
Bmagic |
So, yeah, I broke up our 112 "Daily" triggers into 4 pieces and the action trigger perl script is moving much faster |
| 15:29 |
* csharp |
updated the paste with the PG error |
| 15:30 |
jeff |
it would probably be a good idea to omit such fields from being considered by the reporter, either at template time or at execution time (or both). that doesn't solve your current problem in the way that you hoped, though. |
| 15:30 |
csharp |
Bmagic: awesome! |
| 15:31 |
csharp |
jeff: thanks - that's helpful |
| 15:33 |
csharp |
jeff: do you know if there's an easy way to omit the fields from availability in the reporter? |
| 15:35 |
jeff |
csharp: i don't offhand. others recently non-idle might. :-) |
| 15:35 |
csharp |
jeff: thanks :-) |
| 15:37 |
berick |
should be simple, but requires code. reporter is built from the IDL and it knows if a field is virtual, if it's a link, etc. |
| 15:44 |
csharp |
well, I'm considering creating a view that would basically do what it *looks* like that does |
| 15:44 |
csharp |
I'll open a bug soon |
| 15:54 |
Bmagic |
csharp: would it make sense to chain all of the action triggers sections into a single cron line using the "&&" - eliminating the timing dance. Do the Drones need to "rest" between? |
| 15:55 |
berick |
we chain ours with && in many cases to avoid clobbering the server |
| 15:55 |
Bmagic |
oh good! I was curious if the drones didn't mind |
| 15:56 |
Bmagic |
That thought came to me because I am having to divide the "Daily" up in the first place |
| 16:11 |
rlefaive |
I’m trying to edit a Trigger Event Definition (email template) in the Client, and nothing happens when I click “Save.” I think it must be a permission thing, but what logs would tell me what permission I’m missing (at what level)? |
| 16:13 |
berick |
rlefaive: not exactly what you asked, but make sure you have ADMIN_TRIGGER_EVENT_DEF permission. |
| 16:13 |
rlefaive |
berick: thanks, I have it (System) as well as UPDATE_TRIGGER_EVENT_DEF (System). |
| 16:14 |
berick |
and you're editing an event def owned within the same system where you have the perm? |
| 16:15 |
rlefaive |
Huh! “Owning Library” is set to “CONS” - i didn’t recognize that as a library code! |
| 16:17 |
rlefaive |
OK so that makes sense. :) Next question: why can’t I give myself Consortium-level privileges on those *_TRIGGER_EVENT_DEF despite trying? They keep being set back to System. I’m a Local System Admin and have UPDATE_PERM (Consortium) |
| 16:17 |
Bmagic |
Dyrcona: thank you for directing me to HIGH_LEVEL_HOLD_HAS_NO_COPIES - I am seeing that pop up in the logs when attempting to place a hold on this title |
| 16:19 |
Dyrcona |
You can try adding an override permission for that event, but I don't know if it will work or not. |
| 16:19 |
csharp |
rlefaive: I *think* you're prevented from editing your own account (we have that active in PINES) so you might need to be a higher perm level (or admin) |
| 16:19 |
Bmagic |
dispite adding "HIGH_LEVEL_HOLD_HAS_NO_COPIES.override" - still cannot |
| 16:20 |
Bmagic |
despite |
| 16:20 |
Dyrcona |
Ah, OK. It's not overridable. |
| 16:20 |
Bmagic |
can I come to the conclusion that it's not possible? |
| 16:20 |
rlefaive |
ooh, thanks csharp! |
| 16:20 |
csharp |
rlefaive: I haven't looked at that recently - don't want to send you on a wild goose chase :-) |
| 16:21 |
kmlussier |
berick++ # Added insight for bug 1670457 |
| 16:21 |
pinesol_green |
Launchpad bug 1670457 in Evergreen "Web client: Action menu requires grid selection when some actions are not performed on selections" [Medium,New] https://launchpad.net/bugs/1670457 |
| 16:21 |
Dyrcona |
Bmagic: Not without some code changes. |
| 16:21 |
csharp |
rlefaive: also, you might need to change it at the Admin -> System Administration -> Permission Groups level (depending on what you're trying to do) |
| 16:21 |
Bmagic |
ok, groovy, I think we will just attempt to set it up differently |
| 16:21 |
