| Time |
Nick |
Message |
| 01:23 |
|
sandbergja joined #evergreen |
| 02:03 |
|
rjackson_isl joined #evergreen |
| 06:32 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 07:28 |
|
agoben joined #evergreen |
| 08:00 |
|
rlefaive joined #evergreen |
| 08:09 |
|
collum joined #evergreen |
| 08:32 |
|
dwgreen joined #evergreen |
| 08:42 |
|
mmorgan joined #evergreen |
| 09:00 |
|
idjit joined #evergreen |
| 09:09 |
|
lsach joined #evergreen |
| 09:15 |
|
nfpl joined #evergreen |
| 09:15 |
|
kmlussier joined #evergreen |
| 09:22 |
Bmagic |
Just curious, when doing an apache2 reload I notice that OpenSRF spins up tons of new processes. What is going on there? Are current sessions/connections getting reset? |
| 09:28 |
jeff |
perl handlers re-running their init subs, I believe. |
| 09:29 |
jeff |
even "just" on reload. |
| 09:29 |
Bmagic |
Not a problem for connected clients? Hopefully no one notices when a reload happens on a brick. |
| 09:33 |
|
jvwoolf joined #evergreen |
| 09:38 |
|
yboston joined #evergreen |
| 09:50 |
berick |
Bmagic: an active apache backend will wait until it's done with the current request before exiting during reload |
| 09:51 |
berick |
Bmagic: however, if many apache procs start at once, it can have a load blip on the server |
| 09:52 |
Bmagic |
The load blip is noticable for sure |
| 09:53 |
Bmagic |
The main thing is hiccup in service and it sounds like the users would feel a slowness but not an error? |
| 09:54 |
berick |
guess it depends on the severity of the load spike, but generally speaking apache reloads are safe |
| 09:54 |
Bmagic |
ty |
| 09:55 |
berick |
Bmagic: fwiw, i take servers out of rotation before reloading to be extra safe |
| 09:55 |
Bmagic |
roger that |
| 09:56 |
dbwells |
berick: Do you know if that general safety applies to Apache restarts as well? I've never noticed any ill effects, and sometimes it just has to be done, but I don't want to be totally callous :) |
| 09:57 |
berick |
dbwells: apache restarts will disrupt active connections |
| 09:57 |
berick |
i don't think it's very forgiving |
| 09:59 |
berick |
active meaning in the middle of a request |
| 10:04 |
|
Christineb joined #evergreen |
| 10:07 |
dbwells |
Thanks. I'm curious now to see if transmission just plain stops or some degree of process "cleanup" at least tells the other end, hey, I'm going away. |
| 10:07 |
* dbwells |
has better things to do, probably |
| 10:07 |
berick |
yeah, i don't know |
| 10:26 |
miker |
there's a graceful restart option that waits for requests to complete. there may not be a built in option for that with systemd on some distros, though |
| 10:30 |
miker |
(and if you experience service slowdowns because of apache startup of a dozen backends, you'll probably experience them with normal user interaction sometimes ... think "density wave" a la traffic patterns ... so you proabably want to increase min spare and max kids, and maybe add a vcpu or three. or separate resource "contenders") |
| 10:34 |
Bmagic |
We are going through our apache configs for the big Google SSL August 1st deal... This line "RedirectMatch ^/$ https://ourdomain/eg/opac/home?locg=289" seems to be redundant with "SetEnv physical_loc 289" as far as the resulting tt2->html |
| 10:35 |
Bmagic |
Auditing the configs caused me to find this out. It seems that we can remove the RedirectMatch line and be just fine but I thought I would see what others thought? |
| 10:36 |
pinesol_green |
[evergreen|Dan Wells] Forward-port 3.1.4 upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=daf65af> |
| 10:40 |
pinesol_green |
[evergreen|Dan Wells] Forward-port 3.0.10 upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=827d027> |
| 11:13 |
|
