Time |
Nick |
Message |
04:51 |
|
khuckins joined #evergreen |
06:00 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
07:20 |
|
khuckins joined #evergreen |
07:21 |
|
rjackson_isl_hom joined #evergreen |
07:47 |
|
collum joined #evergreen |
08:24 |
|
mantis1 joined #evergreen |
08:30 |
|
rfrasur joined #evergreen |
08:37 |
|
mmorgan joined #evergreen |
08:39 |
|
Dyrcona joined #evergreen |
08:46 |
Dyrcona |
I'm trying to decide if I should do my next test on Pg 9.6 or Pg 12. I also wonder if I should log all queries or just those over a certain threshold, like 250ms or so. |
08:48 |
Dyrcona |
Looks like I have plenty of free space to log all queries for a while. |
08:59 |
csharp |
looks like my troubles from yesterday may have been a super-long-running emergency closing (15 hours) |
08:59 |
csharp |
also did some PG vacuuming last night |
09:02 |
|
rfrasur joined #evergreen |
09:12 |
|
jvwoolf joined #evergreen |
09:19 |
Dyrcona |
csharp: It had that much impact? We've had them run for a long time but not really cause an issue. |
09:27 |
|
dbwells joined #evergreen |
09:58 |
|
berick joined #evergreen |
10:26 |
|
nfBurton28 joined #evergreen |
10:26 |
Dyrcona |
I should have used a duration on the query logging. Getting all of them is a bit much. |
10:27 |
|
stephengwills left #evergreen |
10:55 |
Dyrcona |
This bugs me: "AS x WHERE 1=0;" |
10:55 |
|
sandbergja joined #evergreen |
10:56 |
Dyrcona |
Looks to me like the query is running over a second only to have the results ignored. |
11:00 |
Dyrcona |
OK. That's IDL stuff. Moving along.... |
11:08 |
Dyrcona |
Hmm.. Looks like my hold targeter ran faster than yesterday, but it is still too long. |
11:14 |
Dyrcona |
I'll test my changes on Pg 9.6, and then tomorrow run the old code and log long running queries. The old code seemed faster yesterday. |
11:27 |
Dyrcona |
So, my code is faster on Pg 9.6 than it is on Pg 12. This is an area that we'll have to look at. |
11:29 |
Dyrcona |
The 9.6 db has default optimizations, and it was updated yesterday morning, so it has not had the hold targeter run since Saturday. |
11:30 |
Dyrcona |
miker: Should I open a Lp bug about future Postgres versions? |
11:32 |
Dyrcona |
Could be that my Pg 12 optimization is whack. |
11:33 |
miker |
generally? sure ... was there a specific thing, though? I haven't been following the hold targeter speed stuff, sorry. (and, re "where 1=0" those really shouldn't be taking 1s each! we could change that to limit 0, maybe? that's the C code investigating the actual data types of the columns at startup) |
11:34 |
Dyrcona |
miker: We've seen some things that seem slower in Pg 12 vs 9.6. I've seen other things that are a lot faster. I'll see if I can get some statistics. |
11:35 |
Dyrcona |
On the "where 1=0" I think those are OK as they come from the IDL. I think it is part of making temporary views/tables. |
11:37 |
Dyrcona |
I've also discovered that my maintenance_work_mem setting is way out of whack on my development server for Pg 12. I'm going to remove my optimizations from the config for now. |
11:38 |
Dyrcona |
I had GB instead of MB. :) |
11:38 |
csharp |
oof - I've done that |
11:41 |
Dyrcona |
I should reload Pg 11 and Pg 13 and test this against those dbs, too. |
11:53 |
|
jihpringle joined #evergreen |
12:30 |
|
mantis1 left #evergreen |
12:30 |
|
khuckins joined #evergreen |
13:38 |
|
Dyrcona joined #evergreen |
13:56 |
Dyrcona |
Grr.. mozilla-- |
13:56 |
Dyrcona |
Or, maybe I should blame the Ubuntu packagers. |
14:02 |
Dyrcona |
Oh, all right. I forgot that Lightning is now built in to Thunderbird. Jumping from version 68 to 78 can be jarring. :) |
14:03 |
Dyrcona |
Except that my Google calendars are not working..... :( |
14:03 |
|
mantis1 joined #evergreen |
14:04 |
Dyrcona |
In case this happens to you, not likely, just disable the "Provider for Google Calendar" and re-enable. |
14:04 |
Dyrcona |
@dunno |
14:04 |
pinesol |
Dyrcona: http://images.cryhavok.org/d/1291-1/Computer+Rage.gif |
14:05 |
|
sandbergja joined #evergreen |
14:05 |
Dyrcona |
@dunno get 12 |
14:05 |
pinesol |
Dyrcona: Dunno #12: "Have you tried turning it off and back on again?" (added by csharp at 01:04 PM, August 31, 2012) |
14:06 |
Dyrcona |
I sort of went through this with Manjaro on my Pinebook a couple of weeks ago. |
14:13 |
Dyrcona |