csharp |
adding perms shouldn't be a problem though |
| 16:22 |
Bmagic |
Dyrcona: We were able to accomplish this for circulate=false, but that is different |
| 16:25 |
Bmagic |
Dyrcona: The Everything permission will make it work though |
| 16:26 |
|
jvwoolf joined #evergreen |
| 16:31 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 16:31 |
berick |
hear ye, hear ye, the afternoon test bell rings |
| 16:35 |
kmlussier |
:) |
| 16:37 |
Dyrcona |
Bmagic: If Everything works, but adding HIGH_LEVEL.....override doesn't, I'd consider that a bug of some kind. |
| 16:37 |
Bmagic |
yeah, there is something wrong in here somewhere, hard to say what exactly |
| 16:38 |
Dyrcona |
The override might work if you modify the client code to include HIGH_LEVEL_HOLD_HAS_NO_COPIES as an overridable event. |
| 16:38 |
Dyrcona |
There's a list somewhere in the client js or xul files. |
| 16:39 |
Dyrcona |
But, I've not tried that. |
| 16:44 |
Dyrcona |
Oh wait. I'm thinking of checkout. Holds go through the OPAC. |
| 16:45 |
Dyrcona |
Yeah...Sorry, about that last bit of nonsense. That was the headache talking. |
| 16:49 |
Dyrcona |
I find it suspicious that purge circulation ran in two minutes when it hasn't been run for years. |
| 16:52 |
Dyrcona |
Yeah. Comparing the count of aged circulations to another database it's like it did nothing at all. |
| 16:59 |
Dyrcona |
So, looks like the stored procedure is still running, but cstore timed out. |
| 16:59 |
Dyrcona |
Anyway, calling it a day. Have a good weekend everyone! |
| 17:04 |
kmlussier |
Have a nice weekend all! |
| 17:09 |
Bmagic |
so, I have two hold policies, a deny rule that is general scope, then a more specific scope for allow, the hold matrix always prefers the general deny. UHGG |
| 17:21 |
gmcharlt |
https://evergreen-ils.org/evergreen-3-0-development-update-7/ |
| 17:37 |
dbwells |
Bmagic: Have you tried adjusting the matchpoint weights? |
| 17:38 |
Bmagic |
I was just looking at that |
| 17:38 |
Bmagic |
I suppose we are on "all equal" but I am fuzzy on how I know which one of the weights we are using |
| 17:40 |
dbwells |
Bmagic: Server Admin -> Weights Association |
| 17:41 |
Bmagic |
ok, so we are using Default |
| 17:41 |
Bmagic |
but I am confused, because it seems to me if the matrix matches more points, it would prefer that rule over one with fewer points |
| 17:53 |
Bmagic |
dbwells: "circ.holds.usr_not_requestor" !!!!!! |
| 17:54 |
Bmagic |
from hours of trial and error, it might be this global setting. It made the hold policies unintuitive, where "requestor permission group" really meant "patron permission group" without me realizing |
| 17:56 |
dbwells |
sounds like a good find, hope it solves it for you |
| 17:58 |
csharp |
Bmagic: matches "fall through" so you may end up with a mish-mash of rules combined into a single rule at the end |
| 17:59 |
Bmagic |
csharp: action.find_hold_matrix_matchpoint has the relevant chunk |
| 17:59 |
csharp |
yeah was gonna mention that |
| 17:59 |
* csharp |
started writing a perl script several years ago to make that more human-accessible but got distracted |
| 18:00 |
csharp |
I think Dyrcona has something like from his MVLC days |
| 18:00 |
csharp |
or tsbere |
| 18:00 |
Bmagic |
so, what setting do you have for that global flag? |
| 18:00 |
csharp |
enabled = f |
| 18:01 |
Bmagic |
k, that is what I just did |
| 18:01 |
Bmagic |
and it made it all happy |
| 18:01 |
Bmagic |
lost about 2 hours of my life on that |
| 18:01 |
csharp |
Bmagic: misery loves company :-) |
| 18:01 |
Bmagic |
I think that setting is the spy! |
| 18:01 |
csharp |
@who is the spy |
| 18:01 |
pinesol_green |
StomproJosh is the spy. |
| 18:23 |
Bmagic |
gnight all |
| 18:23 |
bshum |
Oooo Stretch got a release date! |
| 18:25 |
bshum |
June 17. |
| 18:26 |
bshum |
Guess we'll have some time to test it out before 3.0 |
| 20:13 |
|
jvwoolf joined #evergreen |
| 22:20 |
|
genpaku joined #evergreen |