jvwoolf joined #evergreen |
| 11:22 |
|
jvwoolf joined #evergreen |
| 11:45 |
|
rlefaive_ joined #evergreen |
| 11:58 |
|
yboston joined #evergreen |
| 12:07 |
|
jihpringle joined #evergreen |
| 12:08 |
|
rlefaive joined #evergreen |
| 12:19 |
|
rlefaive joined #evergreen |
| 12:23 |
|
sandbergja joined #evergreen |
| 12:48 |
Bmagic |
Can anyone think of a scenario that would cause Webby to tell the staff that the item that is attempted to checkout is on hold for another patron but when investigated we find that it was on hold for the patron who was checking it out (they came to pick it up) |
| 12:52 |
kmlussier |
That sounds slightly familiar |
| 12:52 |
kmlussier |
I think it's a bug that has been fixed, but I'm not finding it. |
| 12:53 |
kmlussier |
bug 1755502, but that bug was fixed in time for the full 3.1 release. |
| 12:54 |
pinesol_green |
Launchpad bug 1755502 in Evergreen "Alternate hold pickup popup displays when checking out item to hold patron" [High,Fix released] https://launchpad.net/bugs/1755502 |
| 13:09 |
Bmagic |
darn, not in 3.0.x |
| 13:09 |
Bmagic |
kmlussier++ |
| 13:13 |
rjackson_isl |
evergreeners++ for the hard work on all the latest fixes in 3.1.4 and 3.0.10! Impressive :) |
| 13:15 |
|
yboston joined #evergreen |
| 13:21 |
|
nfpl joined #evergreen |
| 13:26 |
nfpl |
Is there a way to set column configurations on the server side to set up defaults for the way we want them to look? We want to modify the fields shown by default in our tables staff side on all workstations so it is consistent. |
| 13:27 |
nfpl |
Also, congrats on 3.1.4. I will be upgrading our system to it next week |
| 13:30 |
idjit |
3.1.4: pi edition. :-) |
| 13:30 |
kmlussier |
nfpl: Not yet, but it may be part of the work berick is doing to move local storage settings to the database. |
| 13:31 |
JBoyer |
nfpl, I believe it currently requires modifying the templates by hand, but once bug 1750894 is closed you'll be able to do that and much more without messing with template files. |
| 13:31 |
pinesol_green |
Launchpad bug 1750894 in Evergreen "Wishlist: Store web staff workstation settings on the server" [Wishlist,New] https://launchpad.net/bugs/1750894 |
| 13:31 |
JBoyer |
kmlussier++ |
| 13:31 |
kmlussier |
JBoyer: Jinx! |
| 13:31 |
JBoyer |
:) |
| 13:32 |
nfpl |
Oh fantastic. I tried modifying the template and it didn't want to listen.... |
| 13:32 |
JBoyer |
(Which I am supposed to be loading on a test server any time now... Where oh where have the tuits gone, long time passing, etc.) |
| 13:33 |
kmlussier |
Yes, I was just thinking the same thing. I might look at it tomorrow. |
| 13:33 |
JBoyer |
nfpl, you may need a "Clear cache and hard reload" to get it to appear, and if you've saved a column config locally that won't be overridden. |
| 13:35 |
|
rlefaive joined #evergreen |
| 13:36 |
nfpl |
I may be in the wrong file? t_checkin_table.tt2 doesn't respond to changes |
| 13:37 |
|
rlefaive joined #evergreen |
| 13:39 |
jeff |
keep in mind also that you might be cachine tt templates on the server side. |
| 13:39 |
jeff |
s/cachine/caching/ even |
| 13:41 |
berick |
bug 1750894 implements the server-side storage. my first follow up to that bug once merged will be a way for staff (w/ appropriate permissions) to save settings for a given gid as an org setting (instead of user/workstation). |
| 13:41 |
pinesol_green |
Launchpad bug 1750894 in Evergreen "Wishlist: Store web staff workstation settings on the server" [Wishlist,New] https://launchpad.net/bugs/1750894 |
| 13:41 |
berick |
you could approximate the same by saving a grid then manually copying the setting value to the analogous org setting |
| 13:44 |
nfpl |