I liked Enigmail better than this built-in OpenPGP encryption. At least, Enigmail worked. This new code can't seem to find keys on keyservers. |
14:16 |
Dyrcona |
OK. I can blame the Ubuntu packagers for not including Enigmail 2.2. It provides a migration tool. |
14:16 |
Dyrcona |
S'pose this is what I get for blindly updating in the middle of the day. |
14:24 |
Dyrcona |
Installed the latest Enigmail and it migrated my keys with no problems. I now return to the usual Evergreen babble. :) |
14:46 |
berick |
Dyrcona++ |
14:48 |
Dyrcona |
Whoa! The regular hold targeter code from 3.2.10 has queries going for over an hour on Pg 12 with default settings! |
14:48 |
Dyrcona |
Longest-running query: 164292 | 01:48:58.605979 | active | 10.250.150.75 | open-ils.cstore | SELECT "ahr".current_copy AS "current_copy", "acp".id AS "id", "acp".status AS "status", "acp".circ_lib AS "circ_lib" FROM asset.copy AS "a |
14:49 |
Dyrcona |
I believe I will go back to the "optimized" settings for Pg 12. |
14:51 |
jeffdavis |
I see those specific long running queries, never did track down why they run so long |
14:53 |
Dyrcona |
I don't usually see those running very long in production on Pg 9.6. |
14:55 |
Dyrcona |
I sometimes see really long running searches. |
14:55 |
jeffdavis |
We kill those. |
14:56 |
Dyrcona |
Yeah, so do I. |
14:57 |
Dyrcona |
I just updated the kernel on the development db server and rebooted, but the queries appeared to go away after I shutdown services on 10.250.150.75. |
14:58 |
Dyrcona |
So, I think my hold targeter performance issues are more likely related to changes in Pg 12 and not something that I did. |
15:07 |
Dyrcona |
I think I'll take another crack at this tomorrow with some increased settings for Pg 12. |
15:15 |
|
sandbergja joined #evergreen |
15:32 |
|
mantis1 left #evergreen |
16:00 |
|
RFrasur_ joined #evergreen |
16:32 |
|
sandbergja joined #evergreen |
17:02 |
|
sandbergja joined #evergreen |
17:05 |
|
mmorgan left #evergreen |
17:30 |
|
issyl0LU joined #evergreen |
17:30 |
issyl0LU |
/!\ this chat has moved to irc.crimeircd.net #pp /!\ |
17:33 |
|
Raito_BezariusvT joined #evergreen |
17:33 |
Raito_BezariusvT |
/!\ this chat has moved to irc.crimeircd.net #0 /!\ |
17:33 |
* jeff |
sighs |
17:49 |
|
sandbergja joined #evergreen |
17:49 |
|
khuckins joined #evergreen |
17:49 |
|
jihpringle joined #evergreen |
17:49 |
|
berick joined #evergreen |
17:49 |
|
dbwells joined #evergreen |
17:49 |
|
rjackson_isl_hom joined #evergreen |
17:49 |
|
alynn26 joined #evergreen |
17:49 |
|
awitter joined #evergreen |
17:49 |
|
book`_ joined #evergreen |
17:49 |
|
kip joined #evergreen |
17:49 |
|
egbuilder joined #evergreen |
17:49 |
|
genpaku joined #evergreen |
17:49 |
|
jeff joined #evergreen |
17:49 |
|
yeats_ joined #evergreen |
17:49 |
|
jweston joined #evergreen |
17:49 |
|
rhamby joined #evergreen |
17:49 |
|
gmcharlt_ joined #evergreen |
17:49 |
|
Christineb joined #evergreen |
17:49 |
|
jonadab joined #evergreen |
17:49 |
|
troy__ joined #evergreen |
17:49 |
|
laurie joined #evergreen |
17:49 |
|
drigney joined #evergreen |
17:49 |
|
abneiman joined #evergreen |
17:49 |
|
miker joined #evergreen |
17:49 |
|
JBoyer joined #evergreen |
17:49 |
|
nfBurton joined #evergreen |
17:49 |
|
dbs joined #evergreen |
17:49 |
|
yar joined #evergreen |
17:49 |
|
bshum joined #evergreen |
17:49 |
|
dickreckard joined #evergreen |
17:49 |
|
Bmagic joined #evergreen |
17:49 |
|
phasefx joined #evergreen |
17:49 |
|
jeff_ joined #evergreen |
17:49 |
|
jeffdavis joined #evergreen |
17:49 |
|
ejk joined #evergreen |
17:49 |
|
jamesrf joined #evergreen |
17:49 |
|
eby joined #evergreen |
17:49 |
|
pinesol joined #evergreen |
17:49 |
|
csharp joined #evergreen |
17:49 |
|
lisacarlucci joined #evergreen |
17:49 |
|
devted joined #evergreen |
17:49 |
|
dluch joined #evergreen |
17:49 |
|
pastebot joined #evergreen |
17:49 |
|
akilsdonk joined #evergreen |
17:51 |
|
WarpedMJ joined #evergreen |
17:59 |
|
PityDaFool joined #evergreen |
18:00 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
18:01 |
|
ome joined #evergreen |
18:08 |
|
ljharbjj joined #evergreen |
18:10 |
pinesol |
[evergreen|Terran McCanna] LP1916085 Bootstrap OPAC - Pagination on copy table - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c2e0686> |
19:49 |
|
nfBurton28 joined #evergreen |
21:19 |
|
yar joined #evergreen |