Time |
Nick |
Message |
00:19 |
|
sandbergja joined #evergreen |
00:32 |
|
dbwells joined #evergreen |
00:47 |
|
AFloyd__ joined #evergreen |
00:48 |
|
yar joined #evergreen |
01:13 |
|
dbwells joined #evergreen |
01:54 |
|
dbwells joined #evergreen |
02:48 |
|
dbwells joined #evergreen |
03:29 |
|
dbwells joined #evergreen |
03:56 |
|
dbwells joined #evergreen |
04:37 |
|
dbwells joined #evergreen |
05:05 |
|
dbwells joined #evergreen |
05:45 |
|
dbwells joined #evergreen |
06:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
06:40 |
|
dbwells joined #evergreen |
07:34 |
|
rjackson_isl_hom joined #evergreen |
08:05 |
|
rfrasur joined #evergreen |
08:21 |
|
mantis joined #evergreen |
08:34 |
|
mmorgan joined #evergreen |
08:38 |
|
Dyrcona joined #evergreen |
08:38 |
Dyrcona |
I have a cstore drone that is currently using about 7GB of RAM and 0% CPU. Think it is OK to just kill it? |
08:51 |
Dyrcona |
Well, I did. |
09:15 |
|
dbwells joined #evergreen |
09:16 |
|
dbwells_ joined #evergreen |
09:18 |
|
jvwoolf joined #evergreen |
09:52 |
|
Dyrcona joined #evergreen |
09:57 |
|
collum joined #evergreen |
10:06 |
Dyrcona |
We can't be the only ones who get reports of white screens and printing issues after upgrades, can we? |
10:08 |
Dyrcona |
Doesn't help, either, that one library reported that Chrome updated to version 90 while they were checking the version for us. New, untested, Chrome version complicates matters. |
10:09 |
Dyrcona |
And, at least one library's issue stemmed from having Hatch enabled without having Hatch installed.... |
10:12 |
Dyrcona |
Hm. I guess 90 is the "current" version of Chrome. Must have been a patch update. |
10:29 |
|
collum joined #evergreen |
10:39 |
rhamby |
A few years ago I heard a lot of white screen complaints after upgrades but it has been long enough that I can't remember exactly how long ago it was |
10:42 |
csharp |
whenever we hear of whitescreen complaints it's usually hatch (or cache/cookies) |
10:57 |
Dyrcona |
Ours seem to mostly be Hatch-related so far: 3.5 really does not like the old version of Hatch. After disabling or updating Hatch, things seem OK. |
11:06 |
Dyrcona |
Did anyone ever figure out a trick to disable/answer that survey question when Angular is installed via npm? That caused my eg2 not to build during the upgrade. |
11:08 |
JBoyer |
Dyrcona export NG_CLI_ANALYTICS=false or include that on the command line: su -c "NG_CLI_ANALYTICS=false make -f ..." |
11:08 |
Dyrcona |
JBoyer: TYVM. I was gonna try 'echo N |' |
11:13 |
JBoyer |
I tried 'yes n' at one point, but depending on how things look re: cpan config that can become exciting. |
11:17 |
Dyrcona |
Yeah, I suppose. The analytics thing is "new" with Node 12 or so. |
11:17 |
Dyrcona |
It happens when prereqs are installed and when you do npm install for eg2, now. |
11:18 |
Dyrcona |
In my specific case, I've already configured CPAN. |
11:19 |
Dyrcona |
Also, emacs++ # Git works over tramp. If you don't know what that means, go back to notepad. :) |
11:34 |
Dyrcona |
So, doing npm install with Node v12+ on Ubuntu 18 changes ownership of the .config directory. Anyone else noticed this? |
11:34 |
Dyrcona |
It's inconvenient, because git complains afterward. |
11:36 |
Dyrcona |
Rather, it looks like it creates .config if it doesn't exist, and then git tries to use .config.... |
12:15 |
|
jihpringle joined #evergreen |
12:33 |
|
collum joined #evergreen |
13:09 |
|
jamesrf joined #evergreen |
13:53 |
|
mmorgan1 joined #evergreen |
13:55 |
|
sandbergja joined #evergreen |
13:56 |
Dyrcona |
So, is it normal for each brick's fmall.js to be different from the others and to have a different file size? |
13:57 |
Dyrcona |
I could see that the classes might show up in a different order, but it seems to me that they should be pretty much the same size. |
14:02 |
Dyrcona |
No two of the fmall.js files are on the same on my bricks. |
14:36 |
csharp |
Dyrcona: mine are all the same |
14:39 |
Dyrcona |
csharp: Do you serve them all from NFS or do you generate them on each brick? |
14:40 |
Dyrcona |
Mine all have the same timestamp during the upgrade. |
14:59 |
|
mmorgan joined #evergreen |
15:11 |
Dyrcona |
Well, mine all differ on the definition of 1 object: complete_lineitem, which could explain the weirdness that I'm seeing in acquisitions. |
15:14 |
Dyrcona |
Hmm. complete_lineitem looks like a regular money billing thing, not acquisitions. |
15:14 |
Dyrcona |
It also appears to be in our custom field section. |
15:18 |
Dyrcona |
And... complete_lineitem is a complete mess. It's using money.billable_xact_id_seq and then looks like a bunch of bib and acquistions columns. I wonder if we even use this? |
15:22 |
Dyrcona |
I honestly don't think this is our problem, but..... |
15:24 |
Dyrcona |
After removing that bad class from fmall.js, they are all the same, now. This might make an interesting case study because it triggers something to make the output different. |
15:32 |
|
mantis left #evergreen |
15:32 |
|
jihpringle joined #evergreen |
15:33 |
Dyrcona |
And, since upgrading to 3.5 from 3.2, we seem to have had an increase in actor drones in use. |
15:35 |
Dyrcona |
Some of our other custom views are broken as far as their sequences in the IDL, but they don't produce different output in the JS IDL files. |
15:41 |
Dyrcona |
Well, looks like I'll be able to drop complete_lineitem. It's not used in any of our reports. |
15:52 |
jeffdavis |
I definitely recommend increasing actor drones. We're at max 200 on 4 different servers and still hit the limit occasionally. |
15:53 |
jeffdavis |
The fix for bug 1918362 should help too. |
15:53 |
Dyrcona |
jeffdavis: Yeah, when I went to 200 before we'd still hit it. I don't think that's the best approach. |
15:53 |
pinesol |
Launchpad bug 1918362 in Evergreen 3.6 "Unchanged workstation settings are re-applied on every checkin" [Medium,Confirmed] https://launchpad.net/bugs/1918362 |
15:56 |
jeffdavis |
You're on 3.5.3 so you should have the other fixes that I know of. |
15:56 |
Dyrcona |
Yeah. |
16:37 |
|
jihpringle joined #evergreen |
17:16 |
|
mmorgan left #evergreen |
18:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
19:16 |
|
dbwells joined #evergreen |
21:43 |
|
dbwells joined #evergreen |