| Time |
Nick |
Message |
| 07:19 |
|
rjackson_isl_hom joined #evergreen |
| 07:29 |
|
collum joined #evergreen |
| 07:56 |
|
BDorsey joined #evergreen |
| 08:12 |
|
rfrasur joined #evergreen |
| 08:24 |
|
mantis1 joined #evergreen |
| 08:33 |
|
mmorgan joined #evergreen |
| 08:38 |
|
Dyrcona joined #evergreen |
| 09:05 |
|
jvwoolf joined #evergreen |
| 09:09 |
|
terranm joined #evergreen |
| 09:50 |
|
mmorgan1 joined #evergreen |
| 10:13 |
|
mmorgan joined #evergreen |
| 10:29 |
|
rjackson_isl_hom joined #evergreen |
| 11:34 |
berick |
heh, explain analyze for a report running 24hrs now. |
| 11:36 |
* berick |
tries same on PG9 |
| 12:01 |
|
jihpringle joined #evergreen |
| 12:19 |
|
collum joined #evergreen |
| 13:59 |
|
jihpringle joined #evergreen |
| 14:12 |
|
jihpringle joined #evergreen |
| 14:33 |
|
stephengwills joined #evergreen |
| 14:57 |
miker |
berick: just looked at your explain ... if you haven't already, could be as simple as creating a functional index on both the circ tables like this: create index concurrently circyear on action.[aged_]circulation (date_part('year', xact_start)); |
| 15:21 |
JBoyer |
Yikes, I had lost track and assumed we both got too busy to notice. D: |
| 15:47 |
Dyrcona |
miker++ |
| 15:50 |
Dyrcona |
We are referring to the explain posted yesterday, right? I don't see another one. (Not that I'm an expert or anything...) |
| 15:56 |
miker |
Dyrcona: that's the one I was referring to today, yes. I wandered away from irc until lunch time today ... sorry for not looking sooner, and apologies to you CPU's heat sink, berick :) |
| 15:57 |
Dyrcona |
Well, I was off yesterday and didn't login to IRC. I'd definitely try the index that miker suggests. Those two Seq Scans look bad with date_part look bad. |
| 15:58 |
Dyrcona |
Oof... I blame the redundancy on it being late in the day. :) |
| 16:30 |
|
jvwoolf left #evergreen |
| 17:30 |
|
mmorgan left #evergreen |
