| Time |
Nick |
Message |
| 02:12 |
|
beanjammin joined #evergreen |
| 06:31 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 07:13 |
|
rjackson_isl joined #evergreen |
| 07:47 |
JBoyer |
miker, berick: Thanks for being patient with my grousing yesterday. I feel like it was becoming unproductive but the main point I was hoping to get at was that I wanted to see more API calls get as fast as pcrud (and forgetting that it's written in C made that a difficult point to reach) |
| 07:50 |
|
kmlussier joined #evergreen |
| 08:02 |
JBoyer |
(I also don't mean to come across as "Hey, somebody go do that!" That's obviously unhelpful as well.) |
| 08:10 |
miker |
JBoyer: I, for one, will never be able to forgive you the lack of subtlety and nuance of a text-only online conversation, as I have NEVER come across in a way unintended. ;) |
| 08:11 |
miker |
srsly, I didn't think you were being harsh |
| 08:12 |
JBoyer |
Were it able to convey either subtlety or nuance from any of *my* communication I would seek the advice of a witchdoctor, because frequently I lack either. ;P |
| 08:25 |
|
rlefaive joined #evergreen |
| 08:32 |
|
agoben joined #evergreen |
| 08:35 |
|
ngf42 joined #evergreen |
| 08:38 |
|
Dyrcona joined #evergreen |
| 08:47 |
|
mmorgan joined #evergreen |
| 09:02 |
|
krvmga joined #evergreen |
| 09:21 |
|
terran joined #evergreen |
| 09:29 |
|
yboston joined #evergreen |
| 09:33 |
|
mmorgan1 joined #evergreen |
| 09:52 |
|
jvwoolf joined #evergreen |
| 09:56 |
|
rhamby left #evergreen |
| 09:58 |
berick |
JBoyer: didn't strike me as grousing at all. and besides I like talking about that stuff. |
| 10:01 |
JBoyer |
I'd be happy to go on at length in St. Charles, then. ;) |
| 10:02 |
berick |
@coffee JBoyer |
| 10:02 |
berick |
or not |
| 10:02 |
* pinesol_green |
brews and pours a cup of Panama Mama Cata Geisha, and sends it sliding down the bar to JBoyer |
| 10:02 |
JBoyer |
At that speed it must be a pour-over. Fancy |
| 10:02 |
berick |
and apparently fusion |
| 10:06 |
kmlussier |
@coffee [someone] |
| 10:06 |
* pinesol_green |
brews and pours a cup of Kenyan AA, and sends it sliding down the bar to mnsri_away |
| 10:07 |
kmlussier |
pinesol_green just needed to wake up. |
| 10:07 |
pinesol_green |
kmlussier: Sorry, we can't do that because, you know, SOFTWARE. |
| 10:21 |
|
Christineb joined #evergreen |
| 10:48 |
|
littlet joined #evergreen |
| 10:52 |
pinesol_green |
[evergreen|Remington Steed] Docs: Add missing section to root.adoc - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=2e157e3> |
| 10:58 |
Dyrcona |
@blame pinesol_green |
| 10:58 |
pinesol_green |
Dyrcona: itself stole Dyrcona's ice cream! |
| 11:35 |
pastebot |
"berick" at 64.57.241.14 pasted "sql tuning" (30 lines) at http://paste.evergreen-ils.org/157 |
| 11:35 |
berick |
sql peeps, wondering if anyone can *cough* explain why the first query is quick and the second is much slower (using hash join) |
| 11:37 |
berick |
more generally, how to convince PG to avoid using a hash join |
| 11:44 |
|
mmorgan joined #evergreen |
| 12:00 |
|
sandbergja joined #evergreen |
| 12:12 |
|
kmlussier joined #evergreen |
| 12:22 |
berick |
heh, didn't answer my question, but found if I inject a join to money.materialized_billable_xact_summary before the all_circ_slim join (and join that to the summary) it avoids the hash join. good enough for now. |
| 12:35 |
|
khuckins_ joined #evergreen |
| 12:42 |
|
beanjammin joined #evergreen |
| 12:43 |
Dyrcona |
Hm. I thought some of the template files had .in forms. Guess I was wrong. |
| 12:47 |
|
jvwoolf joined #evergreen |
| 13:02 |
|
rlefaive joined #evergreen |
| 13:09 |