Okay. It sounds great, guess I will wait. Just thought the defaults would be able to be altered server side as staff here have come up with a standard grid config. I look forward to 3.2 |
| 13:47 |
kmlussier |
nfpl: t_checkin_table should be the right file for the checkin page. In addition to the caching, you'll need to remove any locally-saved settings from the browser's local storage. |
| 14:31 |
|
sandbergja joined #evergreen |
| 14:34 |
|
sandbergja joined #evergreen |
| 14:50 |
* berick |
grabs 1115 |
| 14:57 |
|
bdljohn joined #evergreen |
| 15:03 |
pinesol_green |
Showing latest 5 of 7 commits to Evergreen... |
| 15:03 |
pinesol_green |
[evergreen|Mike Rylander] LP#1766716: Release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=71c4649> |
| 15:03 |
pinesol_green |
[evergreen|Galen Charlton] LP#1766716: (follow-up) expand CSS imports for now - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=a440cb0> |
| 15:03 |
pinesol_green |
[evergreen|Bill Erickson] LP#1766716 Closed dates API emergency filter fix - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3f249b1> |
| 15:03 |
pinesol_green |
[evergreen|Mike Rylander] LP#1766716: Adjust date display logic - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=76ff232> |
| 15:03 |
pinesol_green |
[evergreen|Bill Erickson] LP#1766716 Stamping emergency closing DB upgrade - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=15c4db3> |
| 15:08 |
|
mmorgan1 joined #evergreen |
| 15:15 |
|
remingtron_ joined #evergreen |
| 15:21 |
|
derekz joined #evergreen |
| 15:33 |
|
mmorgan joined #evergreen |
| 15:49 |
|
jihpringle joined #evergreen |
| 15:55 |
|
mmorgan1 joined #evergreen |
| 16:30 |
|
yboston joined #evergreen |
| 16:53 |
csharp |
maybe a bug? when I try to hard-code a "not in list" in the web client reporter UI, it treats what I've entered as a single list item |
| 16:54 |
csharp |
e.g., PO states of 'invalid' and 'complete' become AND "ce91a29a867084d503894c06769de8b4"."state" NOT IN ($_5238$'complete', 'invalid'$_5238$) |
| 16:57 |
csharp |
entering them in the report definition is fine for a recurring report, but some you just don't wanna |
| 16:58 |
csharp |
esp. things you have to know like IDs |
| 16:58 |
|
jvwoolf left #evergreen |
| 17:12 |
|
mmorgan left #evergreen |
| 17:14 |
|
sandbergja joined #evergreen |
| 18:31 |
pinesol_green |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live> |
| 19:03 |
|
kenstir joined #evergreen |
| 19:05 |
kenstir |
Hey evergreeners, your friendly open source mobile app developer here. |
| 19:05 |
kenstir |
I know how to retrieve user preferences like Pickup Library using open-ils.actor.user.fleshed.retrieve; is there a corresponding OSRF request to set it? |
| 19:18 |
|
sandbergja joined #evergreen |
| 20:09 |
|
hbrennan joined #evergreen |
| 20:12 |
hbrennan |
Can I not put a copy alert on a precat item? The Copy Alert button is missing from the Volume/Copy Editor screen, but I don't think it's a permission thing. |
| 20:12 |
hbrennan |
afternoon in Alaska = talking aloud to myself :) |
| 20:21 |
|
sandbergja joined #evergreen |
| 20:44 |
miker |
hbrennan: I'm not surprised. we should really disallow directly editing precats, imo... they "don't exist" except during a circ. you should be able to "catalog" them by adding the copy as a regular holding, even while in use |
| 20:45 |
hbrennan |
miker: I sent a book out that was barcoded but not in catalog. I just wanted to add a note on return for who to give it to. Webby allowed me to add a regular Alert Message. |
| 20:46 |
hbrennan |
I was just really excited to use Copy Alerts with those fancy options |
| 21:00 |
miker |
:D |
| 21:26 |
|
sandbergja joined #evergreen |
| 21:52 |
|
sandbergja joined #evergreen |