|
collum joined #evergreen |
| 13:29 |
|
rlefaive joined #evergreen |
| 13:31 |
Dyrcona |
"He was a Makefile wizard....That deaf, dumb, and blind kid sure wrote a mean Makefile!" |
| 13:32 |
miker |
berick: if this is in-code, cstore and friends know how to order joins now |
| 13:33 |
Dyrcona |
@who knows how to order joins? |
| 13:33 |
pinesol_green |
akilsdonk knows how to order joins. |
| 13:34 |
akilsdonk |
hahaha |
| 13:34 |
* miker |
looks at paste and sees he didn't help |
| 13:50 |
Dyrcona |
@who believes in automation? |
| 13:50 |
pinesol_green |
mmorgan believes in automation. |
| 14:11 |
|
khuckins__ joined #evergreen |
| 14:15 |
|
rlefaive joined #evergreen |
| 14:18 |
dbs |
my current favourite use of "Inspect Element" is to change the maximum # of displayed records in MARC Batch Import from 100 to 250 or 500 |
| 14:40 |
mmorgan |
Heh. pinesol_green is correct! |
| 14:46 |
berick |
miker: just some bare sql, but thanks for reminder |
| 15:07 |
|
rlefaive joined #evergreen |
| 15:21 |
|
jvwoolf joined #evergreen |
| 15:28 |
|
kmlussier joined #evergreen |
| 15:58 |
dbs |
btw, https://medium.com/@firt/pwas-are-coming-to-ios-11-3-cupertino-we-have-a-problem-2ff49fd7d6ea has some info on upcoming Service Workers in iOS; naturally Apple is doing things differently so might still need extra effort to enable offli |
| 16:10 |
Dyrcona |
Yeah, Apple sure thinks different.... They can't even English properly. |
| 16:12 |
Bmagic |
hey check this out |
| 16:13 |
Dyrcona |
?? |
| 16:14 |
Bmagic |
a copy is assigned to a hold |
| 16:14 |
Dyrcona |
And, they walk into a bar? |
| 16:14 |
Dyrcona |
:) |
| 16:14 |
Bmagic |
the library pulls it and checks it in. It goes into the courier. Meanwhile, another copy that could have filled that same hold is checked in somewhere else in the consortium. In that exact instant, the transit is canceled in favor of the new copy |
| 16:15 |
Dyrcona |
Shouldn't happen. |
| 16:15 |
Bmagic |
it's happening |
| 16:15 |
Bmagic |
the action.transit_hold_copy table is helping with the cancel_time |
| 16:16 |
berick |
if copy 2 was checked in while copy 1 transaction was open (same second, basically) i believe it could be possible |
| 16:16 |
berick |
sounds familiar |
| 16:16 |
Bmagic |
it's pretty weird that the cancel_time = auditor.action_hold_request_history |
| 16:16 |
Bmagic |
it's hours later |
| 16:16 |
Dyrcona |
Someone canceled the hold, then. |
| 16:16 |
Dyrcona |
Or, the transit, rather. |
| 16:17 |
jeff |
checkin modifier "retarget local holds" being too aggressive? |
| 16:17 |
|
_sandbergja joined #evergreen |
| 16:17 |
jeff |
"checkout fills related hold"? (doesn't sound like it from the symptoms) |
| 16:18 |
Bmagic |
the new copy was assigned to the hold at "2018-01-18 19:41:23.50588-06" and the cancel time of the original transit is "2018-01-18 12:10:45.956107-06" |
| 16:18 |
Bmagic |
ok, now that I am typing this out loud, I am realizing that the only thing that was the same instant is the transit time of the new copy and the hold copy assignment |
| 16:18 |
Dyrcona |
Seven hour difference. Someone canceled the transit. |
| 16:19 |
Bmagic |
Sorry to bother! Thanks for listening to me though! I thought I had an awesome thing happening |
| 17:03 |
|
mmorgan left #evergreen |
| 17:09 |
kmlussier |
@dessert |
| 17:09 |
* pinesol_green |
grabs some Carrot Cake for kmlussier |
| 17:10 |
* kmlussier |
was hoping for chocolate. |
| 17:16 |
|
jvwoolf left #evergreen |
| 18:05 |
|
kektrain joined #evergreen |
| 18:05 |
|
kektrain left #evergreen |
| 18:31 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 19:11 |
|
Bmagic joined #evergreen |
